Bottlehead Forum
Bottlehead Kits => S.E.X. Kit => Topic started by: SilverS on February 18, 2011, 12:35:46 PM
-
I'm very happy to announce my completion of the S.E.X. amp! It took about a week of work and I'm (at the time of this posting) currently waiting for the caps to discharge so I can flip it over and bring it up to my room for a listen. But before I go about celebrating, I just wanted to confirm a few things:
First, most of my voltage readings showed the "A" side being +10% and the "B" side being -10%, though this probably has something to do with where I put the negative lead, or is a result of the following.
Second, I was given a couple parts that vary from the guide, including a 1.2uF capacitor where I was supposed to install a 1uF capacitor.
Third, it seems one of my tubes is shorter than the other, is that a common occurrence or something I should be worried about?
Also, many thanks to Eileen for shipping a replacement power switch so quickly! It actually showed up right as I was finishing the final solder joint.
-
There was a shortage of 1.0uF caps for a while. I don't think anyone will notice any difference, the circuit is fairly insensitive to that value - as long as the two sides are the same.
I expect that if you swap tubes from side to side, the voltage deviations will follow. That's more likely to be tube to tube variation than brand or shape related - we've found very little difference between different 6DN7 brands. Some people hear enough difference to prefer the coin base variety, but even then the difference is pretty subtle.
-
Yeah I'm just worried that when I turn it on, one channel will be louder than the other.
-
Alright, the only two issues seem to be the 120Hz hum in the left channel and buzzing bee sound on both channels. Work for tomorrow.
-
To start looking for hum and buzz, disconnect everything except the speakers or headphones and turn the volume all the way down. That isolates the electronics from possible external noise sources.
-
I ran the voltage tests again to double check myself and got some interesting readings back on some of the joints:
A7 and B7 both read 10V~
C4 read 4V~
Terminal 2 read .2V~
Terminal 9 read 145V~
Terminal 10 read .5V~
-
Alright, minor update on my situation:
There is no longer a buzzing sound
The 120MHz hum is still present, and I don't quite understand how I can go about fixing it
The right channel is no longer transmitting, but I have a feeling I know how to fix that.
-
A7, B7, C1 and C4 should all be connected to chassis ground by the wire from C1 to safety ground, and you should read nearly zero voltage, AC or DC, at these points. Check wiring and soldering.
Your readings at 2, 9, and 10 appear to be AC; the manual gives appropriate DC measurements but not AC. But T9 (connected also to T12) should be at AC ground through the 1000uF capacitorT12 to T15. Check those connections and solder joints, also the orientation of the capacitor. T10 is on the ground buss, ultimately connected to T18 which is screwed to the chassis plate so there should be no voltage, AC or DC at these points. (Are you sure the measurements are volts and not millivolts?)
By the way, the hum is at 120Hz, not 120 megaHertz! :^) Errors in the multiplier are easy to make, we all do it from time to time, but they can make diagnostics difficult...
-
MHz was a typo sorry ^^
I'll get work on those solder joints though.
-
Alright, the hum is now gone and the amp is sort of transmitting on both channels, but there seems to be an even worse problem to replace it. Its unbelievably quite even at max volume.
-
Well, we're making progress ... sometimes these things get complicated but we almost always get it sorted.
So, what inputs have you tried, and what are the outputs (speakers? headphones?)?
-
My input's I've tried are my iPhone and my computer's soundcard, both being run through my quickie before hitting the amp.
I'm using a pair of Grado RS1i's for listening
-
OK, you've passed the resistance and voltage tests so most of the circuit is correct. Resistance doesn't check capacitors, so take a look at the solder joints at both ends of the two 0.1uF coupling caps, and the two 1.2uF parafeed caps - those are the only ones that are in the signal path.
If it's not there then it is likely to be at the input (level control wiring) or at the output (phone jack and speaker terminal wiring. Look them over carefully, poking with a chopstick perhaps, to see if anything is miswired or a bit loose.
