Bottlehead Forum
Bottlehead Kits => Crack => Topic started by: kscwuzhere on March 29, 2014, 09:48:59 AM
-
Hey so I currently have a crack with speedball driving my Hd650s. My current output tube is the Chatham/Tung-sol 5998 and the driver tube is the Tungsram e80cc. I read somewhere that there are some possible circuit changes you can make to run these two tubes optimally and I was just wondering what these changes were? Do I need to replace or add any additional resistors? Thank you bottlehead community!
-
What voltages do you get at T1 and T5 with the E80CC?
-
Terminal 1: 130V
Terminal 5: 121V
-
Yeah, that's not really so great.
Try replacing the 237 R1 resistors with 470 Ohm resistors to see if the voltage bumps down between 55 and 85V.
-PB
-
Yeah, that's not really so great.
Try replacing the 237 R1 resistors with 470 Ohm resistors to see if the voltage bumps down between 55 and 85V.
-PB
wouldn't this be a permanent change and how would this affect the other tube rolling ?
-
So I went ahead and ordered a bunch of 470 ohm resistors, hopefully that'll do the trick!
-
wouldn't this be a permanent change and how would this affect the other tube rolling ?
Yes, it would be a permanent change. Yes, it won't be ideal for other tubes.
-
So I went ahead and ordered a bunch of 470 ohm resistors, hopefully that'll do the trick!
Please post some feed back here on how you get on with the resistor swap, I tend to favour the E80cc also and am very tempted to go down the same path.
Cheers
-
wouldn't this be a permanent change and how would this affect the other tube rolling ?
Yes, it would be a permanent change. Yes, it won't be ideal for other tubes.
what could happen if I won't change resistors....would I need one of this attached to Crack
-
what could happen if I won't change resistors....would I need one of this attached to Crack
You'll throw off the operation of the 6080. The results won't be great (IMO) with both the Speedball and the stock kit.
-
Thanks Caucasian
-
Please post some feed back here on how you get on with the resistor swap, I tend to favour the E80cc also and am very tempted to go down the same path.
Cheers
I'll make sure to update this thread with the voltage changes after I swap the resistors.
By switching the resistors and running the tubes optimally, what audible differences would result?
edit: Also, I ended up getting the 1/4W 470ohm 1% metal film resistor, would this suffice for the swap?
-
what voltage should be appropriate ,I have this circled
http://www.mouser.com/ProductDetail/Vishay-Dale/RN55C4700BB14/?qs=sGAEpiMZZMt7FrWooXVB1yjOewkIa0Gq
http://www.mouser.com/ProductDetail/KOA-Speer/MF1-4CCT52R4700F/?qs=sGAEpiMZZMv8QEQcPUP6BXh8cFBVkBnm
http://www.mouser.com/ProductDetail/Vishay-Beyschlag/MBE04140C4700FC100/?qs=sGAEpiMZZMukHu%252bjC5l7YUGjKnXnGlgNvS%252baD1SLCZo%3d
http://www.ebay.com/itm/161148100242?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1497.l2649
-
So I ordered the wrong resistor? Are 1/4W 470 ohm resistors fine or should I order something else?
-
1/4 Watt will work fine, they will just be physically kind of large.
-
1/4 Watt will work fine, they will just be physically kind of large.
Hopefully they'll fit!
-
will 1W work well ..I really want to try this new resistors
http://www.mouser.com/ProductDetail/Vishay-Beyschlag/MBE04140C4700FC100/?qs=sGAEpiMZZMukHu%252bjC5l7YUGjKnXnGlgNvS%252baD1SLCZo%3d
-
If a 1/4 Watt resistor is a borderline fit, a 1W really won't work well.
Also, there's no reason to use such a large resistor there, 1/10 of a Watt would be sufficient.
-
Thank you
-
Would anyone knowledgeable (that rules me out) care to comment on what sort of effect running the bhc+sb+e80cc+ts 5998 & this resistor mod have on a Cracks output impedance. I have read the TS-5998 lowers the Cracks output impedance and wondered if the input tube can also effect this?
Thanks
Jamie
-
Output impedance will go down as transconductance goes up.
The transconductance of a 5998 is generally higher than a 6080 at equal operating points. There is some transconductance variation based on the operating point, which suggests a range of 7,000uMho to 21,000uMho (approximately) for the 5998 at typical operating points. Given that the Speedball fixes the current at about 30mA per section of the 5998, then the operating point of the driver stage roughly sets the cathode voltage of the output stage. (75V on the driver plate means the cathode of the output stage will be above that, but not by a whole ton)
If the driver stage plate voltage is set high, then the result will be less available plate-to-cathode voltage on the 5998, which will mean less grid to cathode voltage for that same 30mA of current. This would actually increase the transconductance, but at the expense of potentially limiting the amount of drive voltage that you can apply to the 5998. In other words, you could lower the output impedance, but you'd also lower the maximum voltage output of the circuit in the process. This, unfortunately, would also raise the voltage across the big C4S load under the 5998, and the power dissipation would increase as that voltage crept up, which is not acceptable with the current design.
-PB
-
Tested my Amperex and I'm getting
Terminal 1: 105V
Terminal 5: 106V
do I still have to replace resistors and if I do ..will 1/10w 470ohm be the best choice ?,I have order some resistors earlier but I guess this are to much or should I give them a try ?
http://www.ebay.com/itm/161148100242?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1497.l2649
On another note ;tested my new arrival today (12BH7A)
Terminal 1: 55V
Terminal 5: 56V
EDIT; I just tested my NU 6SN7GT (via adapter ) and getting 88V
-
Yeah, ~105V is on the high side, 55V is very low, 88V is in the safe zone.
Those 470 Ohm resistors you linked to will work well with the E80CC. For the 12BH7, try a 200 Ohm resistor to bring the plate voltage up.
-PB
-
I replaced the resistors today (I didn't know how small they were :o )
Terminal 1: 81V
Terminal 5: 84V
I'm very happy with the outcome
-
Sweet, you're right in the groove with those voltages.
-
I replaced the resistors today (I didn't know how small they were :o )
Terminal 1: 81V
Terminal 5: 84V
I'm very happy with the outcome
Were there any audible improvements?
-
I have ordered some metal films for this mod it will be fun to try and is easily reversible the e80cc hardly gets removed anymore its defiantly on par if not slightly better than the best of my 12au7s
Out of interest what e80cc brands are you guys getting good results with? Using a Siemens E80CC here same plate structure and big o getter as the one below but marked only Siemens E80CC and the pins seem to be a brighter than normal silver colour. It also has a small silver V2 sticker on it?
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fcdn01.trixum.de%2Fupload2%2F7%2FA%2F7AXvV2NU1jJC135394892364P5614.jpg&hash=a7dd205c3fd92f9201651c6b186f3798894a881f)
-
They look like Tungsrams.
Have also tried Philips E80CC gold pin, rebranded by Mullard. This tube is one of my top five noval driver tubes.
-
Out of interest what e80cc brands are you guys getting good results with? Using a Siemens E80CC here same plate structure and big o getter as the one below but marked only Siemens E80CC and the pins seem to be a brighter than normal silver colour. It also has a small silver V2 sticker on it?
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fcdn01.trixum.de%2Fupload2%2F7%2FA%2F7AXvV2NU1jJC135394892364P5614.jpg&hash=a7dd205c3fd92f9201651c6b186f3798894a881f)
I'm currently using a Tunsgram e80cc which I found to jive amazingly with the hd650s (pictures attached)
-
Terminal 1: 60V
Terminal 5: 61V
YES !
believe it or not but most of my 6SN7s have similar soundstage on Crack to E80CC Amperex..love it ,this grey skirt VT-231 sounds very good
I'm very glad I can use both
-
So I finally got done with the mod, voltages check out and I'm extremely pleased with the results!
I ended up replacing the two R1 resistors (237ohm) on the small speedball boards with 470ohm resistors. I then wanted a way to switch back to the original set up for 12AU7 rolling, so I wired a SPST (single-pole single throw) in series with another 470ohm resistor, which was then wired in parallel with the newly installed 470ohm resistors. When the switch is turned on the resistance through R1 drops to 235ohm and voltages check out!
Thank you bottlehead community!
edit: In terms of sonic differences, I've noticed a more controlled low end combined with an increase in detail in the mids. Current set-up is tungsol 5998 + tungsram e80cc + hd650s.
-
Neat idea which I might steal it if you don't mind ;D
-
Steal away! That's what its all about!
-
great mod kscwuzhere,thanks for sharing
-
Absolutely! I got the idea from the ecc32 thread so I'd thank the contributors of that thread :)
-
Absolutely! I got the idea from the ecc32 thread so I'd thank the contributors of that thread :)
6SN7,12AU7 =( T1 75V / T2 76V )-very happy with this ;D
E80CC=( 85/86V )
12BH7A=(55/57V )
Thank you Sir !!
-
Absolutely! I got the idea from the ecc32 thread so I'd thank the contributors of that thread :)
6SN7,12AU7 =( T1 75V / T2 76V )-very happy with this ;D
E80CC=( 85/86V )
12BH7A=(55/57V )
Thank you Sir !!
Nice work! Love the finish on your build.
-
OK started looking for a pair of switches lots of choices but what voltage rating should I be looking at to keep things safe?
-
This: http://www.jameco.com/webapp/wcs/stores/servlet/Product_10001_10001_317287_-1 (http://www.jameco.com/webapp/wcs/stores/servlet/Product_10001_10001_317287_-1) is what i used
-
OK started looking for a pair of switches lots of choices but what voltage rating should I be looking at to keep things safe?
Voltage rating isn't important. The switch will have less than 2V across it, and technically will never be switched when any DC voltage is present.
I still do recommend gluing them to the PC boards.
-PB
-
OK started looking for a pair of switches lots of choices but what voltage rating should I be looking at to keep things safe?
Voltage rating isn't important. The switch will have less than 2V across it, and technically will never be switched when any DC voltage is present.
I still do recommend gluing them to the PC boards.
-PB
When applying the glue, put some on the pc board and wait ~2mins for the glue to dry a little. Then, place the spst in the appropriate position and hold it for another minute or so. This method will ensure minimal glue overapplication, which prevents the glue to splurge from the sides and will ensure a tight fit.
Gotta thank Paul for this was his advice to me!
-
Super thanks, will probably will use a dab of hot glue to fix the switch in place, already have the resistors the switches should turn up in a couple of days.
-
Super thanks, will probably will use a dab of hot glue to fix the switch in place, already have the resistors the switches should turn up in a couple of days.
Make sure to post some pictures when you get done with it :):)
-
Super thanks, will probably will use a dab of hot glue to fix the switch in place, already have the resistors the switches should turn up in a couple of days.
You might possibly want a glue with a higher melting point ...
-
Super thanks, will probably will use a dab of hot glue to fix the switch in place, already have the resistors the switches should turn up in a couple of days.
You might possibly want a glue with a higher melting point ...
Thanks Paul my glue gun is a industrial high temp one the glue melts at 195°C so should be fine but will take some readings with the inferred thermometer to check first.
-
OK no sweat, should not go much over 60C. I was thinking the home craft type.
-
So I replaced my resistors and installed the switches all went well and I am delighted with the results I am hearing. The music feels tighter more coherent with better textures and substance to it.
So a big thank you to those who have paved the way.
-
what reading are you getting when switched ? (12AU7/E80CC) ,looking good Jamie !
P.S. I used very light "rocker switches " so direct soldier worked great for me ,switches feel sturdy,all I need to do is lift the unit on one side without flipping to make the adjustment,this mod should be sticky in my opinion
-
Switched to run E80cc
T1 80.01
T5 79.98
Switch to run 12au7
T1 74.4
T5 76.7
-
better then what I'm getting -20v jump
-
I wonder why the difference? Perhaps its down to mains voltage 240 here.
