Bottlehead Forum
Bottlehead Kits => Crack => Topic started by: Mark Ferring on February 24, 2016, 06:28:41 AM
-
Built standard Crack. All voltages good, ran for the last month and a half with no issues.
Started building Speedball. Ran into a problem.....
All LEDs on (2 9 pin socket and 4 on small board), went thru the troubleshooting flowchart leg for OA or OB above 110 and all checks out OK. However, my voltages are off by a factor of 2 at all 4 small board locations.
That is, OA "and" OB are around 150, not 75ish..........And IA "and" IB are around 380, not 190ish.
BTW, B-A/B is zeroV
This seems to imply something that is common to the whole board and not a particular solder connection or component (since the entire board has this problem).
Any thoughts as to what might be affecting these voltages by a factor of 2? Is it possible that the 12AU7 tube has a problem?
Mark Ferring
-
If your using the same tube that you have been using it's not logical that it would be the problem. Check that you have the correct resistors in each position.
-
Thanks for the reply Doc. I appreciate you helping with this.
Yes, I do believe that I have the resistors in the correct position. Here is a picture.
-
What is funny to me is that I am getting both the same 2X factor on both the A and B side. So either there is a common single issue, or I made the same mistake on both sides.
-
Looks OK on the top. Do you have both tubes installed when you are doing the voltage check? Heaters glowing? Do the LEDS on the nine pin socket light? Sure you are grounding your black test lead at T12?
-
Just to be sure, I did it again.....
Both Tubes installed? Yes
Heaters Glowing? Yes
LEDS on Nine Pine Socket light? Yes
Grounding black test lead at T12 Yes
For full disclosure, I did have the PN2907 on the B side installed backwards initially. I swapped that around and the picture is where we currently are. I did power up the amp with the PN2907 in that backwards configuration. Not sure if that would possibly have any effect......
Mark Ferring
-
You can measure resistance between each possible pairing of the three transistor leads and see if any read shorted, i.e. like less than 50 ohms. However it does not seem likely that a problem with one 2907 would affect both channels. And if all of the LEDs are lighting my hunch would be that either the voltage measurement itself is inaccurate for some reason or the R1 resistors are not correct or not soldered in well, creating too little current draw thru the tube and thus elevated voltages. They appear to be the right value in the photo. You might want to measure their resistance just to be sure.
Maybe if we could see a picture of the underside of the board we could spot something.
-
Measured both R1 resistors without power and they read ~245 ohms.
Attempted to reflow all four solder joints for the R1 transistors.
Re measured Voltage: Same.
Few pix of the back....
Darnnnnn
-
Oh, and I measured all the pairing combinations for the transistor. They were all open circuit.
-
OK the resistors seem to be in the ballpark. Maybe try a different DC voltage range scale on your meter to see if the voltages read about the same. Sometimes meters can read weird on one range setting and fine on another in certain circuits. You might also reflow all of the connections at terminal 3. That connection is common to both channels. One possibility I can think of is that the connection of the big white ceramic resistors has gone cold from the new connections made at T3, and that means the 6080 is not conducting properly. Since this is a circuit where the two tubes are directly connected that could pull the 12AU7 and small board voltages way off.
-
Just reflowed the T3 and re measured. I used two different voltmeters and two ranges on each. Still the same voltages....
I will say that for a short period after I noticed the PN2907 was backwards, the voltages were in the appropriate range. I thought all was well at that point. Then after I desoldered and rotated the transistor to the proper location, the voltages were wrong again. So there might be something simple here. But heck if I know what it is. I have checked and rechecked all the circuit connections and reflowed most of them already. Fairly frustrating at this point of course!
Happy to try anything else. It simply looks like it should work at this point.....
-
OK, I don't have a good theory for why the pn2907 would affect both channels. Let's try to avoid yanking out parts until other less invasive things are tried. Could you measure voltage at terminals 7 and 9? That will help give us some idea of the functionality of the 6080.
-
This may be clutching at straws, but it also looks from the underside shots like possibly one leg of the removed and reversed pn2907 might not be going all the way thru the hole. You might just reflow those 2907 connections one more time, paying attention to how the solder pulls in around the pads.
-
T7 216V
T9 219V
Reflowed 2907 pins. No change in voltage.
-
My logic on this is that the only thing that has been physically disturbed in the connections of the 6080 during the small PC board installation is terminal 3, where the big cement resistors that determine that current and voltage connect to ground. Sorry to keep coming back to this but a bad solder joint at T3 seems the most likely candidate if the 6080 is installed correctly and it glows when you power on. There are four things connected at terminal 3 - a black ground buss wire that goes from the volume pot to 3L, the black wire from the center terminal of the nine pin socket to terminal 3L, the black wire from the B- pad on the little PC board to terminal 3U, and what would seem to be the potential problem, the twisted leads of the 3K 10W cement resistors inserted into terminal 3U.
Power off, unplug from the wall socket and a wait for a couple of minutes. Then try measuring resistance from terminal 7 to terminal 12 and terminal 9 to terminal 12. Should be a little under 3K for each one if that connection at T3 is solid. You might try wiggling the resistors at the T3 end gently to see if you can make the meter jump around.
-
No worries coming back to the same thing. I get your logic completely and am happy to keep trying things. I really like following any leads that would have potential voltage effects on both the A and B side. I have built a few kits in my day, but I have tremendous respect for your history and experience. I will gladly follow any lead you suggest.