If that doesn't turn anything up, we may have to trace the circuit. That means a source of fairly steady sound. I'd look for a computer sound generator to make a sine wave signal which we can trace, but a highly compressed pop song will also work. For starters, measure the output from the computer, with meter on AC volts - you should have something like 0.10v to 1.0v. Then we'll go directly to the SEX amp, to keep the chain simple. Be sure to start with the volume control turned down - once we fix it, the Grados will blast you!
Do the checks above, and if necessary I'll post some places to check and voltages to expect.
-
Back to square one it seems, the right channel is not working and the left channel is giving me the 120Hz hum (but at least it's transmitting sound).
:(
-
If the behavior changes intermittently without any changes to the circuit, than it is highly likely you have a bad or missing solder joint. A bright light, magnifying glass, and plenty of time are your friends in that case!
-
Sometimes a friend is the right friend for this case. You may be looking at the problem and seeing something else. After you do an inspection, if you find nothing let a friend do it too.
-
Could not agree more, Grainger. Even a non-techinical friend brings fresh eyes to the table.
I can't count how many times I've over-looked the same obvious problem, then had someone else spot it for me. After you've worked on something closely for a while, I suspect everything starts to look just as it should, even if it isn't correct.
-
Thanks for the suggestions, I'll have my dad give it a look over and see if he can find anything. I'm assuming he should just trace the circuit as best as possible?
-
I meant that your dad should look for odd looking solder joints and connections that have no solder. If he can trace circuits that is good too.
-
If he can trace the circuit, that would be great. Alternatively, you may just want to read each of the wiring steps to him in order, and have him follow along in the actual amp. That will end up giving you a terminal-by-terminal look at each connection, I think. Take the steps in order, and don't move on to the next step until you are both satisfied that the one you are working on is correctly executed.
A little time consuming I realize, but you should find anything that looks suspicious to you.
-
Ok first thing we did was hook some speakers up before doing anything and guess what?
The right channel on the speakers is working, but not the left. And the left channel is still the only one that comes though when I plug in headphones.
Where should we go from here?
-
Well, I would certainly suggest inspecting all the wiring to the headphone jack, the headphone jack, and the wiring that goes from the headphone jack to the speaker terminals. Since a bunch of that is done in twisted pair in which both wires are black, I can see where it would be easy to make a mis-step. When in doubt, of course, use your Ohm meter on each end to make certain that you are working on the same wire.
-
I could also just be that headphone jack is shot, I probably resoldered the bottom terminal 4-5 times (miswiring/resoldering etc). I'll definantly check the wiring tomorrow though.
Thank you very much to everyone for all the help, I can't tell you how relieved I am to have you all around for advice :)
-
Ok, I check the wires connected to the headphone jack and turns out one wasn't soldered, but that didn't fix anything, I think its just that the headphone jack is completely shot from all the soldering/desoldering/resoldering I've done to those connections.
-
Alright big update! Both channels are working
The only problem is the 120Hz hum is still in the left channel. Where should I check to fix this issue?
-
Did you replace the headphone jack to get both channels working?
Here are things I've posted before, not sure if you've done all of them:
>> To start looking for hum and buzz, disconnect everything except the speakers or headphones and turn the volume all the way down. That isolates the electronics from possible external noise sources.
Try to measure the output hum at the speaker terminals, using the lowest AC voltage scale you have. (It may be too low to measure with your meter, but if we get any data it will help figure this out.)
>> A7, B7, C1 and C4 should all be connected to chassis ground by the wire from C1 to safety ground, and you should read nearly zero voltage, AC or DC, at these points. Check wiring and soldering.
>> Your readings at 2, 9, and 10 appear to be AC; the manual gives appropriate DC measurements but not AC. But T9 (connected also to T12) should be at AC ground through the 1000uF capacitorT12 to T15. Check those connections and solder joints, also the orientation of the capacitor. T10 is on the ground buss, ultimately connected to T18 which is screwed to the chassis plate so there should be no voltage, AC or DC at these points. (Are you sure the measurements are volts and not millivolts?)
