Bottlehead Forum
Bottlehead Kits => Crack => Topic started by: rburrows on June 25, 2011, 09:16:50 AM
-
Hello Bottleheads,
I recently purchased a built-up Crack & Speedball unit from a user on head-fi.org. The Seller touched up the solder joints and verified functionality before sending off. The amp arrived very well-packed, but for whatever reason I am unable to hear sound when the unit is powered on. I am hoping a simple mechanical disturbance occurred in transit but I could use a little help in isolating the issue.
I confirmed proper functioning of source, interconnects, and headphones so I'm pretty sure it's a problem with the unit. Here are the things I have observed:
- After powering on, tube filaments light up in a seemingly normal fashion
- No LEDs in the circuit are lit when powered on, even after some time.
- During a couple power-up trials, I did notice a very faint, crackly signal in the left channel only if the volume control was maxed or near-maxed. It would fade to nothing in a matter of seconds. I cannot replicate this consistently.
- I went through the resistance checks from the manual and found that terminals 1,5,7,9,b3,b6 are all reading infinite resistance. Other terminals match the manual.
I am not very experienced with electronics, and nothing obviously visually wrong with the circuit stands out to me. The seller suggested the following:
"Reading through the manual, points 7 & 9 are what connect to b3 and b6, as well as connect to the larger speedball board. So that's something...
The part that gets me though is that power is out on both sides, which is odd because most of the circuit is two independant halves. Even the large speedball board is mostly mirrored across the middle. I find it very unlikely that there would be the same break/damage on both sides at the same spot. A shared path would be the (I think) ground. There's a wire from point 3U to G (page 25 of the speedball manual) that you could double check. I'm not entirely certain if that's a ground path, but it's the only thing I can think of at the moment."
That connection looks good but I am not sure how to confirm on the multi-meter.
Any advice would be greatly appreciated!
Thanks,
Robert B.
-
The very first thing to do is look over all of the solder connections with a magnifier and see if you can spot what came loose. Go ahead and do the voltage measurements too, and report the deviations from the manual. That will speed up the process of figuring out what is wrong. Since no LEDs are lighting it sounds like the problem is between the power transformer and the amp circuit - the rectifiers, the filter caps, and the associated resistors and connecting wires.
-
Hmm, voltages are strange and weak:
1: .8
2: 0
3: 0
4: 0
5: .8
6: 0
7: .1
8: 0
9: .1
10: 0
11: 0
12: 0
13: 0
14: 0
15: .1
19: 0
20: 0
a1: .8
a2: 0
a3: 0
a4: 0
a5: 0
a6: 1.3
a7: 0
a8: 0
a9: 0
b1: .8
b2: .2
b3: .4
b4: .8
b5: .2
b6: .8
b7: 0
b8: 0
-
If the heaters are glowing, you can measure nothing but essentially zero DC volts everywhere, and the meter is being used properly you have a problem at the power transformer high voltage secondaries. Switch your meter to AC volts, at least 400V scale, and measure the AC voltage across power transformer terminals 6 and 7 and then terminals 9 and 10. You should see around 150VAC.
-
I am measuring 170 VAC between both 6&7 and 9&10. Using V~ 700 on my DMM.
For the VDC measurements I was using the V⎓ 200 setting on my DMM. at the next level up , V⎓ 1000, those decimal values became zero.
-
OK, power transformer is good. Measure the DC voltage at terminal 21, red test lead to terminal 21 and black test lead to terminal 12. Use the 1000VDC range.
-
Measuring 0 VDC between terminals 12 and 21.
-
OK, turn power off, unplug amp. If you have a diode check setting on your ohmmeter use that, if not, use around the 2K ohms or 20K ohms range. Measure across each UF4007 rectifier connected the strip on the right side of the power trans. They should read either around 500 to 1000 ohms or infinity depending upon which way the test leads are oriented. If you see a really low reading, like under 100 ohms, the rectifiers may be blown.
-
I'm not sure if these readings make sense, I took them in both 20k and 200k ranges. I was getting all infinite on 2K range.
20K 200K
18->20 : 19.66 : 106.4
18->21 : 18.15 : 92.2
19->20 : 19.55 : 106.5
19->21 : OL : 109.5
-
Silly me, I found the diode check setting on my DMM. Each rectifier is reading around .570 on the scale.