-
Yeah I'm getting the following readings:
12AU7:
T1 74V
T5 76V
E80CC:
T1:84
T5: 82
Maybe try a difference switch?
-
Yeah I'm getting the following readings:
12AU7:
T1 74V
T5 76V
E80CC:
T1:84
T5: 82
Maybe try a difference switch?
that's pretty close to what I'm getting
6SN7,12AU7 =( T1 75V / T2 76V )* [88V before]-very happy with this ;D
E80CC=( 85/86V )*[105V before]
12BH7A=(55/57V )*[SAME Before]
I'm not sure if this have anything to do with the switch but as long as I'm in the "55~88v " window there is nothing to complain about :)
-
Those readings are perfect! How have you been enjoying the mod?
-
I love this mod ,wish 12BH7A voltage was little higher ,since this tubes sound awesome..
-
I just installed this mod today, with two 470ohm resistors running in parallel for normal function and a single 470ohm for the e80cc. I also tried the 12bh7 with my t1s running on each circuit, and found that I preferred it to my current Mullard ecc82. So, I've decided that I want 3 different resistance settings, one stock (270ohm), one @470 for e80cc's, and one @200 for 12bh7's.
Would throwing the stock R1 resistor back in and adding a 634ohm resistor on one throw of a SPDT to net 470ohms (e80cc), and a 768ohm resistor on the second throw to net 200ohms (12bh7) let me accomplish this? I figure that the off position will give me the stock resistance with a On-Off-On SPDT: http://www.jameco.com/webapp/wcs/stores/servlet/Product_10001_10001_72557_-1 (http://www.jameco.com/webapp/wcs/stores/servlet/Product_10001_10001_72557_-1)
Edit: Or it seems I could just add another 348ohm (I think?) resistor as the second "on" position parallel to the 470ohm I've already installed in order to achieve the 200ohm resistance for the 12bh7?
-
This mod is really brilliant, I've decided to go for it and hope I don't break anything in the process :)
Does the 470 ohm take the E80CC with any power tubes or just the 5998? If I decide to use the E80CC with a 6080, do I toggle the switch to the 237ohm or 470ohm? I love the E80CC + Bendix 6080WB combo for faster music.
-
I believe this mod will work with the 6080wb under the 470ohm resistance position. How would you compare the sound of the e80cc+5998 vs e80cc+6080wb? Which cans are you using?
-
I believe this mod will work with the 6080wb under the 470ohm resistance position. How would you compare the sound of the e80cc+5998 vs e80cc+6080wb? Which cans are you using?
With the E80CC, the 5998 has superior details, transparency and soundstage size. The 6080WB betters the 5998 in terms of dynamics, drive, and bass punch.
For jazz, vocals, and most slow music, I strongly prefer the 5998 + E80CC. With electronic and rock music, the 6080WB has the added drive and weight that makes it more enjoyable.
I'm intrigue by the 470 ohm mod and wonder how much improvement it brings to the sound pf the 5998 +E80CC combo. One problem I have with the 5998 +E80CC is that the gain is a bit too high and I have only a small play in the volume knob.
I'm using the Beyer T90 as my main set of cans. My other set is a Alessandro MS2i which doesn't match well with the Crack.
-
It's a very noticeable improvement with the e80cc resistance mod. I only decided to try it was after A/B'ing kscwuzhere's mod. I still don't prefer the e80cc to my Mullard ecc82 on my T1's, but for the 650s the difference made the e80cc a real winner IMO.
-
.
-
Specifically what improvement do you guys hear when moving from 237 ohm to 470 ohm resistor for the E80CC + 5998 combo?
I'll probably go get myself the 470 ohm resistors tomorrow to try out. Is it worthwhile to get some higher quality resistors since I'm at it?
-
I used just some ordinary 470 Ohm 0.25w Metal Film resistors £1.19 free delivery for 10 at that sort price its a throw away mod if you don't like it just stick the old ones back in. My original ones came out quiet easily just heated up one lead at a time and pulled it out with some forceps once the solder melted.
http://www.ebay.co.uk/itm/120531828406?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1497.l2649
-
Specifically what improvement do you guys hear when moving from 237 ohm to 470 ohm resistor for the E80CC + 5998 combo?
I'll probably go get myself the 470 ohm resistors tomorrow to try out. Is it worthwhile to get some higher quality resistors since I'm at it?
For my HD650s, it really tightened up the bass (after a/bing it with against the 237ohm, I found the old resistance to make the bass noticeably muddier) ,slightly improved the imagining/instrument separation and soundstage definitely widened.
-
Can anyone just confirm the 200ohm net resistance needed for the 12bh7? I haven't seen anybody put this on a switch yet and just wanted to make sure.
-
Can anyone just confirm the 200ohm net resistance needed for the 12bh7? I haven't seen anybody put this on a switch yet and just wanted to make sure.
You'll have to smoke it and see... The 12BH7 curves aren't well enough resolved to really make a good estimate of what the current should be at 1.5V of bias to get 75V on the plate.
As R1 drops, the plate voltage will increase. I might try something around 100 Ohms first. You can run the amp with a 12BH7 plugged in and no 6080, then adjust the resistance of R1 till you see ~75V. If you have a few different 12BH7's, you can then swap them all in the socket to confirm that they are all in the appropriate range.
Depending on how much current you end up running, a small heatsink or a beefier transistor may be a good idea.
-
Can anyone just confirm the 200ohm net resistance needed for the 12bh7? I haven't seen anybody put this on a switch yet and just wanted to make sure.
You'll have to smoke it and see... The 12BH7 curves aren't well enough resolved to really make a good estimate of what the current should be at 1.5V of bias to get 75V on the plate.
As R1 drops, the plate voltage will increase. I might try something around 100 Ohms first. You can run the amp with a 12BH7 plugged in and no 6080, then adjust the resistance of R1 till you see ~75V. If you have a few different 12BH7's, you can then swap them all in the socket to confirm that they are all in the appropriate range.
Depending on how much current you end up running, a small heatsink or a beefier transistor may be a good idea.
How should I go about installing a small heatsink, (where on the board would i do this)? Which transistor should I replace with a beefier one to run the 12BH7A with the proper resistance?
-
Start by determining the R1 that gives the correct plate voltage.
-
Alright, time to order a pack of resistors and start trying! Would testing these using some 21" alligator leads be alright just for multimeter purposes? I realize the leads are far too long for listening purposes, but would they be alright to keep the solder/unsoldering time down in testing?
-
I would just use a pot in the right range to dial in what you want. Then measure it when your done.
-
Haha well, I went ahead and threw the 470ohm resistors we had in parallel to each other to get about 160ohms with 3, and about 117 ohms with 4.
@160ohm
T1: 64.7V
T5: 69.5V
@117ohm
T1: 79V
T5: 75V
I'm about to give this a listen. I did add some ~1.5" leads in order to tack on resistors, so hopefully that doesn't make itself too obvious of a bottleneck.
Edit: Plugged in, this thing sounds crazy! Bass extension, detail retrieval, soundstage are all the best I've heard on the t1, even though the treble can be a bit more cutting than the Mullard ECC82. I must say though, the tubes are getting extremely hot with this setup, even after just 10 minutes or so of play. What can be done heat sink / transistor wise to fix this while still allowing for stock resistance use?
-
Looks like 147 Ohms would be a sweet spot.
This is about 6mA, which in an MJE-350 being cooled by air will give a temperature rise of 75C beyond whatever temperature is under the hood of the Crack. The MJE-350 has a maximum heat rating of 150C, but will eat itself in your case if that temperature goes beyond ~130-ish.
IIRC, the temperature measurements that we have seen have been between 30 and 40C, so adding 75C to that still leaves you in safe territory. Replacing the MJE-350's with MJE-5731A's (they face in the opposite direction) would decrease the temperature rise to ~47C, which gives a much higher safety margin (and longer life most likely).
-
Alright, I think I'll put some MJE-5731ag's on order, as it seems that the "a" variants aren't very available. When installing these, will the back still butt up to the heatsink in the same way, even given that "they face in the opposite direction" or will some orientation changes need to be made?
I plan to test these numbers with a Sylvania 12bh7 that I have coming in tomorrow, and then committing on the proper resistors to get a 3-way switch running properly for the e80cc, 12bh7, and 12au7
-
So guys if I wanted to try the E80cc tube I should "our its best" to replace my 237 ohm resistors with the 147 ohm ones for the lower voltage? I don't intend to install a switch just yet ..I'll try the tubes out for sound 1st.
I received some E80cc tubes today but haven't tried them yet.
-
The E80CC's need something along the lines of 470ohms across R1, which is why many of us have been replacing R1 with a 470ohm and putting a second 470ohm on a switch in parallel, so when switched on, the resistance totals 235ish, and both resistances are usable. The 12BH7's are the dual-triode tubes that we're finding require a resistance of ~147ohms.
-
When installing these, will the back still butt up to the heatsink in the same way, even given that "they face in the opposite direction" or will some orientation changes need to be made?
Butt up to which heatsink? These transistors go on the smaller PC boards, they do not go in place of the TIP50's on the bigger PC board.
-
The E80CC's need something along the lines of 470ohms across R1, which is why many of us have been replacing R1 with a 470ohm and putting a second 470ohm on a switch in parallel, so when switched on, the resistance totals 235ish, and both resistances are usable. The 12BH7's are the dual-triode tubes that we're finding require a resistance of ~147ohms.
Thanks :)
-
Edit: Plugged in, this thing sounds crazy! Bass extension, detail retrieval, soundstage are all the best I've heard on the t1, even though the treble can be a bit more cutting than the Mullard ECC82. I must say though, the tubes are getting extremely hot with this setup, even after just 10 minutes or so of play. What can be done heat sink / transistor wise to fix this while still allowing for stock resistance use?
Yeah, this is the difference between just plugging any old tube in and actually sitting down and figuring out what needs to be done to properly operate said tube.
I don't understand the question about heat and the stock resistors.
-
... This is about 6mA, which in an MJE-350 being cooled by air will give a temperature rise of 75C beyond whatever temperature is under the hood of the Crack. ...
For what it's worth, ever since we had problems with the early Foreplay III upgrade and the Paramount, I have designed for a maximum chip temp of 100C and assumed 60C under the hood - so a 40C maximum temperature rise.
-
That's a good reference, thanks for the reminder. It seems like an MJE-5731 should be OK on most line voltages. (I calculated for a drop from 200V to 75V, which is worst case scenario)
-
I don't know if anyone else has mentioned but this resistor mod has also completely removed the zero volume hum.
Also, really rather surprised but quite delighted to hear the audible improvements. As with most upgrades to the Crack, it's 'just the same but more'. Nothing I've done has removed or fundamentally altered the qualities of the amp, it just keeps opening out and stretching a little further in every direction.
-
I don't understand the question about heat and the stock resistors.
I just wanted to make sure these new transistors on the small speedball boards will work just fine with everything else in "stock" configuration, which seems to be the case.
I wanted to throw this out there to anyone thinking of playing with the 12bh7's too- I just grabbed a vintage Sylvania 12bh7 to check the operational differences against a Tung-Sol 12bh7, and got the following readings:
@117ohm, w/SYLVANIA
T1: 64V
T5: 69V
So it seems the proper resistance definitely changes between these 12bh7 tubes.
That all said, the soundstage on my t1s has never been so massive and three-dimensional with any other tube.
-
That all said, the soundstage on my t1s has never been so massive and three-dimensional with any other tube.
Try retrofitting for the 5687.
-
Just threw in a couple more resistors,
@93ohm, w/SYLVANIA
T1: 74.8V
T5: 70.8V
Try retrofitting for the 5687/
I'm intrigued - any recommended branding? I'll try and scoop one of these up for play. Would resistance be the only mod needed for this dual-triode as well?
