Bottlehead Forum
Bottlehead Kits => Legacy Kit Products => Crack-a-two-a => Topic started by: Toobuzz on May 22, 2021, 07:26:24 PM
-
Hello,
I built the Crackatwoa and listened for a day. It was outstanding and great fun to build, of course. I also changed out the caps and listened some more. Today I installed the Twoquiet along with some bigger caps. Once I completed the Twoquiet upgrade I ran the resistance checks and all passed (still do). I then registered the following voltages:
Low current C4S
IA 107
OA 73
KregA 1.1
bRegA 107
IB 148
OB 70
KregB 3.25
bRegB 148
High Current C4S (B tube socket)
IA 195
OA 107
bA 0
IB 0
OB 78
bB 107
The other higher current board measures perfectly. After going through the manual twice, I focused on the bad high current board. I removed the jumpers and all excess solder/flux. I took my time and reflowed all pads and ran new jumpers. I also inspected, checked for continuity/resistance all connections on the B tube socket as well as solder lugs 6-10.
Here are the new voltages:
Low current
IA 49
OA 33
KregA .36
bRegA 49
IB 148
OB 70
KregB 5.5
bRegB 148
High current
IA 198
OA 49
bA 0
IB 0
OB 30
bB 49
The B side LEDs of this HC board also went out after reflowing and reattaching the board. All 4 LEDs on the low current board have not illuminated since attempting this upgrade.
Was I correct to focus on this HC board? Is there something that these numbers suggest I focus on next?
I may do the same rework on the low current board but I’m just guessing at this point. She was working great before doing the Twoquiet and that upgrade didn’t really touch any of the areas that seem to be giving me issues. Any help would be greatly appreciated!
Sam
-
I reflowed the A side of the low current board and ensured all connections were clean around terminals 1-5 and 11-15. I even reattached the white wire on pin 2 of the 7-pin socket under the offending board. I swapped the 7-pin tubes and am still getting these second set of voltages from my OP. Is there some way to rule out a bad transistor? I suppose I could swap the high current boards, and if the issue doesn’t follow the board then I must be missing a mis-wire or short.
-
Can you tell me about the soldering products you're using? I would very strongly recommend not adding flux to your build, as there is already flux in your solder. If you pull both 6AQ5 tubes and run the amp, does the offending voltage pop back up? (It should go up closer to 200V)
-
I use a Hakko FX-888D and Cardas Quad Eutectic. I’ve never used flux on any of my builds, although I sometimes wish this solder was a smaller gauge. After pulling the 6AQ5 tubes I only see a small difference in voltages.
LC
IA 60
OA 40
KregA 0
bRegA 60
IB 205
OB 70
KregB 1.25
bRegB 205
Would swapping the high current boards help isolate the issue?
-
Which flux does your Cardas solder have?
Are you 100% certain this build was working prior to the modifications?
Next I would pull the wire out of the IA pad on the low current board that goes to 6U, then see if the voltage at OA on the high current board pops back up.
-
I am using Cardas Audio Quad Eutectic With Rosin Core. I have a .25 lb spool. I pulled the red wire from IA on the LC board (other end still attached to 6U) and the OA on the HC board is only up to 64V.
I am 100% sure that is was working, both in stock form and with the Mundorf SO's on the LC board. The EVO 100uf are new as well as the TwoQuiet (as they kinda needed to go in at the same time), but those mods don't seem related at all to this issue.
Thanks for helping me. While a bit stressful, these challenges actually maximize the enjoyment later on. I was happy to have built the BeePre2 w/o any help, and then I was proud to have troubleshot an issue where I lost the right channel about 10 days in. I'm still pretty new to the hobby but I feel like I can read a schematic and troubleshoot these things on my own, but this one is stumping me.
-
My concern regarding the solder comes from your photos. There's light colored flux gooped all over everything. That's not the color I would expect from rosin core flux and seeing it just all over the place is inconsistent with the flux flowing out form inside the solder.
On the high current C4S board, disconnect the wire leaving OA and going down to the 5 lug terminal strip below it, then see if OA on the high current C4S board pops up into range.
-
Well, I desoldered the wire leaving OA on the HC board by gently pulling down with tweezers, and wouldn't you know, the short wire fell to the chassis as it broke off from the 5-lug terminal strip. This seems likely to be the root cause, if this wire has become compromised! It passed my basic continuity checks but it was obviously extremely week. I'll rework that connection and post my results.
Thank you for leading me directly down the (or a)correct path!
BTW, that flux certainly looks disastrous on the A tube base, but that was the only photo I used a flashlight to assist with lighting. That clear/blue haze is a weird reflection from the flashlight and really just a shiny bulb of solder. The yellow/gold splatter is all me using too much solder I suppose.
-
I repaired the wire, connected all boards, and put all tubes in. I am still getting the same voltages.
I did get a 200V reading on OA of the high current board when there were no 6AQ5s plugged in, IA on the low current board was detached and OA on the high current board was detached.