-
OK, assuming that your meter reads like most that means 570 ohms on the 2K ohm scale which is about right. If the rectifiers are intact and everything is connected properly you should have voltage at 21.
So these are the possibilities:
-The rectifier readings are actually .57 ohms and they are shorted out. It's not common that all would be dead, usually one or two blow if there is a short downstream from them in the circuit.
-Something in the power supply wiring may be shorted to ground. Look for a bare lead touching the chassis or a grounded terminal.
-Something is wrong with the connection at 21 or 12. It is possible that the black wire that goes from 12 to the headphone jack and then on around to the other ground connections is open somewhere.
-Just to cover all possibilities, it could be that your meter is messed up on the DC volts reading. This however does not explain why the LEDs aren't lighting up, so that is a pretty slim possibility.
-I think you are on the right track thinking that something got moved the wrong way in transit. At this point my best guess is that close examination of all of the connections at every terminal to make sure everything is still connected could find the problem. I would also ask the seller to pick his brains about what might be wrong, as he may recall a problem area.
Hang in there, lately we've been getting the majority of folks with problems up and running in a day or two.
-
Can I just say that Bottlehead (Doc, Paul, et al) are the greatest customer service folks in the business? My hat is off to you guys....
John
-
A-ha. So I found that the wire between 18L and 10 was loose at the 10 terminal. I re-seated the wire and added a fresh bit of a solder and now the Crack is making sweet music!
Wow John, no kidding.
Thank you Doc for getting me on the right track so quickly and so clearly. Troubleshooting with you was a breeze
-
A-ha. So I found that the wire between 18L and 10 was loose at the 10 terminal. I re-seated the wire and added a fresh bit of a solder and now the Crack is making sweet music!
Wow John, no kidding.
Thank you Doc for getting me on the right track so quickly and so clearly. Troubleshooting with you was a breeze
-
Spent the afternoon demoing some albums on the Crack, and it's sounding really nice. But I have a bonus question :)
I'm noticing a slight hiss in the left channel only. It remains at a constant low volume independent of volume knob level. Doesn't seem to matter if interconnects are in or not. The right channel is noise free. (just FYI there is a separate, lower pitched hum that is dependent on the volume knob, but is only slightly noticeable when maxed. This is even across the two channels. It's not an issue at all at normal levels but I'm just pointing it out for contrast)
Checked on a couple sets of headphones and the left channel hiss is there for each. Could this be another case of disturbed soldering joints, introducing noise? If that seems plausible I'll go through the connections again.
Thanks!
-
The hum could be a tube or an iffy ground connection. It could also be caused by other things but Im better served letting others help you with that. The "hiss" could also be a tube (tube rush) but again, could be other causes. Couldnt hurt to check the solder joints again, probably focusing on the grounds. Always a good idea to have some spare tubes as well. They dont have to be the highly sought after tubes, just spares. Always good to have spares to troubleshoot a possible problem tube and to get you back up and running in case a tube is problematic or goes south. Fortunately, many of the 6080/6AS7 and 12AU7 types can be had for pretty cheap. Others will chime in with more possibilities and better, more specific advice.
-
The tubes being at fault had crossed my mind. Maybe I'll order some spares (like you say, they're good to have around in any case) and check against those before meddling with the circuit more. Thanks! -Robert
-
I have some tubes on order and will report back as to their effect on the left channel hiss.
Another thing I observed in the running circuit is that though the 8 speedball LEDs are lit, the 2 LEDs wired to the input tube are not. The voltage readings at their terminals seem exceptionally low (see complete listing below). Is this normal or could they be blown? If something is wrong, what are the risks/detriments to running the amp like this? The amp sounds great right now, but perhaps I'm slowly destroying it or missing out on what it's actually capable of?
1: 46
2: 172
3: 0
4: 173
5: 50
6: 0
7: 83
8: 0
9: 83
10: 0
11: 0
12: 0
13: 173
14: 0
15: 191
20: 0
21: 210
a1: 49
a2: 0
a3: .56
a4: 0
a5: 0
a6: 46
a7: 0
a8: .24
a9: 0
b1: 45
b2: 173
b3: 84
b4: 49
b5: 173
b6: 83
b7: 0
b8: 0
-
I dont have the speedball upgrade yet so I dont know what the voltages are supposed to be. But, I think the original 9 pin LED's are supposed to still light up. You need better advice than I can provide. But I know that Doc often mentions double checking to make sure that the right resistors and transistors are installed on the right boards. Double check that if you havent and then, I guess, hang on for better help. Im sure it will be forthcoming. It's been slow on the board, understandably with the holiday. Im sure more Bottleheads will be checking in soon.