-
I think you could run 10mA on the 1.5V LED, but you'd want the MJE5731A for that. Do also note that the 5687 has an annoyingly different pinout, so you'll have to do some work there as well.
-
I finally installed the 470 ohm mod and am very pleased with the result. Bass is more controlled and the mids are more detailed and dimensional.
I wonder if anyone with this mod will actually use the 235 ohm switch position. With this mod, is there any 12AU7 tube that is superior to the E80CC?
When using a 6AS7G (with E80CC), should the switch remain at 470 ohm or changed to 235 ohm?
-
When using a 6AS7G (with E80CC), should the switch remain at 470 ohm or changed to 235 ohm?
The switch is to optimize the plate voltage of the E80CC, so if the E80CC is plugged into the amp, the 470 Ohm resistor should be there.
-
Just had a crazy idea.
I wonder if it would be possible to put a voltage regulator between the anode of the input tube and the grid of the cathode follower, setting the output voltage at 75V. This would provide a steady bias to the 6080 at the level required by the circuit.
As I understand it, the voltage into the regulator would need to be greater than the sum of the output voltage and the voltage drop across the regulator. This in turn would require revision of the operating point of the input tube. I wonder if a plate load could be chosen that would
1. Result in a linear operating point for 12AU7 (and variants), 12BH7 and E80CC
2. Result in a plate voltage that met the requirements for running the said regulator
If this could be done, you could swap input tubes between these three families without having to turn off the amp and flip switches: Tube rollers heaven. Wonder how much it would cost to implement.
Does anyone have any thoughts to add. Do you think it is possible?
-
I wonder if it would be possible to put a voltage regulator between the anode of the input tube and the grid of the cathode follower, setting the output voltage at 75V. This would provide a steady bias to the 6080 at the level required by the circuit.
Voltage regulators have low impedance and require capacitors at the input and output. Therefore, you would be shorting all the signal to ground at the plate of the 12AU7, and the grid of the 6080. On top of that, the regulator will work its butt off to null out any signal present at the input from appearing at the output.
The plate voltage of the driver tube can move around without a whole ton of consequences, it's when it goes down into the 50's or up past about 90V that things may get out of whack. Being obsessed with getting exactly 75V is neither particularly healthy for one's psyche nor a concomitant to getting better sound out of the Crack.
-
Told you it was a crazy idea!
Thanks for the learning experience.
-
I wonder if anyone with this mod will actually use the 235 ohm switch position. With this mod, is there any 12AU7 tube that is superior to the E80CC?
I still preferred my ECC82 to the E80CC when I installed the mod, it was simply a gateway to the 12BH7 for me. I'd say that these results are very dependent on your other equipment - E80CC is an excellent tube for the hd650s, but not my t1s (IMO).
-
I wonder if anyone with this mod will actually use the 235 ohm switch position. With this mod, is there any 12AU7 tube that is superior to the E80CC?
I still preferred my ECC82 to the E80CC when I installed the mod, it was simply a gateway to the 12BH7 for me. I'd say that these results are very dependent on your other equipment - E80CC is an excellent tube for the hd650s, but not my t1s (IMO).
Interesting I am very pleased with the mod with my T1's and use the E80cc for roughly 90% of my listening.
-
Yeah I liked the e80cc at first but now I'm finding it a bit to much , It has bit more detail but its brighter and gives me ear fatigue to quick with my hd600s
I like a smoother warmer sound and going back to 12au7 sounds better for me :) I will try it again tho :o
-
plus i get more volume control with the 12au7
-
I will disconnect my pio capacitors and try it running with just the mkp films tomorrow, I did spend some time today swapping between my 12au7 and e80cc types today its very close but defiantly prefer the e80cc in my set up. A whisper more dynamic and revealing.
Running gec 6as7g brown base power tube
12au7's inputs rolled today
RCA
Tung Sol black glass
Tung Sol JTL
Mullard 4003 box plate
-
I will disconnect my pio capacitors and try it running with just the mkp films tomorrow, I did spend some time today swapping between my 12au7 and e80cc types today its very close but defiantly prefer the e80cc in my set up. A whisper more dynamic and revealing.
Running gec 6as7g brown base power tube
12au7's inputs rolled today
RCA
Tung Sol black glass
Tung Sol JTL
Mullard 4003 box plate
You should definitely try rolling with the 12bh7a. I got a chance to listen to onelivewire's set-up the other day and his T1's are a new beast with the Sylvania 12bh7as, seriously. The detail and dynamics of the e80cc is definitely all there with the 12bh7a but the 12bh7a is much more forgiving and has a larger soundstage (width and depth). Imaging also improved significantly with the T1's!
-
I'll just add my 2 cents, though ksc pretty much laid out my feelings. The 12bh7a at the proper resistance is not as bright as the e80cc, and definitely has the same level of imaging and dynamics, which was lacking from 12au7s. I couldn't handle the obvious sibilance in lesser recordings that came with the e80cc + T1s. The e80cc matched the 650s very well though.
I just finished installing a 3 way switch to let me run the 12bh7a, e80cc, or 12au7s, so I can try to do a direct A/B with the former 2 tubes soon.
On another note, the MJE5731As are in the mail, too! I've been looking at the 5687 pinout, and here are the differences I see between that and the 12bh7/au7 pinout:
12au/bh7
1 - Plate (2)
2 - Grid (2)
3 - Cathode (2)
4 - Heater (2)
5 - Heater (1)
6 - Plate (1)
7 - Grid (1)
8 - Cathode (1)
9 - Heater (2/Midtap)
5687
1 - Plate (2)
2 - Grid (2)
3 - Cathode (2)
4 - Heater (2)
5 - Heater (1)
6 - Cathode (1) [Redirected from stock pin 8]
7 - Grid (1)
8 - Heater (Midtap) [Redirected from stock pin 9]
9 - Plate (1) [Redirected from stock pin 6]
So pin 8 must go to 6, 9 must go to 8, and 6 must go to 9.
I realize I may have displayed that poorly, but does it seem to be correct? If these are the only changes necessary for the 5687, couldn't these three leads go through some 2-way on-on switches, with the biasing LED on it's own switch between pins 6 & 8? It seems previous 5687 modders used blue LEDs to achieve a different bias, which I'm guessing would eliminate 12au7s and the lot for tube rolling - but it seems the 5687 could run just fine with the stock HLMP6000s? Sorry for all the questions!
-
Yeah, that will get the pinout to be correct.
For the 5687 to reach 75V on the plate with 1.5V LED's on the cathode, you'd need to run 20mA of plate current! (The stock 12AU7 is about 3mA).
12mA of plate current and just over 2V of bias on the 5687 makes for a more reasonable operating point.
-
Yeah, that will get the pinout to be correct.
For the 5687 to reach 75V on the plate with 1.5V LED's on the cathode, you'd need to run 20mA of plate current! (The stock 12AU7 is about 3mA).
12mA of plate current and just over 2V of bias on the 5687 makes for a more reasonable operating point.
In order to achieve this bias and install it on a switch so that we can roll the crack with the 12au7, should we replace both the 1.5V Red LEDS on A3 and A8 with 2V Blue LEDS (being on A3 and A6 instead) or just one of the LEDS? Thank you PB!
-
You would need a switch to swap out both LED's on the tube socket to something else.
This gets a little bit beyond the bounds of what should be switchable on an amplifier, if you go far enough down this trail, you'll turn your Crack into a tube tester.
-
You would need a switch to swap out both LED's on the tube socket to something else.
This gets a little bit beyond the bounds of what should be switchable on an amplifier, if you go far enough down this trail, you'll turn your Crack into a tube tester.
What LEDs should we switch it out with? Would the following LED be okay?
http://www.digikey.com/product-detail/en/HLMP-CB1A-XY0DD/516-2275-1-ND/2428165
-
That LED would work with about 5mA of plate current. (~175 Ohm R1)
One thing to note, that LED will be effing bright. I wouldn't hesitate to spray them with black paint (mask the leads) before you install them.
-
LEDs are highly variable in their voltage even if they are said to be the"same color". They are also quite variable in how much noise they generate, and in their dynamic impedance. The LEDs we use seem to be (so far - knock on wood!) fairly quiet and fairly repeatable. We haven't measured or otherwise tested other LEDs, and the specifications for most LEDs give a wide range of voltages and no noise data, so the specs are no real help.
-
Would installing these LEDS in series with a switch or in parallel with a switch allow us to achieve the correct plate voltage? Otherwise we'll just have to chance it with some LEDs.
-
Are there any other desirable Speedball part changes when swapping the MJE350 for MJE5731A other than the transistor facing the opposite way?
-
Would installing these LEDS in series with a switch or in parallel with a switch allow us to achieve the correct plate voltage? Otherwise we'll just have to chance it with some LEDs.
Paralleling diodes doesn't really word, you have to have just one or the other.
-
Since there may be a chance that the LEDs may be noisy, I want to order a variety of LEDS, should the following be a good selection?
http://www.vishay.com/leds/list/product-81641/
http://www.digikey.com/product-detail/en/HLMP-CB1A-XY0DD/516-2275-1-ND/2428165
http://www.jameco.com/webapp/wcs/stores/servlet/Product_10001_10001_334749_-1
http://www.electron.com/led-1.8mm-blue-130lb7c.html
http://www.electron.com/led-5mm-blue-520lb7c.html
Thank you and sorry for the plethora of links!
-
The first one is a surface mount part, and you don't want to deal with that.
I'd suggest just starting with one pair, I haven't yet run into LED's causing issues in this situation unless the circuit is very, very sensitive.
-
Alrighty, I'll get those LEDs and toggle switches ordered and update ASAP! Thanks PB
-
Can't wait! You guys really kick ass for providing this level of customer interaction. I can wholeheartedly say that I wish all business could make themselves a part of the community with such social outreach. I can't say thank you enough!
-
Once I get this mod done for the 5687, would I be able to roll the 7119's with the crack too?
-
Yes.
-
Awesome! Onelivewire and I are gonna give this thing a shot once all the parts come in and I'll update the thread accordingly.
-
Switches/biases have been installed, just waiting for the proper resistors and tubes to come in now! Exciting stuff!
-
Ok well I find myself liking the two 12bh7's that arrived in the post this morning I have been listening to them on the 12au7 setting but am mindful of how well the e80cc sonics improved with the resistor mod once better matched so once again interested in following progress here. :)
-
I thought I would join the club and add a three way switch to SB allowing me to select between three resistance values for E80CC 12AU7 and 12BH7.
I checked actual vales for resistance before installing speedball and they came to the expected values. I inserted both an E80CC and a 12AU7 (don't have a 12BH7 yet) and measured voltages of between 75 and 80 on the correct switch settings.
Listening on the E80CC everything is sweet (switch position off). On the other two settings there is a horrible hum in both channels that never varies despite changing volume via Submissive or volume pot.
Did I mess up design, implementation or do I have a dodgy part?
-
On the other two settings there is a horrible hum in both channels that never varies despite changing volume via Submissive or volume pot.
Did I mess up design, implementation or do I have a dodgy part?
Is this with the E80CC or the tubes that those positions are supposed to feed?
This is part of the risk of switching R1's.
-
1. Tested with 12AU7
2. Is the hum part of the risk of this mod? If so, what causes it and why isn't everyone reporting the hum? If not, what exactly is the risk you are referring to?
For the record, I will take risks all day if I can learn from them, especially with a bargain amp like the Crack.
-
2. Is the hum part of the risk of this mod?
Yes
If so, what causes it and why isn't everyone reporting the hum?
Not everybody is going to build this the same way. (There's no manual)
-
Paul, do you and the rest of the staff support modding in our community?
Is there anyone else reading this who is willing to help an idiot and a Loser (ie, me) trouble shoot?