I'm still getting those same voltages where IA on the low current and OA on the high current are only showing 50V. The other changes I can report are that the LEDs on the low current board are all lighting (dimly), but one of the 2 LEDs in the 9pin socket is not lit. This is new as both were lit this morning.
-
After dinner I’ll swap the high current boards. If the problem follows the board then that obviously narrows down the location the issue.
While she was running yesterday I noticed that the chassis plate was much warmer than the Mainline or BeePre2. Not alarmingly so, but I though it worth mentioning. I figured it was due to tight fitting ceramic tube sockets.
-
I did get a 200V reading on OA of the high current board when there were no 6AQ5s plugged in, IA on the low current board was detached and OA on the high current board was detached.
Now reconnect OA on the HC board to the terminal strip but leave IA on the LC board detached. I would not swap any boards at this time. We still don't have things narrowed down enough for that to be helpful.
-
Ok, we’re getting somewhere. After reconnecting OA to the high current board (from terminal strip), without 6AQ5s, and without OA attached on the low current board I am getting 200V on both IA and OA of the high current board.
Adding the tube makes OA drop to 50V.
-
That would tend to point to a shorted TL431 on the center low current C4S board.
-
Ok, excellent! What can I do with this? This doesn't look shorted to me, or am I missing something?
-
I would recommend removing it and replacing it. You can't just look at it and say with certainty that it's shorted.
-
Thanks for your help PB! I’ll work on getting a replacement.
-
Ok, I replaced the regulator on the A side of the low current board and unfortunately I’m still getting 60V on IA.
Just to double check, is this a compatible part?
https://www.mouser.com/ProductDetail/on-semiconductor/lm431aiz/?qs=fAPocZeZcjgLlO1ZPti8PA%3D%3D&countrycode=US¤cycode=USD
-
Yes, that's acceptable. With all the tubes in and the amp running, can you give me IA/OA on the high current C4S board on the bad side, as well as Kreg and the OA/OB voltage on the middle board on the bad side?
This is somewhat likely to just end up as a repair on my bench, so keep that option in mind also, though you'll need to replace the caps on the low current C4S board with what we provided first.
-
High Current Board on bad side
IA 195V
OA 60V
Low current board
IA 60V
OA 40V
Kreg A 9.5V
IB 150V
OB 70V
Kreg B 5.5V
Please let me know if there is anything else that I can try!
Is there anything to the fact that the KregA voltage was always low (1V or even less) and after the regulator change it’s now high (9.5V)?
Sam
-
Your initial 1.1V on Kreg A is resolved, so that's good. Now pull the 6AQ5 and verify that OA on the high current C4S board pops up. Once you have verified that, put the 6AQ5 back in and pull the 6080, then perform the same test to see what OA on the high current C4S board settles to.
-
Without the 6AQ5 I’m still only getting 60V on OA of the HC board. Could there be an issue with the 7pin socket itself?
-
Without the 6AQ5 I’m still only getting 60V on OA of the HC board. Could there be an issue with the 7pin socket itself?
Pull the 6080 and the 12AU7 and the other 6AQ5, then recheck to see if the OA voltage there pops back up.
-
Yes. With no tubes at all the OA is back up to 200+V on the HC board. It’s the 6080 alone that makes it drop.
Does This make Q1 and Q2 on the B side of the HC suspicious?
-
Well hold on a sec, try with just 6AQ5s, then the OA voltage on the HC board should cruise down to about 150V, then try with the 6AQ5s and the 12AU7 and everything will be OK other than the OB voltages on the big C4S boards. You would also expect those Kreg voltages to be quite low, but I think the regulators will still work without the 6080 sucking down current.
-
You are correct. It’s not perfect. With only the 6AQ5s the OA on the HC board did come down to 161V.
Adding the 12AU7 (still no 6080) I get:
HC board
IA 195
OA 162V
bA 0V
OB 0.1V
bB 162V
LC board
IA 162V
OA 71.5V
KregA 2V
bRegA 162V
IB 188V
OB 69V
KregB 2V
bRegB 188V
At least the LM431A seems to be working well. Is that right, since the Kreg voltages match on both sides?
-
Yeah, the Kreg voltages are too low for the regulator to work effectively.
So to continue debugging, the two paths would be:
1. Undo the cap mods to see if things go back to normal. If there's something amiss with the output coupling caps that could present itself as an issue with the 6080 in the socket but not be an issue at all with no 6080.
2. Try a different 6080. Since the amp was working before the mods, my money would be on the caps being somehow problematic.
It's worth mentioning that Audyn and Dayton 100uF film caps will fit the space in the amp pretty well. Those Mundorfs don't look like they fit all that well and the extra 100V of voltage rating makes them bigger than would otherwise be necessary.
-
I tried a different tube and the stock 100uf caps. Unfortunately, no change. I’m confident in my wiring and soldering (famous last words), and I’d really like to tackle this myself. I’ll buy a whole new set of parts, whatever is necessary. If you think it’s my wiring then I’ll order some and redo a whatever sections are needed.
Based on my last set of measurements, and knowing that the caps were good, doesn’t that point to Q1 and Q2 on the B side of the HC board (the B Side of the HC board is the only place where the LEDs won’t light)?