-
I just went down to give the amp one more test before going to sleep and its back to only working on the left channel again... This afternoon it transmitted on both channels when I plugged in my earbuds. So could it just be the low impedance of the Grado's that is messing up the transmission? Also, I'll do what you said to test for miss solders tomorrow afternoon and see if I can't get rid of the hum.
-
I'm wondering if it might be something about the fit of the headphone plugs in the jacks. Maybe check to see if the earbuds/both channels and Grados/one channel situation is consistent. I had a brand new set of cans at the Head Fi meet a couple weekends ago. Halfway through the day they went mono. Cans were fine, amp was fine. Turned out someone had overstressed the 1/8" to 1/4" headphone plug adapter and shorted it to mono.
-
I can't help but wonder if the Grados work on other headphone outputs? The lack of Left channel might be the Grados. Then again it might not be the Grados. I just have to ask.
-
I believe I have found the problem causing the hum:
T21 ~ 143V (supposed to be 183V)
T22 ~ 143V (supposed to be 183V)
T29 ~ -14V (supposed to be -17V)
T12 ~ -14V (supposed to be -17V)
So, I should just check all my solder connections to the terminals on the B side or could it just be that power isn't properly getting to that side? Sorry for asking yet again but I don't know where to start looking.
As for the left channel only transmission, it's probably just that the Grado's plug and the headphone jack aren't playing nice, I'll have a new jack by the end of the week though so that should fix any issues there.
-
Oh and a minor update the Hum is gone, but still no right channel to speak of.
-
I want to piggyback on this thread as I am having the same issue... sort of. I just finished my SEX kit as well and when playing my ipod w/ Grado RS-2 headphones plugged in, the amp seems to work fine out of both channels. However, when I plug in my speakers, the right channel is not working.
I troubleshot to ensure that the speakers and cabling were not the issue, and they were not
So, I retraced all of the steps, paying very careful attention to the wiring from the headphone jack to the binding posts, which seems to be correct. I triple checked my solder joints which all seem to be good (I even moved one wire from the twisted teflon pair of wires to the outside terminal of the bottom pair of terminals on the headphone jack, as the inside terminal was crowded and made me suspicious that that terminal may have been a bad solder joint with how crowded it was). However, the problem remains the same.
I'm at a loss as to what else I can try, so I'm going to list a couple of inconsistencies that I found with the resistance/ voltage checks to see if someone can make sense of what might be going on...
RESISTANCE
19, 39 = 5.5K ohms at minimum volume (should be at 0 ohms w/ volume at minimum)
A4, B4 = 6K ohms at minimum volume (should be at 0 ohms w/ volume at minimum)
VOLTAGE
PT terminal 1= 0 VAC (should be at 120 VAC)
PT terminal 7= 188 VAC (should be at 0 VAC)
PT terminal 10= 188 VAC (should be at 0 VAC)
*Terminals 6 and 9 on PT are at 188 VAC as well.
-
SilverS, those voltages are low, but not a lot low. Makes me wonder if your power line voltage is low, or if your meter is reading accurately. Let us know when you've replaced the headphone jack, I still suspect some internal damage for the intermittently missing channel. The fact that it once worked indicates the electronic portion is wired correctly.
cmason84, the voltages on the power transformer 6=7=9=10 do vary with different meters, so if the voltages at the tube sockets are correct you can ignore those power transformer measurements.
The zero volts at power transformer terminal 1 indicates you have the neutral connected to that terminal, not the hot. Check your wiring carefully, then if necessary check your house wiring - it may be that your power is wired incorrectly. (I am assuming you have true 3-wire sockets with a good safety ground.)
You say you get both channels on the headphones - first thing is to be sure it's really both channels, and not one channel in both ear pieces. The latter would indicate a bad connection between the headphone plug and the jack, which might be a mechanical problem rather than an electrical one. If you do have both channels in the phones, then the problem is in the wiring from transformer to jack and speaker connectors.
-
Thanks Paul, all is well now.