For instance, what are the common causes of hum, how do you determine the cause and how do you eliminate it?
-
Support is maybe a little vague. We can make suggestions and assist with ideas, but for us to be able to sort out hum from a modification we haven't tried, we would have to build a Crack and attempt to modify it to exactly your specifications.
As far as hum from switching R1's, the only experience I have was finding noise when long wires were used to try different resistor values.
-
Totally understand Paul.
You often go above and beyond the call of duty. Thanks.
Have tested the problem further changing cathode follower.
6080 - hum on startup, but goes away after approx 20 seconds
6AS7G - hum on startup, but goes away after approx 20 seconds
7236 - hum on startup, increasing in volume after approx 20 secs
5998 - Hum on startup, increasing in volume after about 20 secs
7802 - Hum on startup, increasing in volume after about 20 secs
All the tubes types with persistent hum have a higher mutual conductance than those that do not. In another thread Paul mentioned that mutual conductance is a "tendency to oscillate".
What would the sonic symptoms of oscillation be? My hum would be around 120 Hz (speaking as a musician, not an electrician).
Do you think my hum could be a symptom of oscillation?
I have read about grid stoppers in this context? Might a grid stopper help?
Thanks again.
-
For the 10 cents that a pair of resistors will cost you, go ahead and give the stoppers a shot.
-
MJE 350
OK. I have just read a thread re hum problem where the problem was a faulty transistor in the speedball.
My Crack is a new build. I installed switch in Speedball prior to installing Speedball in Crack. The Crack worked fine with all the mentioned tube types prior to installation of speedball. Perhaps, therefore, the problem is with the Speedball and not the switch as I ordered my amp about six weeks ago and received it about two weeks ago.
Will replace MJE 350 and see what happens.
-
Ok well I find myself liking the two 12bh7's that arrived in the post this morning I have been listening to them on the 12au7 setting but am mindful of how well the e80cc sonics improved with the resistor mod once better matched so once again interested in following progress here. :)
I just finished modding the crack to run the 5687 and the improvement is absolutely incredible. The soundstage and the imagining on my 650s has enhanced tenfold! The 5687 is an even bigger improvement than from the e80cc to the 12bh7! I'll post pictures soon :)
-
Just had a look at the data sheet. Crickey!
Looks like it will draw about 20 mA with a bias of -1.5 V and a plate voltage of 75 V. I assume you have made a number of mods to make this work?
Looking forward to seeing what you have done.
-
Just had a look at the data sheet. Crickey!
Looks like it will draw about 20 mA with a bias of -1.5 V and a plate voltage of 75 V. I assume you have made a number of mods to make this work?
Looking forward to seeing what you have done.
Yeah, I upgraded the transistors on the small speedball boards and had to bias some LEDs. I installed everything on switches so I can switch between 12au7/e80cc/12bh7/5687 :)
edit: Oh and the pin out on the 5687 is a little different too so some mods had to be made there too! The more and more I listen to this tube the more I'm being blown away because every aspect in the sound has improved, most noticeable difference is in the soundstage(incredibly whole-bodied) and imaging(instrument separation).
-
Ok well I find myself liking the two 12bh7's that arrived in the post this morning I have been listening to them on the 12au7 setting but am mindful of how well the e80cc sonics improved with the resistor mod once better matched so once again interested in following progress here. :)
I ended up changing the original SPST set up to a SPDT set up with an On(235ohm)-Off(470ohm)-On(150ohm) to run the 12au7, e80cc, and 12bh7a, respectively. I'd recommend modding it with the SPDT design for ease of use :) For the resistance of the 5687, I'm in the process of installing a spst in parallel to one of the ON positions on my SPDT toggle to reach the correct resistance; not the most elegant, but itll do.
-
I have my crack running the Raytheon 5687 with my T1s right now, and I can say that the imaging has definitely improved. Bass response may be a bit reduced by comparison to the Sylvania 12bh7a, but the detail retrieval seems to be improved. The biggest difference is, as Paul suggested, the 3-dimensionality of the soundstage. Where other tubes would generally place all of the musicians relatively equidistant from me in different sized rooms, the 5687 is able to place some instruments right by my ears and others far away in the room. I don't know if anyone else uses binaural recordings, but the percussion-jazz in "Explorations in Space and Time" really came to life through this tube.
Edit: I should add that I am running R1 at 117ish ohms to get 77V on Terminals 1 and 5.
-
Changed out the MJE350's; installed the 470 switch-resistor mod, and big sound change like everyone else. Bass, detail, and hearing new things. My problem and surprised no one else mentioned it, is a big jump in gain. I can only just crack my volume pot. Would appreciate suggestions on the the best way to proceed. Potentiometer or a resistor in the volume pot leads? Is this normal. Have others had this gain jump? Ordered a Sylvania 12bh7 today so lots more to go. Beats building more headphone cables( although I do have some Litz wire coming}.
-
The 12BH7 shouldn't be that noticeably different than the 12AU7.
There are, however, likely cases of other tubes that are close to a 12BH7 being relabeled as a 12BH7. Some computer dual triodes that are kind of a cross between a 12AT7 and a 12BH7 come to mind. I wouldn't adjust anything until you have a few 12BH7's to triple check.
-PB
-
So the mod is finally done and my crack now works at the correct resistances for the 12au7/e80cc/12bh7a/5687.
In order to wire this mod to run the 5687 you're going to need 2 dpdts and the following leds and transistors:
http://www.digikey.com/product-detail/en/HLMP-CB1A-XY0DD/516-2275-1-ND/2428165 - blue LEDS for 5687
MJE5731A - install these in place of the transistors on small speedball boards in reverse orientation
You're going to want to cut the connections on the following lugs of the 9 pin socket:
A6 - coming from terminal 1
A9 - coming from B7
RED LEDs
Connect terminal 1 to the one of the inputs of the dpdt. The two outputs for this position should be going to A6 (for 12au7) and A9 (for 5697), respectively.
Connect the wire from B7 to the other input of the dpdt. The two outputs for this position should be going to A9 (for 12au7) and A8(for 5687), respectively.
Cut a piece of wire and connect it from the center lug of the 9 pin socket to the two inputs of the second dpdt.
Connect the RED leds to the two outputs of one of the ON positions on the dpdt going to A3 and A8 (make sure to orient correctly).
Connect the BLUE leds to the two outputs of the other ON position on the dpdt going to A3 and A6 (make sure to orient correctly).
Once you've installed this all thats left is to make sure the 5687 runs at the correct resistance so adding another resistor with a switch in parallel may be necessary. Luckily for me, my Raytheon 5687 runs at 70V at 140ohms which is the exact resistance that my RCA black 12bh7a needs in order to run at 76V.
So I have a 470 ohm resistor at R1 with a ON-OFF-ON DPDT in parallel to it. One of the ON positions has another 470ohm resistor to make R1 run at 235ohm. The other ON position has a 200ohm resistor in parallel to the 470ohm, making R1 140ohms, which runs both my 12bh7a and the 5687.
I'm sorry if the instructions were a little unclear so please let me know if anyone has any questions. I've attached some pictures but they wont be of most help since its getting very crammed up under the chassis. My Crack now has a cool blue glow when the 5687 is running.
Sonically, the 5687 is my favorite tube out of the lot. Its sound signature makes the hd650s sound very balanced with a quite incredible soundstage (the biggest difference with the 5687 tube). My second favorite is a tie between my RCA 12bh7a and my Telefunken e80cc which has a more lush bass than the 5687 but has noticeably worse imagining and soundstage.
-
Just to clarify before I order a bunch of tubes, with the mod working with the 5687 mean that I'd be able to roll both the 7119s and e182cc's with my crack now? Or would further modifications be needed?
-
Here's my build. I still prefer my Yellow label Sylvania 12bh7a (long grey plates, o-getter) to the Raytheon 5687 for the T1s, for its warm sound signature, despite slightly less articulate imaging.
Album: http://imgur.com/a/TTaME (http://imgur.com/a/TTaME)
Underside
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FroJHG0V.jpg&hash=07e9ccdb3200f913275d07a2fe0495f0bf27ab05)
Pinout Switches
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2Fu0Uuovs.jpg&hash=e33fc5b49e69ef454f7d40260d2101cf9b26b443)
Variable Resistance Switches
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2Fb1YB5f2.jpg&hash=eb4023c15dfc37f9d42dcf4451b3c448ae926679)
Running 12BH7A
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FPX5VyNV.jpg&hash=d9f5aaa70b331492152dc8c5a15b8537ab3e28c7)
Running 5687
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FRK0ek6i.jpg&hash=a8d3a2d12f62fda1ea91ac0a6ce4865e3687799f)
-
Thank you onelivewire for pictures that are actually helpful!
-
Yes a picture is worth a thousand words. Nice job by the way you have implemented the changes much neater than I could have done.
-
Hi
to help those like me who need to see the schematic of the 7119/5687 mod, I've made this based on the excellent kscwuzhere work, Kudos to him !! :D
I have not kept the center lug of the 9 pin socket wired, as I was not quite sure it still made sense. The wire was taken directly from terminal 3. Could someone enlighten me on this ?
[EDIT]I've finally retained this central pin to simplify the soldering. It may have a shielding effect too.
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fwww.4shared.com%2Fdownload%2F2XQN8tW1ce%2F7119_wiring.jpg%3Flgfp%3D3000&hash=117d962bc18c08e83cec2132f18b3ae85a060b69)
-
Hi
to help those like me who need to see the schematic of the 7119/5687 mod, I've made this based on the excellent kscwuzhere work, Kudos to him !! :D
I have not keeped the center lug of the 9 pin socket wired, as I was not quite sure it still made sense. The wire was taken directly from terminal 3. Could someone enlighten me on this ?
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fwww.4shared.com%2Fdownload%2FtfFupbJkce%2F7119_wiring.jpg%3Flgfp%3D3000&hash=c2e0b937c5940ef505350d19334f300b2167e6c7)
Thank you snarri! That's absolutely perfect!!
Only reason I did t take the cable directly from terminal 3. Was because I thought the center lug of the 9 pin socket needed to be grounded. I just added a cable from the center to the dpdt. My guess is that your technique should work!!
Also gotta thank onelivewire (we brained this together) and pb for being an amazing guide!
-
My simple E80CC/ 12AU7 switching mod. (One for each small SB PCB of course)
Even these small dip switches are more than within spec requirements.
Uses the already available PCB wiring.
Thanks to Caucasian, Jamie and the rest of You for great advice.
-
Terminal 1: 60V
Terminal 5: 61V
YES !
believe it or not but most of my 6SN7s have similar soundstage on Crack to E80CC Amperex..love it ,this grey skirt VT-231 sounds very good
I'm very glad I can use both
This is what i want to do with my CRACK run 6/12SN7's with it but first run it stock to be familiar with the stock sound then MODDING Time.I will need more INFO on these MOD later.Thanks!
-
My simple E80CC/ 12AU7 switching mod. (One for each small SB PCB of course)
Even these small dip switches are more than within spec requirements.
Uses the already available PCB wiring.
Thanks to Caucasian, Jamie and the rest of You for great advice.
With parallell and 12AU7 I get 72v on terminal and 76 on term 5
In off position with E80CC I get 105v and 107 v (470ohm 0,125 w)
Is this acceptable? Or even understandable?
-
With parallell and 12AU7 I get 72v on terminal and 76 on term 5
In off position with E80CC I get 105v and 107 v (470ohm 0,125 w)
Is this acceptable? Or even understandable?
So your values for the 12AU7 are fine but the e80cc ones are on the high side. Try messing around with different resistors (I bought a pack of resistors ranging from 10ohm - 900ohm) using alligator clips and test the voltage at T1 and T5. Not every tube performs exactly the same for example my buddy's Sylvania 12bh7a needed 91ohms to run at 75V and my RCA 12bh7a needed 140ohm to run at 75V; same tube variety, different resistance requirement. I'd say anything from 60-85V is on the safe side!