-
If you had just built this amp and it had never worked, then I would be more suspicious of the current sink load for the 6080. At this point you can swap the high current boards to see what happens. If the problem switches channels, replacement of at minimum the TIP50 and 2N2222 transistors would be in order, but there is generally a reason for a failure like this to happen, and ascertaining that may present a bit of a challenge.
-
The issue followed the board to the other side. I see, you are doubtful that a bad component is the root cause. If I replace the bad components the new ones might go too. Is it worth a shot to try replacing them and see what happens after that? I did see and replace some sketchy wiring on terminal 6. Perhaps that contributed.
Thanks again for all your attention to this!
-
Yes, I would replace the transistors on that half of the board, but keep an eye on the amp for a while to see if more problems develop. Sketchy wiring at 6U could cause a fault that took out your 431 chip, but that shouldn't really cause problems with the 6080 and its current sink load.
-
Alright, I replaced the 2N2222A and the TIP50 on the problem high current board. All measurements are now spot on except the Kreg on the side with the bad HC board (Now the B side of the LC board).
On the LC board I get
KregA 5.5V
KregB 2.5V (side with suspect HC board)
By the way, this is pretty much what I originally had when it was working (except KregA was low). I may have had 2.8V on the Kreg, but I figured it was close enough to 3V. Im no longer going to be that cavalier. Now, since the issue has moved to the other side, following the board, that gives me confidence that the underlying issue is not wiring or sockets. It’s also not the LC board as it performs fine when connected to the other HC board.
Any suggestions for getting that Kreg up?
I did put in some equivalent parts (until stock parts arrive).
I’m using an 2N2483 (NTE says it’s a 2N2222A equivalent) and a TIP50 D1. These are on high current side and wouldn’t seem to affect the Kreg, but disclosing for sake of completeness.
-
2.8V is barely enough. Does this follow one of the 6AQ5s if you swap them side to side?
-
Yes, it does! Sort of. Simply swapping the 6AQ5s brought the KregA up to 3.75 and the KregB down to 4.35. Technically all voltages are in range now! I don’t have that warm and fuzzy felling yet, however. Is the next logical thing to try a fresh set of 6AQ5s?
Whoa! That TwoQuiet went from “real quiet†to pretty darn amazing sounding!
-
6AQ5s are one of those tubes that really need to be awoken to work all the way. When I test them before sending them out, it can take a while for them to come up to full emission. I would leave the amp on for three or four hours, then check the voltages. I would bet that you'll find that both of your Kreg voltages climb after some run in time.
-
Thanks again, PB! I really enjoyed working through this with you and I am super grateful. Even though I may have come across more as frustrated (I was). This forum is 1000x better than a tutorial or YouTube video in terms of learning about and hence, appreciating the amp. I consider this experience excellent value for money!
-
We are always here to help, and the C2A is certainly one of the more complicated circuits to debug. Hopefully your fixes are all that's required for years of reliable operation.
-
So all has been well for the last few days, all voltages well within range. However, I lost the right channel after a 12au7 tube swap this morning. She was fine for 20 minutes then I powered down to swap the 12au7. Im hoping you can help me zero in on what would cause OB on HC board to jump up to 145V? OA on the LC board is also reading 145V.
Is this the part of the circuit that involves the plate of the 12au7 and the MJE350 and PN2907?
These are the only 2 measurements (resistance and voltage) that are affected right now.
Thanks!
-
Can you put the other 12AU7 back in?
-
I did try that. Still seeing that 145V.
The Kregs are both rock solid at 4.4V now, btw.
-
What's OA and OB on the low current center C4S board?
-
OA = 145V
OB = 91V
-
OK, likely the only issue with your amp is that high OA voltage on the center PC board. That can happen from any of the following and this is the order I would use to debug them:
1. The solder joint at pins 4/5 on the 9 pin socket isn't solid and one half of the tube isn't heating up. This will also manifest as only half of the tube glowing when you look at it.
2. The solder joint on the center leg of the MJE350 feeding OA isn't 100% soldered and working properly. Just reheat it until you see the solder wick into the board to be sure.
3. Possibly an LED isn't working properly, but they don't break on their own. A voltage of 145V on OA and a DC voltage across an unlit LED of about 12V DC would indicate a failed LED. It's really important that when you see an LED not illuminating that you don't fixate on it and resolder it to death.
4. Some other wire is broken or loose.
-
I found the issue and I’ll spare you a pic. 12au7 rolling made the pins move and the lead attached to the center pin of the right channel snapped. I’ll put a little slack in my replacement.
-
You can also install a wire between pins 3 and 8 for the time being to get the amp back up and running.
-
Thanks for the tip! Do you think it would be wise to order some new resistors and a little more wire (in addition to the LED) and clean up that 12au7 socket. It’s definitely the sloppiest part of what I’ve done here. I was thinking the heater wire is fine but maybe redo everything else attaching to the 12au7 lugs?
-
I would just replace the LED and enjoy the amp.