The power was wired correctly (voltage check sheet lists incorrect voltages for terminal 1 and terminal 2 on PT; they're switched).
The problem was the wiring from the jack to the binding posts. When the directions said to solder the wires to the red and black binding posts, I was looking at the red and black on the underside of the chassis instead of the topside of the chassis. All I had to do was swap the wires at the two inside binding posts and I'm getting sound from both of my speakers now.
-
I have already replaced the headphone jack, which fixed the hum. As for it transmitting on both channels I checked again, and it turns out it was just transmitting a the left single to both sides. So even at the time when I posted to worked on both, it really was just a mono signal being transmitted on both sides.
-
Good, that's progress!
OK, we'll do some signal tracing, starting at the input and output. First, find a song that is fairly consistently loud. Unplug the source from the SEX amps and measure the AC voltage at the RCA plugs coming from the source. You should see a fraction of a volt on each channel - I'm just making sure your meter can measure a signal that small, and that both channels are working on the source.
Now it's the right channel that's dead. That would be hhe "A" tube which is on the left when the amp is upside down, correct?
Then plug that source into the right channel (the dead one) and measure the voltage at T19, with the amp still turned off. You should see the same voltage if the volume control is all the way up. Then check at the tube socket pin 4 - that's the grid, input to the tube, and should be the same as T19.
If the signal has gotten that far, then we'll check the other end. Measure the AC voltage at T20, where the output transformer brown wire is connected. Do this with the amp turned on. You should see a much greater voltage. If it's there, check at T5 which is the transformer output - should be less again, maybe a volt or two.
-
I'm really confused right now. I just tried testing those terminals and when I had the negative lead connected to the ground wire from the power socket, and the red one in hand touching nothing, it was reading 1-1.5V~. Then when I connect the positive lead to terminal 19 it drops to .0003V~
Also I had a question about terminal 19. I recall when soldering resistors to those terminals, the guide called for the B side to have the resistor soldered to the lower hole of terminal 19, while the A side should have it soldered to the top hole of terminal 19. Is this a mistake in the guide? And perhaps why I'm not getting a signal from the right channel?
Also, considering speakers are picking up the right channel wouldn't that mean the terminals around the tubes and the tubes are all working properly?
-
When the meter lead is not touching anything, it acts like an antenna and picks up whatever electrical fields are around. Once it is connected to something, the low impedance of that something (less than the 10,000,000 ohms of the meter input anyhow!) shorts out the antenna function, leaving the measured voltage.
The top and bottom holes of the terminals are all the same metal, so they are all connected together. I'd call it an "inconsistency" in the manual ... sounds better that way. :^)
You said previously "As for it transmitting on both channels I checked again, and it turns out it was just transmitting a the left single to both sides." Now you say the right channel is working with speakers. So I don't know what the real problem is anymore. If both channels work with speakers but not with headphones, then the headphone jack/transformer output/speaker terminal wiring is in error somewhere.
-
Yeah I can see where the confusion is coming from. Let me sum everything up to a final conclusion:
The speakers since day 1 of the issues have only been transmitting on the right channel.
The Headphones since day 1 have only been truly transmitting on the left channel (when I said they were transmitting both it was just the left channel being sent through both sides for some reason)
So where do we go from here?
-
Thanks, that makes better sense now - guess I just got lost in teh many posts.
OK, the amp is generating output on both channels. The output (autoformer secondary connections are T5 and T25 for the hot lead (the black primary and white secondary are both connected here. The cold side of the outputs are the secondary black wires, connected at T10 and T30.
The hot outputs (T5 and T25) go to the headphone jack, and from there to the hot (red) speaker binding post on the top side for that channel. You may have to trace the physical wires, because the output transformer secondary winding has a fraction of an ohm resistance, making it hard to tell with a meter which speaker terminal is connected directly, and which is connected through the winding. Also, the speaker terminals in the manual are coded red for one channel, black for the other channel - but on top there's a red and a black for each channel - it's the binding post color, not the base color referred to. I'm not sure if the current speaker terminals are identical to what's in the manual now, so I can't be certain this is clear, but this seems to be the most likely place for a wiring error.