-
OK, finally here we are 8)
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fwww.4shared.com%2Fdownload%2F3SMRnnbkce%2FCrack-1.jpg%3Flgfp%3D3000&hash=21f667fe585dd04f217440ebdbdad0ebd6cc775e)
After hours of surgical operations, dealing with really tiny crowded spaces where binocular vision is sometime impossible, making desoldering or soldering a challenge, broken short stock 22AWG PTFE wires, everything is now in place and my crack has achieved its ultimate mutation on the input section :
- half-DPDT with a 330R in parallel of the stock 237R still in place to achieve the 138R needed for my 12BH7A (82V). I don't have E80CC, so no need for another resistor combination. Chances are good I won't have to change anything for the 7119 as it happened for kscwuzhere.
- 2 x DPDT switches for the Standard / 7119 configuration. The wiring with the leds (HLMP-6000 / HLMP-CB1A-XY0DD) is made according the drawing I posted a few posts upward. I finally kept the central pin of the socket wired to the ground and restarted from there to the switch. I modified the drawing accordingly.
- 2 x MJE5731A to replace the MJE350. These parts are incredibly hard to find right now, nowhere in stock in the major companies with months of wait, don't know why. I finally ordered on ebay to someone having motorola parts with short pins (new ?). Whatever, they work as intended (reverse from MJE350 !)
- In front on the right, the Valab 23 stepped attenuator 100k. Behind are the capacitors I'll describe in the capacitor porn thread http://bottlehead.com/smf/index.php/topic,4744.90.html (http://bottlehead.com/smf/index.php/topic,4744.90.html).
Modifying the crack during the initial assembly would have been so much easier ! I've replaced the wires I needed to change with 20 or 22AWG PTFE, for a neater result, the overall diameter being smaller than with PVC. Everything eventually at risk has been isolated with shrink tube.
I now just have to wait for my 7119 PQ in transit…
-
OK, finally here we are 8)
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fwww.4shared.com%2Fdownload%2F3SMRnnbkce%2FCrack-1.jpg%3Flgfp%3D3000&hash=21f667fe585dd04f217440ebdbdad0ebd6cc775e)
After hours of surgical operations, dealing with really tiny crowded spaces where binocular vision is sometime impossible, making desoldering or soldering a challenge, broken short stock 22AWG PTFE wires, everything is now in place and my crack has achieved its ultimate mutation on the input section :
- half-DPDT with a 330R in parallel of the stock 237R still in place to achieve the 138R needed for my 12BH7A (82V). I don't have E80CC, so no need for another resistor combination. Chances are good I won't have to change anything for the 7119 as it happened for kscwuzhere.
- 2 x DPDT switches for the Standard / 7119 configuration. The wiring with the leds (HLMP-6000 / HLMP-CB1A-XY0DD) is made according the drawing I posted a few posts upward. I finally kept the central pin of the socket wired to the ground and restarted from there to the switch. I modified the drawing accordingly.
- 2 x MJE5731A to replace the MJE350. These parts are incredibly hard to find right now, nowhere in stock in the major companies with months of wait, don't know why. I finally ordered on ebay to someone having motorola parts with short pins (new ?). Whatever, they work as intended (reverse from MJE350 !)
- In front on the right, the Valab 23 stepped attenuator 100k. Behind are the capacitors I'll describe in the capacitor porn thread http://bottlehead.com/smf/index.php/topic,4744.90.html (http://bottlehead.com/smf/index.php/topic,4744.90.html).
Modifying the crack during the initial assembly would have been so much easier ! I've replaced the wires I needed to change with 20 or 22AWG PTFE, for a neater result, the overall diameter being smaller than with PVC. Everything eventually at risk has been isolated with shrink tube.
I now just have to wait for my 7119 PQ in transit…
Absolutely beautiful implementation Snarii! Now just for those 7119's.......
-
So your values for the 12AU7 are fine but the e80cc ones are on the high side. Try messing around with different resistors (I bought a pack of resistors ranging from 10ohm - 900ohm) using alligator clips and test the voltage at T1 and T5. Not every tube performs exactly the same for example my buddy's Sylvania 12bh7a needed 91ohms to run at 75V and my RCA 12bh7a needed 140ohm to run at 75V; same tube variety, different resistance requirement. I'd say anything from 60-85V is on the safe side!
Thank You and it is a good point. I only did this mod because it made sense. 470 in parallell makes 235 ohms and the switch would be easy and small. I am sure I will find something.
-
Thank You and it is a good point. I only did this mod because it made sense. 470 in parallell makes 235 ohms and the switch would be easy and small. I am sure I will find something.
It's all about trial and error. Try using a 500, 550, or a 600ohm resistors for the e80cc position!
-
Or how about installing a self bias circuit?
-
Hi all,
my two 7119 PQ (yes, I like to always keep some spare) arrived this morning. Very good shape, like new indeed. They were made in Heerlen in 1966. One last visual wiring check, we're never too sure, and power. A few seconds then the 2 two blue leds light up, then the small speedbal cards, OK !
Time to check the plates. As expected, the 138ohm (= 237 // 330) 12BH7A resistor mod worked for the 7119, as I get 77.3V on the plate of the 2 triodes. Now let's get the tube the time to cool down, then put a 6H13C on its slot, and lets play some jazz….
Woww !! Stunning !
Absolutely no regrets spending some time integrating the 5687/7119 mod !
We are really one step beyond every progresses I've made so far installing all these gears. Transparency, liveliness of the music, improved soundstage, presence and focus of placement of the instruments, sensation of space, … what to say more, this 7119 tube is killer and pairs beautifully with the HD650. The 12BH7A was my favorite, but it never went this far, even properly driven by a good plate voltage.
3 months after my complete DIY beginning on the crack, I think my tuning journey has now ended, thanks to all at Bottlehead for this wonderfull amp, and those of you on the forum making me try something else, now time to enjoy the music !
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fwww.4shared.com%2Fdownload%2F_ivNbdkNce%2FCrack-7119.jpg%3Flgfp%3D3000&hash=98be522cf91b279a7cfb8987f66fd32b7d0b4ce7)
-
Hi all,
my two 7119 PQ (yes, I like to always keep some spare) arrived this morning. Very good shape, like new indeed. They were made in Heerlen in 1966. One last visual wiring check, we're never too sure, and power. A few seconds then the 2 two blue leds light up, then the small speedbal cards, OK !
Time to check the plates. As expected, the 138ohm (= 237 // 330) 12BH7A resistor mod worked for the 7119, as I get 77.3V on the plate of the 2 triodes. Now let's get the tube the time to cool down, then put a 6H13C on its slot, and lets play some jazz….
Woww !! Stunning !
Absolutely no regrets spending some time integrating the 5687/7119 mod !
We are really one step beyond every progresses I've made so far installing all these gears. Transparency, liveliness of the music, improved soundstage, presence and focus of placement of the instruments, sensation of space, … what to say more, this 7119 tube is killer and pairs beautifully with the HD650. The 12BH7A was my favorite, but it never went this far, even properly driven by a good plate voltage.
3 months after my complete DIY beginning on the crack, I think my tuning journey has now ended, thanks to all at Bottlehead for this wonderfull amp, and those of you on the forum making me try something else, now time to enjoy the music !
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fwww.4shared.com%2Fdownload%2F_ivNbdkNce%2FCrack-7119.jpg%3Flgfp%3D3000&hash=98be522cf91b279a7cfb8987f66fd32b7d0b4ce7)
Absolutely beautiful build! After I installed the 7119 for my hd650s I felt the same way! I felt like I maxed out the hd650s potential and that's why I decided to upgrade to the hd800s haha! My journey has begun all over again.
-
It's all about trial and error. Try using a 500, 550, or a 600ohm resistors for the e80cc position!
So its not just something wrong with my E80CC?
-
So its not just something wrong with my E80CC?
Definitely not, no tubes are identical. Your e80cc can run at 108V or whatever value you were getting and your crack will sound wonderful but for a couple of bucks worth of experimentation I thought it was worth it to find the correct resistance to run the e80cc tube at the correct plate voltage.
-
I will certainly look into it. Maybe some resistor sockets of sorts for easier interchangability.
-
So its not just something wrong with my E80CC?
If you have unusually high voltages with the same resistor that is working for the rest of the crowd, you may have a tired E80CC.
-
If you have unusually high voltages with the same resistor that is working for the rest of the crowd, you may have a tired E80CC.
My thoughts exactly. Was supposed to be a NOS and everything.
By the way, I have read the 12AU7 datasheets :
http://free-hosting.infodiv.com/audiotube/small_signal_triode/small_signal_triode0018/small_signal_triode0018-01.html
To me, it seems a plate voltage of 100v should be a usual operating voltage, not 75v.
Again, there must be something I do not understand.
-
By the way, I have read the 12AU7 datasheets :
http://free-hosting.infodiv.com/audiotube/small_signal_triode/small_signal_triode0018/small_signal_triode0018-01.html
To me, it seems a plate voltage of 100v should be a usual operating voltage, not 75v.
Again, there must be something I do not understand.
There are literally an infinite number of operating points for any tube. The plate curves for a tube will reflect this. (They are usually a few pages into the datasheet).
The number on page 1 are maximum ratings. The 100V rating is an insulation rating.
-
There is a listed operating point at 100v; it's just another example. The example operating points are just that - not recommendations, just examples of the infinite possibilities. In this case, the 100v example is also at zero bias, a point that will have significant nonlinear grid current and is unsuitable for audio. It might reflect use as a vertical deflection amplifier in B&W televisions.
-
Thank You both very much. I feel privileged to learn so much in so brief time!
-
Just dropping in to give some updates on my 5687 experience. Recently I've become fond of kscwuzhere's Amperex 7119, which certainly became less sibilant after burn-in. However, just a couple days ago I received a Sylvania Gold Brand 5687 (Gold Pin variety). With the t1s, this tube is incredible. Easily takes the cake as my favorite tube for t1s. Soft, yet articulate, this tube truly takes away just about all the harsh glare that I find in bad recordings, while sacrificing nearly no articulation. I highly recommend this tube to any t1 owners running this mod.
-
There are literally an infinite number of operating points for any tube. The plate curves for a tube will reflect this. (They are usually a few pages into the datasheet).
The number on page 1 are maximum ratings. The 100V rating is an insulation rating.
Some feedback.
Decided another attempt at ordering an E80CC instead of changing my resistors.
Received an unmarked Tungsram where the plate voltages are 78ish and 81ish with 470ohms
On my now heavily hotrodded crack with GEC brown based 6080, I am convinced the soundstage widened, upper mids and treble much clearer and stronger but no annoying sibilance. I use HD580s and am a very happy camper.
-
I just received an RCA bronze plate 5687 as research eluded to it's warmth, which was something I was looking for my hd800s. After letting the NOS tube warm up for a couple of hours it has become my new favorite with the hd800s! Sonically, mid bass to mids have become much more lush and sound more whole-bodied and less airy than the 7119. Treble seems to be rolled off a little more than the amperex but is slightly more sibilant (won't surprise me if this goes with more burn in). The amperex 7119 bass is tighter and has an overall quicker presentation than the rca.
For more rock/metal orientated as well as poorer recorded music I find myself preferring the warmer RCA 5687s. Whereas for jazz, electronic and more "tamed" music, I find that the Amperex is my go to tube (particularly enjoy the speed the hd800's provide with these genres).
-
Here's a sneak peek at my solution for easy switching between 12AU7, 12BH7, and E80CC tubes. It fits right into the existing resistor holes on the Speedball boards.