-
Another thing I observed in the running circuit is that though the 8 speedball LEDs are lit, the 2 LEDs wired to the input tube are not. The voltage readings at their terminals seem exceptionally low (see complete listing below). Is this normal or could they be blown?
FWIW - I damaged one of the LED's on my input tube while installing my speedball, and it actually resulted in a voltage that was way too high at A8. Here is my trouble shooting thread that discusses, among other things, how you can measure resistance across an LED, causing it to light up, and the measurements you should see (quickly jumping to infinite resistance), if your meter doesn't have an LED setting. (See http://www.bottlehead.com/smf/index.php/topic,1825.0.html ) I found that if I set my meter to 200 or 2000 ohm sensitivity, my undamaged LED's would light up (while the damaged one did not). If I set it any higher, none of them would light up.
You might try this with your amp powered off, in a dark room just to make sure your LED's are actually bad before you contemplate replacing them. It seems to me that if your LED's were bad at A3 and A8, you would be seeing unusually high voltages, like I did. Then again, I'm not an engineer and this is really just lay speculation. Best of luck!
-
Hello again,
After some tube rolling I have found that the left channel hiss is always there, but its volume is subtly changed by different tube combos. I haven't investigated any further on this front.
As for the LEDs, I performed some diode checks on all of them to get a sense of their relative behavior.
On the big speedball board: the LEDs on the edge of the board do not light, but read .955. the inner ones light but give no readings.
On the two smaller boards: the LEDs closer to the black MJE350 unit read 1.2 with no light, and other ones light but give no readings.
On the input tube: A3 LED reads .245. If I reverse the probes (ground on the non-metallic side on the LED) I get .228. For A8, I get the audible tone from the multimeter, and nothing when reversed. Given the cramped layout here, I'm measuring at the terminals and not directly on the LED leads. Not sure if that matters. No lights in any case.
All speedball LEDs light up when the amp is on. It is just the input tube LEDs that do not.
To get more reference points, I also tried Hopeful's method of using the low ohmeter settings, but I couldn't get any LEDs to light up or report values this way.
-
Not sure if this was mentioned before, but check the value of the R1 resistors on the small PC boards.
-
reading 235 ohm for each R1 resistor on the small boards, so those seem OK.
Does it make sense that if the bias LEDs were dead and/or improperly soldered, the voltages could be mostly right and the amp could pass a good signal (but perhaps not its best) to the headphones? I recently re-measured voltage with different tubes, and they are closer to the manual now than my previous listing, though still a bit low:
1: 55
2: 171
3: 0
4: 171
5: 64
6: 0
7: 90
8: 0
9: 95
10: 0
11: 0
12: 0
13: 173
14: 0
15: 191
20: 0
21: 210
a1: 63
a2: 0
a3: .572
a4: 0
a5: 0
a6: 54
a7: 0
a8: .243
a9: 0
b1: 54
b2: 173
b3: 90
b4: 63
b5: 173
b6: 95
b7: 0
b8: 0
Thanks,
Robert
-
The LEDs usually either work right or don't work at all, but I suppose they could be partially functional. I would double check that the small metal can transistors on the small PC boards are 2n2907s and the ones on the big board are 2n222as. If nothing else is obviously wrong I would maybe go ahead and change the LEDs on the A socket to see if it makes a difference. I might also ask the seller to verify that he saw the proper voltages at pins A6 and A1 and A8 and A3 when he built the unit with the stock tubes. It would help to know if the deviant measurements have always been there or if they developed after shipping.
-
So I replaced the bias LEDs on the input tube. It was a little tricky with all the other circuitry in place, but with patience I got everything set. Now they light and the voltage readings are in line with the manual.
The sound quality has bumped up a couple notches. It was good even with the broken LEDs but I sensed it wasn't running at peak performance. The left channel hiss is also gone
-
Excellent! Glad to read that you've got it sorted out. Enjoy.
-
Very happy to hear you got her up and running!