Similarly the cold outputs go to the headphone jack (lowest terminals, which are connected) and from there to the black speaker binding posts.
It might help to first unsolder the wire that goes from T5 to the headphone jack, and also the one at T25. In both cases, these are one wire of a twisted pair. Then those wires can be traced with an ohmmeter, as can the cold wires (the other wire of the twisted pairs).
-
Alright I desoldered the wires connected to T5 and T25, no sound on either channel now. I'd also like to add I never plan on hooking speakers up to the amp, so trouble shooting the binding posts is at the bottom of list for me.
-
So, now we trace the wiring to the headphone jack and binding posts. (I am including the speaker binding posts because an error in that wiring could cause problems with the headphone wiring, and I don't want to complicate matters by deviating from the manual.) At every terminal you test, inspect the solder joint while you are there in case the solder hasn't flowed properly.
First check the ground wiring. With one meter lead to terminal 17 or 18, check that there is zero resistance to terminals 17, 18, 15, and 10 (and to A3 on the tube socket). Then check resistance to the lowest terminals on the headphone jack. Finally, you should find zero resistance from this ground to the two black speaker binding posts (one for each speaker).
All OK? Then measure the resistance to ground of the wire that used to go from terminal 5 to the headphone jack - it should be infinite. Do the same for the red speaker binding posts which should also be infinite resistance to ground.
Repeat for the other side, terminals 37, 38, 35, 30, and B3 for ground, and infinite resistance at the wire that used to go from terminal 25 to the headphone jack, and the red binding posts.
If all that is OK, then we will trace the hot leads. With one meter lead connected to the wire that used to go from terminal 5 to the headphone jack, put the other lead on the middle terminal of the headphone jack on the side nearest the volume control - the should be connected with zero resistance. At the other end of the 120 ohms resistor, you should measure 120 ohms. The resistance to the other resistor on the headphone jack (the top terminals) should be infinite. Finally, at one of the red speaker binding posts you should read zero resistance, and infinite resistance at the other red terminal.
Repeat for the other side - meter to the wire that used to go from terminal 25 to the headphone jack, read resistance at both ends of the resistor on the top of the headphone jack (should be zero and 120 ohms) and the other red speaker binding post.
-
Here are the results of the tests:
All resistances in Ohms
T17 ~ 1.1
T18 ~ 1.4
T15 ~ 1.1
T10 ~ 1.6 - 1.8
A3 ~ 1.6 - 1.8
Black Binding posts (L/R) ~ inf./inf.
T37 ~ 1.1
T38 ~ 1.4
T35 ~ 1.1
T30 ~ 1.6 - 1.8
B3 ~ 1.6-1.8
Red Biding posts (L/R) ~ inf./inf.
As for the wires going from T5 and T25 to the headphone jack, was I supposed to desolder them from the headphone jack or the terminal strips?
-
Man, it's really hard to describe things fully with just words! Sorry about that, I means to desolder from the terminal 5 and terminal 25.
The ground buss resistances are all low, showing they are connected properly. However, those resistances should not differ by even 0.1 ohm, since they are all connected by short lengths of wire. It may be the solder connections are not good, or it may be that the connection to the meter was not good enough to get that kind of accuracy. It can be quite difficult to get a good, low resistance connection with a meter probe or even a cliplead.
You did not report the resistance from ground to the bottom terminals on the headphone jack, but they should also be near zero. There are two twisted pairs of wire, from the headphone jack, one to each speaker output pair. One of those wires is the ground and should go from the bottom terminal of the headphone jack to the black binding post. You did report that none of the speaker binding posts are connected to ground, but the black ones should be. So there is a failure of ground connections somewhere along that line.
-
Whats really confusing me is that the speaker posts get their signal from the headphone jack, which means they're getting the right channel from the headphone jack but the headphone jack doesn't appear to be getting the signal in the first place.