With no jumper installed:
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FxdKfJYe.jpg&hash=e66254306896f246eb970b2d2cde994010e2a687)
With jumper installed for 12AU7 operation:
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FGYlebXF.jpg&hash=c8ebfd64aa169c4177f7b3db8bbc0a4e963e280a)
The back:
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FusZwaXp.jpg&hash=2282576145b06ce5fb489d596076d8671d59f515)
I will post more pics once I get the 2nd one made and both installed.
-
how cool is that...John
-
Creative solution!
-
Here's a sneak peek at my solution for easy switching between 12AU7, 12BH7, and E80CC tubes. It fits right into the existing resistor holes on the Speedball boards.
With no jumper installed:
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FxdKfJYe.jpg&hash=e66254306896f246eb970b2d2cde994010e2a687)
With jumper installed for 12AU7 operation:
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FGYlebXF.jpg&hash=c8ebfd64aa169c4177f7b3db8bbc0a4e963e280a)
The back:
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FusZwaXp.jpg&hash=2282576145b06ce5fb489d596076d8671d59f515)
I will post more pics once I get the 2nd one made and both installed.
Any update with the mod? I'm curious to see the results!
-
Well, I've been sitting on this Rogers(Canadian?) D Getter 12BH7A that I got with a bunch of other stuff about a year ago now :P I was digging through my electronics kit, and found some 150ohm resistors so I decided why not give it a go and put them together.
Glad I did.
In a stock SB crack the 12BH7A was sitting in the low 60V on T1 and T2. Dropping R1 to 150Ohm brought up T1 and T2 to ~70V. Acceptable given I just soldered the resistors them straight in without much thought to reverting. Honestly, re-reading this thread from the start tonight indicated to me that I was likely be happy with this choice anyhow.
I gotta say before modification the 12BH7A showed a ton of potential, but was a bit fat in the transients and flabby on the bottom end at ~60V. Bringing the plate up to 70V put this tube in the zone! Still using the stock MJE350s, so I'm hoping their not going to go into runaway with the extra load to dissipate... because I'm not reverting. NOPE. ;-) No sir. I won't like it.
Whole new texture appearing across the spectrum compared to the 12AU7s I've give a fair shake over the last year. I really don't roll tubes that much these days. I may just setup switching because I'm curios about the E80CC, but that's about it. I've read the E80CC sacrifices some bottom end for clarity, so I'm likely prefer this 12BH7A.
Strangely with the 12BH7A I find the RCA 6080WA stock tube totally new. My 5998 has been my staple since I got it - even selling the 412A in favour of this particular 5998. But the 6080 has a midrage texture reveal that is a bit lost on the 5998. Interesting.
-
Oh yes.
Thanks CB and Paul for your input in this thread. Its appreciated.
-
Oh yes.
Thanks CB and Paul for your input in this thread. Its appreciated.
Why not install a SPDT ON-OFF-ON so that you can roll between the 12au7, 12bh7a and the e80cc? That's what I did.
After rolling all these tubes at the right resistance I decided to retrofit for the 5687 and 7119 and I must say, I'm never going back to the 12au7, 12bh7a, and e80cc; the 5687 destroys these tubes. It has way better imagining and the soundstage is massive. Details and data retrieval are as good as the e80cc but it has the lower end warmth that the 12b7a had. After installing the 7119/5687 I found myself questioning why the Crack wasn't designed to run this tube in the first place. The way i retrofitted it also allows me to switch back to the 12au7 but I havent found myself ever wanting to.
-
What are the potential (haha) drawbacks of trimpots in this application? Cycle life is one I have seen so far. But I'm looking for a fit with effectively infinite turn cycles. Say 100ohms in series with a 500ohm trimpot.
http://www.mouser.com/ds/2/428/accutrim1240-7897.pdf
Edit: OY! Price could be one issue.... An the Power Derating Curve...
-
Oh yes.
Thanks CB and Paul for your input in this thread. Its appreciated.
150 Ohms is about 6mA. The voltage across the C4S is roughly 80V, and with 6mA of current, you'll dissipate 1/2 a Watt through the MJE-350. A tiny heatsink probably wouldn't hurt here, and the MJE5731A will handle this with aplomb.
-
What are the potential (haha) drawbacks of trimpots in this application? Cycle life is one I have seen so far. But I'm looking for a fit with effectively infinite turn cycles. Say 100ohms in series with a 500ohm trimpot.
A trim pot would function here, though you'll end up with a fair amount of useless adjustment if you parallel the two when the trimmer is set to a low resistance. A 100 Ohm resistor in series with a 500 Ohm/10 turn trimmer may work a bit better.
These are indeed expensive little buggers, and they would have to be glued down so that adjustments could be made reliably.
-
I think I can just use some of the free space on the C4S board to get a 100 Ohm resistor in series with a trim pot easily enough.
-
Why not install a SPDT ON-OFF-ON so that you can roll between the 12au7, 12bh7a and the e80cc? That's what I did.
After rolling all these tubes at the right resistance I decided to retrofit for the 5687 and 7119 and I must say, I'm never going back to the 12au7, 12bh7a, and e80cc; the 5687 destroys these tubes. It has way better imagining and the soundstage is massive. Details and data retrieval are as good as the e80cc but it has the lower end warmth that the 12b7a had. After installing the 7119/5687 I found myself questioning why the Crack wasn't designed to run this tube in the first place. The way i retrofitted it also allows me to switch back to the 12au7 but I havent found myself ever wanting to.
The MJE5731A and some cheaper cerment trim pots are on their way in the mail. You've definitely peeked my interest in the 5687 and they seem to be readily available. I have a about 5 flea-bay 12BH7's on their way as well. It was hard to turn down 4 for $10 and a NOS Tung-Sol '59 milspec for $20. I'll give the 12BH7 a bit of a listen and then likely hard wire for the 5687 once I get the urge again.
I'd also like to re-cable my HD800, but I think I'll wait for the mainline. Then this crack stuff might become moot. :)
This photo initially lead me to believe an HD800 cable kit might surface from BH, but has yet to materialize:
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fbottleheadnginx.bottleheadcorp.netdna-cdn.com%2Fwp-content%2Fuploads%2F2014%2F01%2Fmainlinehd800.jpg&hash=2bdd49d68e1d87de04d2f0a1ed7bbceba04181bd)
-
It's too difficult to hand build a reliable connection with those terrible plugs. My cable breaks each time we take it to a Head Fi meet. They really need a molded plastic strain relief.
-
So I've been lurking on this thread and willing to try modding my crack for the 12BH7a and 5687's tubes but would still allow me to switch to the 12au7's.
From my understanding, I would need to get the following parts:
2 x DPDT On-Off-On
1 x DPDT On-Off-On
2 x MJE5731A
2 x HLMP-CB1A-XY0DD LED's
I'll also need to test out different resistors to see what matches the tube for my 12BH7's.
Could you guys tell me if I have any missing or incorrect parts?
Thanks!
Jim
-
So I've been lurking on this thread and willing to try modding my crack for the 12BH7a and 5687's tubes but would still allow me to switch to the 12au7's.
From my understanding, I would need to get the following parts:
2 x DPDT On-Off-On
1 x DPDT On-Off-On
2 x MJE5731A
2 x HLMP-CB1A-XY0DD LED's
I'll also need to test out different resistors to see what matches the tube for my 12BH7's.
Could you guys tell me if I have any missing or incorrect parts?
Thanks!
Jim
Everything seems right to me. I recommend getting more RED leds and blue leds as they tend to break easily during construction. A simple resistor pack may be a good idea, as well as alligator clips to test them out.
-
Thanks kscwuzhere,
Do you know where I can find the voltages for 12BH7's. What range is good to run them at?
-
Thanks kscwuzhere,
Do you know where I can find the voltages for 12BH7's. What range is good to run them at?
My 12bh7a ran perfectly at 70v with 140ohms
-
My 12bh7a ran perfectly at 70v with 140ohms
Should all 12BH7a's run around that voltage?
-
Should all 12BH7a's run around that voltage?
I would be satisfied if ten 12BH7's that tested good hit that voltage (or thereabouts). If only one 12BH7 was used to dial in the R1 value, then a worn or even an overly fresh specimen could throw things off a bit for the average tube.
-
My first 12BH7A (unknown test value) hit 70v/75v @ 150ohm on R1. I dropped in MJE5731A to replace the MJE350 and they seem to work well enough as drop in replacements with no change to the output voltage. Experimentally putting in cermet trim pots tonight. The trim pots I grabbed are 500ohm with 20 turns and seem to be stable to around .1 ohm at room temperature. We'll see how that goes once in the amp. I'm thinking in series with a 75ohm effectively in R1 to have more room to play with.
It definitely be nice to be able to dial in each side of the triode!;-)
-
It definitely be nice to be able to dial in each side of the triode!;-)
You really want to set each half to have the same current running through it, not whatever current is required to produce the same plate voltage as the other side.
-PB
-
You really want to set each half to have the same current running through it, not whatever current is required to produce the same plate voltage as the other side.
-PB
Okay. Can you expand on that?
-
Okay. Can you expand on that?
If you have a really dead triode and you adjust the current strongly to counteract that, then plug in a good tube, your voltages will be way off. There's also the general philosophy that if you don't need to provide an adjustment for a particular aspect of an amplifier, you probably shouldn't.
The Crack is reasonably tolerant of wandering driver stage voltage, so that's not as much of a concern as it might be in a circuit like the Paramount 2A3. Or, for another example of matching plate voltage, the Mainline offers an adjustment to set the plate voltage to the ideal value, which allows the amplifier to produce the most power.
-
How do I set up the 12BH7's using one DPDT? It's hard to tell from the picture (from Snaari's mod) where the wires go and where to put the resistors. I should have went the easier route by installing two SPDT's instead but I thought I could make it easier by just having one switch for the 12BH7's. Oh well. If you guys can help me install this or show me some clearer pictures to how the wires are connected I would really appreciate it. Thanks!
Jim
-
You really want to put one switch on each board.
-
http://www.4shared.com/download/3SMRnnbkce/Crack-1.jpg?lgfp=3000
This is the picture of Snarri's mod where he was able to use one DPDT. If I could get a better angle of the DPDT it would be very easy.
I don't really want to buy a pair of SPDT's if I don't need one.
-
If you need to pop a board off, it will be pretty hairy to get all that apart.
-
If you need to pop a board off, it will be pretty hairy to get all that apart.
That's very true. I guess I can work on the 7119 switch in the meantime if I order some SPDT's. The thing is I probably won't need it after installing the 7119's.
7119 here I come. :)
-
Great news. Just installed the switches for the 7119's and 12AU7's!
Now to test the voltages... What should I be testing and what readings should I be getting between the two tubes?
-
After a week of modding I finally got my SPST's and DPDT's installed to run the 12BH7/7119's!
For the resistance mod, I added a 330 ohm instead of a 350 ohm because I did not have it in my resistor pack. Voltages came out a bit higher than the range mentioned.
T1/T5:
7119: 88/88
12BH7: 81/78
12AU7: 100/100
Are these voltages okay?
-
The 100V kind of sucks, are you getting that with the stock resistors, or the 330 Ohm resitors? (The 330's are not to be used with the 12AU7)
-
The 100V kind of sucks, are you getting that with the stock resistors, or the 330 Ohm resitors? (The 330's are not to be used with the 12AU7)
I have the resistance mod turned off for the 12AU7's so it would be with the stock 235 ohm resistors.
-
So I tried using a higher ohm resistor to try to get my voltages down. First I tried to add a 22 ohm in series with my 330 ohm but my voltage ratings only went down slightly. (Btw, before I added the 22 ohm in series with the 330ohm my 12AU7's were around 85/85. In my last post I think I may have forgotten to turn the switch off XD.)
T1/T5
7119: 86/86
12BH7: 76/73
12AU7: 84:84
I replaced the 22 ohm with a 47 ohm resistor but got near ratings. Then I replaced the two resistors in series with a single 470 ohm resistor
T1/T5
7119: 83/83
12BH7: 75/72
12AU7: 82/82
Should I keep replacing it with higher resistors till I get between 70 - 80?
-
The high voltage just indicates a 12AU7 that isn't super fresh.
-PB
-
Thanks PB.
-
Just installed the Tungsram E80CC alongside a Sylvania 7236 (replaced 237's with 470 Ohm Resistors on the A and B boards) and my background hiss was immediately gone (faulty 12AU7 I guess). It sounds glorious.
I still get crackling at high volumes, but I don't play at those insane volumes anyway so it's all good.
One question, I was only supposed to replace those resistors, correct? 1 237 resistor on A and B? Nothing else?
-
One question, I was only supposed to replace those resistors, correct? 1 237 resistor on A and B? Nothing else?
Yes.
-PB
-
Thanks to Snarii's diagram of kscwuzhere's mods, I've modded mine using a DPDT switch for the bias current and 4PDT to switch between 5687 and 12au7 wiring topology. The DPDT switch allows me to switch the current on both channels simultaneously.
The resistors I used are 470 ohm in the off position for E80CC. 475 ohms in parallel for 12AU7, and 196 in parallel for 12BH7. I'll have to do some measurements with the 5687 to see if I can get away with 138 ohms.
I'll likely regret it long term, but I epoxied the 4PDT switch to keep it firmly in place since I didn't want to wait for the special tape I ordered, which happened to arrive today that I used for the DPDT. I was worried about the adhesion of run-of-the-mill double-sided tape when used in a high temperature environment, so I bought 3M VHB 5952 tape, which withstands up to 250F long-term exposure. At $15 for 1/2" x 5yds it was on par price-wise with the better stuff Home Depot carries. And 3M does not publish maximum temperatures for the consumer-grade stuff, they only spec application temperature.
I'm running a Sylvania 12BH7 at the moment which sounds even better when biased properly. My Tungsol 5687's will be here this week and I'm hoping to be wowed based on the reviews I've read.
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FsDWfrixl.jpg&hash=7fd81ae88b3b0a2a1583433aa4f8beae058d4097)
-
Thanks to Snarii's diagram of kscwuzhere's mods, I've modded mine using a DPDT switch for the bias current and 4PDT to switch between 5687 and 12au7 wiring topology. The DPDT switch allows me to switch the current on both channels simultaneously.
The resistors I used are 470 ohm in the off position for E80CC. 475 ohms in parallel for 12AU7, and 196 in parallel for 12BH7. I'll have to do some measurements with the 5687 to see if I can get away with 138 ohms.
I'll likely regret it long term, but I epoxied the 4PDT switch to keep it firmly in place since I didn't want to wait for the special tape I ordered, which happened to arrive today that I used for the DPDT. I was worried about the adhesion of run-of-the-mill double-sided tape when used in a high temperature environment, so I bought 3M VHB 5952 tape, which withstands up to 250F long-term exposure. At $15 for 1/2" x 5yds it was on par price-wise with the better stuff Home Depot carries. And 3M does not publish maximum temperatures for the consumer-grade stuff, they only spec application temperature.
I'm running a Sylvania 12BH7 at the moment which sounds even better when biased properly. My Tungsol 5687's will be here this week and I'm hoping to be wowed based on the reviews I've read.
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FsDWfrixl.jpg&hash=7fd81ae88b3b0a2a1583433aa4f8beae058d4097)
Beautiful modding! What headphones are you driving? What kind of music do you listen to? I've tested a lot of 5687/7119 varieties on many different headphones with my mod, I could potentially have some recommendations depending on your headphones/music choice.
-
Beautiful modding! What headphones are you driving? What kind of music do you listen to? I've tested a lot of 5687/7119 varieties on many different headphones with my mod, I could potentially have some recommendations depending on your headphones/music choice.
The cans I listen to most often are Sennheiser HD800. My music tastes are quite varied although I tend to gravitate towards jazz or vocals. Classic rock, rock, some classical, most types of jazz, new age, hip hop every now and then. It's easier to say that I don't typically listen to rap, opera, or country.
On the way are NOS tung-sol 5687 and RCA 5687. In my repertoire of input tubes:
12AU7: RCA long black plate, Westinghouse, Sylvania (stock), RCA clear tops (currently in my MC240)
5814: CBS-Hytron (angled D-getter), GE
12BH7: Sylvania, RCA (currently in my MC240)
Mullard CV4003
My output tubes:
GE 6080 (stock), Sylvania GB6080, Mullard CV2984, Tung-sol 5998
My favorite input tubes are the Sylvania 12BH7 and CBS-Hytron 5814. For output I like the 5998 and GB6080.
-
The cans I listen to most often are Sennheiser HD800. My music tastes are quite varied although I tend to gravitate towards jazz or vocals. Classic rock, rock, some classical, most types of jazz, new age, hip hop every now and then. It's easier to say that I don't typically listen to rap, opera, or country.
On the way are NOS tung-sol 5687 and RCA 5687. In my repertoire of input tubes:
12AU7: RCA long black plate, Westinghouse, Sylvania (stock), RCA clear tops (currently in my MC240)
5814: CBS-Hytron (angled D-getter), GE
12BH7: Sylvania, RCA (currently in my MC240)
Mullard CV4003
My output tubes:
GE 6080 (stock), Sylvania GB6080, Mullard CV2984, Tung-sol 5998
My favorite input tubes are the Sylvania 12BH7 and CBS-Hytron 5814. For output I like the 5998 and GB6080.
I run the Hd800s too! My fav combination of all time is the Bendix 6080wb slotted cross columns with either the Amperex 7119 PQ or the RCA 5687. They jive really well with the hd800s. The Amperex 7119 is my go to for jazz, or anything not quite as brutal as heavy progressive rock or some metal in which my RCA 5687 is my go to. The bendix outshines the GEC 6as7g, WE421a, and tungsol 5998 with the HD800s imo, the data retrieval is unmatched.
-
Two of my 5687 tubes arrived today. With 138 ohms I'm running 74.5 volts, so I also can use the same resistor for 5687 and 12BH7 (didn't actually measure 12BH7, but assumed 138 was the proper value based on what I read here).
Wow this 5687 runs hot. I'm showing 180F with an IR thermometer vs 125-130 for all the other tube flavors.
My initial impression is better imaging, well controlled bass, maybe a more luscious/liquid midrange. This is with my Gold Brand 6080.
-
Well, I decided to go the extreme route and made a new speedball PCB to accommodate a switch and various resistors to support 3 different bias points. I didn't feel comfortable with the taped switch and wanted something neater.
Here's the bare PCB:
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FA5d69jSl.jpg%3F1&hash=63ee2c6e4bffcdf166031be3f923b2ba33ea20c3)
Stuffed with parts:
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FR31SnPZl.jpg&hash=2c835e653f50a5070e200d22fe37c4bf9b22fabb)
Installed in place of the dual PCBs:
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2F1kGMm0ll.jpg%3F1&hash=0d654c43411f70e72e5a3a5d655d5625708346fc)
The fit wasn't perfect (had to dremel the screw holes a little as I didn't realize the 3rd standoff would interfere), the pads for some of the parts are on the small side, the holes for the diodes are too big, but it worked first try. It's also slightly simpler requiring only one input (B+) and one ground.
I just got my Tungsram E80CC tube in today and I'm listening to it with a Bendix graphite plate 6080 as I type this. The E80CC and 5687 both sound better than the various 12AU7. I'll have to switch back to the CBS-Hytron 5814 at some point since I recall liking it a lot too, as well as the Sylvania 12BH7.
-
Good stuff!
-
Very nice layout, love the symmetry of it :)
-
incredible!!
-
Very cool!
-
The fit wasn't perfect (had to dremel the screw holes a little as I didn't realize the 3rd standoff would interfere), the pads for some of the parts are on the small side, the holes for the diodes are too big, but it worked first try. It's also slightly simpler requiring only one input (B+) and one ground.
Interesting. I'd love to know what tools/service are required to create a custom PCB for low scale count like that. A very clean solution. I believe the original PCB was one piece like this as well... correct me if I'm wrong, I saw a single PCB for the C4S circuit for the input tube on a Crack previously on here and prefer the look.
I assume the third stand-off you refer to is the middle one that mounts the C4S for the output tube. My first thought would have been to notch out space for it on the long edge rather than the two screw holes. Is there room for that? From the photos it looks possible.
-
Interesting. I'd love to know what tools/service are required to create a custom PCB for low scale count like that.
There are numerous companies on the net that will create boards from your gerber files in small volumes at affordable prices. Some companies offer free software to allow you to create boards, but I haven't tried them since I have Orcad schematic capture & layout.
I assume the third stand-off you refer to is the middle one that mounts the C4S for the output tube. My first thought would have been to notch out space for it on the long edge rather than the two screw holes. Is there room for that? From the photos it looks possible.
Notching out the top would work too. You just need to keep away from the bottom side trace that carries B+.
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2F3rpOBbql.png&hash=9ebd020be347327f6a2b312fb5a2a94a55712dab)
-
In order to wire this mod to run the 5687 you're going to need 2 dpdts and the following leds and transistors:
http://www.digikey.com/product-detail/en/HLMP-CB1A-XY0DD/516-2275-1-ND/2428165 - blue LEDS for 5687
Is your amp silent when using the 5687? I substituted the above referenced LED with a different one with identical Vf (@20ma) but 2 orders magnitude lower luminous intensity, and although my voltages are right where they should be, I get a very faint hum when no signal is present. The hum is there with 3 different 5687 tubes, so I figure it could be the LED since I've seen mention of noisy LEDs before. The amp is dead silent when switched to use with 12AU7.
-
Is your amp silent when using the 5687? I substituted the above referenced LED with a different one with identical Vf (@20ma) but 2 orders magnitude lower luminous intensity, and although my voltages are right where they should be, I get a very faint hum when no signal is present. The hum is there with 3 different 5687 tubes, so I figure it could be the LED since I've seen mention of noisy LEDs before. The amp is dead silent when switched to use with 12AU7.
I get no hums whatsover, its black as can be.
-
I've made another PCB to clean up my 5687 switch. It helps to clean up the implementation shown here (http://bottlehead.com/smf/index.php?topic=5989.136) because you no longer need multiple wires to certain connections. I haven't installed it yet because I don't have any more of the solid core wire that came with the kit, and I usually end up breaking one or more of them because the solid core doesn't like to be bent more than once. I'll install it once I have extra wire on hand.
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FviOpCJjl.jpg&hash=b86f5a36f1686f351e4d05fc77980463f021084a)
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FB8I3Jywl.jpg&hash=e344eb5acba04d24b408ff331061262a30eb1058)
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FpUeLoVfl.jpg&hash=09ac318005f0add053dd6a88c06538518ca4c4f8)
-
Beautiful work my man! I know what I'm gonna do now.
-
Any updates? I wish i could make a PCB myself.
Everybody still in love with their 5687 or got something better?
Deciding on pin-out switch mods.
-
I just tried out a 12BH7A and love it, however, voltage at T1/T5 are low, 50, and 53 - considering replacing the R1 resistors for lower resistance to bring the voltage up - one question (that may have been answered already, sorry if it was):
Is it unsafe to run it at low voltage? what is the danger if so?
-
Is it unsafe to run it at low voltage? what is the danger if so?
It's not unsafe, but that voltage biases the 6080 in the Crack, and if this voltage is too low, the tube will not be properly biased, and you won't get optimal performance out of the amplifier.
-
It's not unsafe, but that voltage biases the 6080 in the Crack, and if this voltage is too low, the tube will not be properly biased, and you won't get optimal performance out of the amplifier.
Understood, I'll wait until I get my new headphones to see if I still like the 12BH7A as much as I currently do before making any changes.
Currently running HD 650's, in the mail a HD 800 S is on the way.
You guys are fond of the HD 800's from what I've seen around, any interest in the new variant?
-
Hello from Switzerland
First, i want to say thank you to the pioneers of this topic (especialy kscwuzhere,onelivewire, Snarii, and Maxhawk) and of course thank you to the Moderators in this Forum. With the information in this thread i was able to create (let create) two pc boards to handle the diffrent voltatges on t1 and t5 and to switch between stock and 5687/7119 tubes.
Here some pictures with a lousy smartphone cam:
(https://img5.picload.org/image/wicrorw/dsc_0139.jpg)
Hm... in this picture the blue light shines like a house with freaky christmas decoration. In reality its not that intense.
(https://picload.org/image/wicroog/dsc_0136.jpg)
As you can see (or not) i drilled a hole in the chassic plate to fix the small pc board. Spontaneous i was not able to get some nylon standoffs therefore i did it that way. I would not recommend this, because when you or someone else switch between the stock/5687 mod accidentally you could possibly break a tube. In my case the amp is far away from strangers hands and the switch is behind the input tube, so i got that goin for me :)
A note to the post #211 from maxhawk. On the small pc board to switch between stock/5687 there is a solder joint named B+. On the big pc board and on the two small ones of the the speed ball uprgrade the B+ solder joints are connected to 2U. But in my understanding this solder joint should be connected with T1 (1U). It was a bit confusing for me :)
Best regards
-
Another 5687 switch board here.
Really like this this tube (Amperex 7119 PQ), lush, big soundstage with loads of details.
Worth it IMO.
All three (5687, 12bh7 and e80cc) are special sounding tubes in there own way.
Nice flavors or "sound seasoning" compared to the standard 12au7's.
-
I'm hoping one of you technically savvy forum members can help. On Head-Fi I bought a modded Crack from a BH forum member -- great build and very enjoyable sound. What I can't get from the builder is a clear explanation about which way the switches should be set for 12AU7 and which way for E80CC. Here are pictures of the resistor switch assembly he devised, taken by the builder of the amp. If you can use these pictures to deduce the proper switch positions for 12AU7 vs E80CC, I would appreciate it. Especially if you can explain in "for dummies" terms (as I readily confess I am). Thanks in advance.
6SN7,12AU7 =( T1 75V / T2 76V )-very happy with this ;D
E80CC=( 85/86V )
12BH7A=(55/57V )
Thank you Sir !!
-
Those switches are way too small. You should upgrade to something like this.
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fwww.hometrainingtools.com%2Fmedia%2Fcatalog%2Fproduct%2Fcache%2F1%2Fimage%2F9df78eab33525d08d6e5fb8d27136e95%2FE%2FL%2FEL-SWITCH2.jpg&hash=59cd93e1f4f7277a448a8fe87073756bfa22b9ff)
-
Those switches are way too small. You should upgrade to something like this.
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fwww.hometrainingtools.com%2Fmedia%2Fcatalog%2Fproduct%2Fcache%2F1%2Fimage%2F9df78eab33525d08d6e5fb8d27136e95%2FE%2FL%2FEL-SWITCH2.jpg&hash=59cd93e1f4f7277a448a8fe87073756bfa22b9ff)
I'm actually thinking of something a little more retro and heavy duty...
-
I'm hoping one of you technically savvy forum members can help. On Head-Fi I bought a modded Crack from a BH forum member -- great build and very enjoyable sound. What I can't get from the builder is a clear explanation about which way the switches should be set for 12AU7 and which way for E80CC. Here are pictures of the resistor switch assembly he devised, taken by the builder of the amp. If you can use these pictures to deduce the proper switch positions for 12AU7 vs E80CC, I would appreciate it. Especially if you can explain in "for dummies" terms (as I readily confess I am). Thanks in advance.
In the speedball circuit there's a series resistor from B+ to the 2N2907 transistor that sets the current flowing through the tube. The stock 12AU7 uses a 238 ohm resistor while the E80CC requires 470 ohms. In the photo you can see that the switch is connecting a 470 ohm resistor in parallel with the resistor on the board (which looks to be 470 ohms). When the switch is open it's set for an E80CC. When the switch is closed, it's set for a 12AU7.
I'm not familiar enough with rocker switches to tell you which position is which (I only know toggle switchs). You can take a DVM and measure the resistance on the switch between the middle and right terminal. 470 ohms = E80CC and 235 ohms = 12AU7.
-
In the speedball circuit there's a series resistor from B+ to the 2N2907 transistor that sets the current flowing through the tube. The stock 12AU7 uses a 238 ohm resistor while the E80CC requires 470 ohms. In the photo you can see that the switch is connecting a 470 ohm resistor in parallel with the resistor on the board (which looks to be 470 ohms). When the switch is open it's set for an E80CC. When the switch is closed, it's set for a 12AU7.
I'm not familiar enough with rocker switches to tell you which position is which (I only know toggle switchs). You can take a DVM and measure the resistance on the switch between the middle and right terminal. 470 ohms = E80CC and 235 ohms = 12AU7.
Thanks very much, Maxhawk. I will dig out the VOM and give it a try. Thanks again.
-
marked
-
For those planning on the upgrades described here. A word on 12BH7 as i've found from testing 10+ 12BH7 the optimum is a bit lower then the 140 ohm often advocated. This current trough the tube, still functions fine with 5687/7119/182CC with 3,2V bias.
For the R1 value's -> with E80CC i use 475 ohm, for 12AU7's the 237ohm from standard Crack for 12BH7 (mind the MJE 5731A transistor) I use 113 ohm. With your normal E182CC it sits at 85V with the 8mA of current, with weaker ones this goes over 90V, so beware.
With 12BH7 and E182CC on the same R1 with different bias point's. it's a bit of a balancing act, but this works perfectly.
But there's always more tubes to experiment with.
If i'd wanted to try a more current demanding tube to get in the optimal voltage range -> lower R1 even further? Can i do this?
What's the maximum current that Speedball and the powersupply will handle?
And how do i calculate from R1 value -> current?
-
R1 = 0.855/current is the formula. You don't have to be perfect on the resistor value, just close. R2 needs to pass about 10% of the current your setting, or 1mA, whichever is more.
So, the power transformer in the Crack can supply a TON! of HV current. The first limiting factor will be those 270 ohm 5W resistors. The current circuit uses about 75mA of current and I wouldn't want to run much more than about 80 before the life of those resistors will drop. If you wanted to run more current, you could swap out the 270R/5W resistor that sits on top of the power transformer with a 300R/10W resistor, then make the other 270R/5W resistor into a 200R/5W resistor. You can reliably run 90mA through a 200R/5W resistor, and the 300R/10W resistor won't get all that toasty.
The other issue is the dissipation of Q2 up front. The MJE5731A is good for about 1W of dissipation, so even at 10mA I wouldn't be too worried about heatsinking it.
-
Thank you PB. For your detailed explanation.
This sure helps a lot. I'll get my Crack up to 10mA and leave it at that. That's perfect for the tube in mind.
It's the BL63. Not many use it with Crack as it's expensive, takes too much heater current (1.3A!) for the Crack to provide, and this much plate current will cook the transistor you mentioned. And there's no detailed datasheet / curves available, so we have no clue how it'll settle in Crack. This is a true recipe for disaster.
So, that's why i've decided to try it in my Crack :) . With A2293 as output pair, there's 1.6A left for the driver and the BL63 became an option. Still it lacked any data.
I've added a curve i traced on this tube. You can see why i'm planning on 10mA, it comes right in the suggested plate voltages with -3,2V bias (blue led). This is the same as with E182CC/5687 and can be used with the flick of a switch. I've now tried it at 7.5mA and this turn out to show 55V on the plate with the normal red LED bias.
For those interested in these curves. I've also added E80CC and ECC82 to compare. Do note, the ECC82 doesn't appear to be very healthy.
-
The BL63 traces would be easier to look at if the Y axis only went up to 30mA.
The Crack-a-two-a power transformer would be up to the job of heating that tube as a substitute.
-
Thank you for the feedback Paul. Did just do another curve with the lower plate current setting on the screen.
The C2A is on my wishlist. When funds permit it's my next BH purchase.
Here's the BL63 curves (wow these look good) with a 6SN7 as comparison.
-
Just wondering if anyone has applied the 12AU7/E80CC/12BH7 switch mod to the newer version of the Speedball that uses the single PCB for the driver. Any hints?
Thanks
-
Any hints?
The circuit and part names are identical, so it should be exactly the same to build this out on the newer Speedball.
-
On the data sheets and in engineering practice, it has long been standard to refer voltages to the cathode - so "plate voltage" technically means the voltage difference between plate and cathode. Similarly, "grid voltage" is the difference between the grid and cathode voltages (which is why it's usually a negative number).
This stems from the days (more than a hundred years ago) when tube circuits usually grounded the cathode, and were powered by three batteries - the A battery heated the filament/cathode, the B battery supplied current to the plate circuit, and the C battery set the grid voltage. That has survived in the use of "B+" for the positive high voltage from the power supply, and in the terms Vb for plate voltage and Vc for the grid voltage. In more academic literature, especially international literature Va for anode voltage, Vg for grid voltage, and Vk for cathode voltage.
Originally the term for voltage was "electromotive force" a.k.a. EMF, and voltages were Eb, Ec, or Ea, Eg, Ek. That's how I learned Ohm's Law, and I am still struggling to remember to write V instead of E. Similarly, current was measured as an "intensity" which is how we came to use I for current.
You may have noticed the substantial opportunities for confusion here! :^)
-
.....
You may have noticed the substantial opportunities for confusion here! :^)
:-[ ...
-
How do you calculate the size of the Va of the 6080? Is it simply a matter of subtracting the Va of the 12AU7 from the total B+?
What would be called the "plate voltage" of the 6080 would be the difference between plate and cathode voltage on the 6080.
This is influenced by the plate voltage that appears at the 12AU7, whether you have the Speedball installed, and how potent the particular 6080 is that you're using.
-
You would start at the plate curves for the 12AU7. Without the Speedball, you would draw a 22.1K load line that starts at about 175V DC, then estimate where that would hit the imaginary 1.57V grid voltage line. Go straight down from there and you get your 12AU7 plate voltage.
With the Speedball, you would draw a horizontal line at 3.2mA and estimate where that would hit the imaginary 1.57V grid voltage line, then go straight down from there to estimate the plate voltage.
The 6080 is a little less conventional, but for DC operating points and load lines it isn't so bad. You can draw a 3K loadline on the 6080 curves that hits 175V. Now on that line, you must find the point where the grid voltage magnitude (make the number positive), plus the 75V you already have, plus the plate voltage on the X axis equals your B+ (175V). Working out some terms, that is the magnitude of grid bias plus the plate voltage equals about 100V. On the curves, this happens at about 70VP and 30V of bias, so just over 30mA of current.
With the Speedball, you draw a horizontal line at 30mA and find the spot where grid bias magnitude and plate voltage add up to about 100V.
If the plate voltage on the 12AU7 moves, then you have to redo the analysis. If your line voltage moves, you have to redo the analysis. I prefer to be happy with estimates!
-
Yup, that's a decent estimate.
-
Bias = 30
We give bias in voltage. The bias with the Speedball is a variable to solve for. Current is held constant at 30mA. The bias voltage is the conclusion to the work, not the starting point.
Va - Vk = 70 (determined from curves where plate voltage + bias = 100)
Yes, plate to cathode voltage on the 6080 is about 70V. From the curves, you would get from this that bias voltage is about 30V.
Vg = 70 (supplied from plate of 12AU7)
Vg - Vk = Bias
70 - Vk = 30
Therefore Vk = 40
If the grid voltage is 70V and the bias voltage is 30V, then the actual DC voltage at the cathode is 100V.