To that end, unfortunately, T7 and T9 were both in the 2930-2500 Ohm range. I wiggled the big cement resisters around (and every other wire going into T3) and could not get the reading to change.
Quick question. One of the leads on each of the 237 Ohm resistors is really close to one of the leads of the 2907 transistor. Are there any potential issues if the solder on those two joints touch each other? I don't think they do but my old eyes are having a hard time telling (and this is adding my reading glasses to my driving glasses for extra magnification!).
One other quick question for the learned master......I don't remember the math from my old engineering circuits class 40 years ago, but is there any single place in the circuit where a bad connection or component could cause this almost perfect doubling of all the voltage readings?
-
Can you power the amp back up and measure A3 and A8? should be around 1.5-1.6VDC.
-
Hmmmm
They both read 2.5
-
OK, let's take a small sideways step. Carefully check the AC volts across power transformer terminals 4 and 5, which supply the heater current. Should be around 6.3VAC. Just want to verify the meter calibration.
If that is OK, I think maybe it is time to consider replacing the PN2907 that was in backwards. I don't have an exact scenario in my head of how one 2907 would throw both sides off and pull the cathode LEDS up from 1.56V, but that should more or less take you back to where things went haywire.
Another thing you can try is pulling the 6080 and remeasuring DC voltages OA, OB and A3 and A8. I don't imagine they will change, but it might give some more info.
-
Maybe one more step before replacing the 2907 -
I keep coming back to T3 connections being the culprit. One more reflow of all the connections on T3 might be worth trying. If that doesn't help you might want to try temporarily disconnecting IA/IB, OA/OB and reattaching the original 22K resistors to verify that voltages go back to normal. That would narrow the issue down to the board itself and not the ground connections.
-
Transformer across T5 and6: Measured 5.3V for two different meters
6080 pulled: No voltage change at OA/OB or A3 or A8 (actually small drop of voltage on OA/OB to around 141V)
Reflowed T3 one more time. Did not change anything.
Reconnect original 22K resistors: I foolishly clipped them off at a point where I cannot use them.
I would be happy to go back to the original 22K resistor config or try a new 2907. Is there a way to source those parts from you? I live in the middle of nowhere in the mountains of New Mexico and don't have access to an electrical supply house. I would be happy to pay of course.
-
Transformer across T5 and6: Measured 5.3V for two different meters
Hopefully that's a typo and you measured from T4 to T5. 5.3VAC is kind of low. What is your AC mains voltage at the wall outlet?
If there is just a little stub of lead left you could temporarily clip lead or tack solder the 22K resistors to the red and white IA/OA pair and IB/OB pair of wires after disconnecting them from the PC board.
You can order parts by emailing [email protected]
-
No typo. Measured again (after going back to original configuration) and still got 5.3V. Outlet is 119.8V
Thanks for the experienced tip (regarding resistors). I got them hooked back up and .........holy cow....Here is what I got.
I reran all of the original resistance measurement checks. All were as expected per the manual.
Now for voltage. I think we found our general problem. The below seems to imply it is not the board.....I assure you that originally they all measured normally. I then listened to my HD650s (with much pleasure) for over a month. However, now they are all messed up. I will only provide the measurements that were not "expected"
T1 163
T2 385
T4 385
T5 161
T7 226
T9 226
T13 385
T15 431
T21 478
A1 162
A6 163
A9 5.3 vs 0
B1 164
B2 391
B3 230
B4 161
B5 390
B6 233
B7 5.4 vs 0
-
No typo.
Sorry, I think I wasn't very clear. I was asking for an ACV measurement from T4 to T5 on the power trans and you said you measured 5.3V from T5 to T6 on the power trans. What I meant was I hoped that T6 was a typo as it should be from T4 to T5.
Add to that that you are seeing 5.3V (is this AC? DC?) at A9 and B7 and I am further confused. T4 of the power transformer is one end of the 6.3VAC heater secondary winding. It should be connected to ground at either T22 or T14. And that should mean that the other end of the heater secondary at T5 of the power transformer - which is connected to A9 and B7 - is effectively connected to DC ground as well and should also read 0VDC. If you are seeing 5.3V it is either 5.3VAC and you need to switch your meter back to DC or something is floating that should be tied to ground.
All of the other readings should be DC volts too.
See if you can sort that heater supply stuff out first and then we will move on to the other high readings. At this point and assuming the meter readings are all DCV I would be inclined to look for a loose ground connection somewhere.
-
OK, if the meter has been set correctly to DC on the last round of measurements and you were in fact measuring 5.3VDC at terminal A9 I have another thing to try. This would assume that the 12AU7 might have a short from heater to cathode.
Pull both tubes out of the amp. Power up the amp and measure AC volts from power trans terminal 4 to power trans terminal 5. Should be 6.3VAC or a little higher. Then power down and unplug from the wall.
Switch meter to DC volts. Connect black test lead to T12, connect red test lead to A9. Power back up and measure DC volts at A9. Should be 0VDC. Be careful going into the amp after this, the filter caps will store some charge for a while without the tubes being in there to drain them down at shutoff.
-
Another possibility is terminals or leads attached to terminals on the A socket touching each other, that shouldn't be touching.
-
Weeeelllll. I think I have likely been a prize idiot. I was measuring all of the terminal voltages in the AC setting. Now that I went correctly to the DC setting, all of the voltages were normal. I would guess that the small board is also OK and that I was using AC vs DC as well.
I will now add the board back to the circuit and see if this is indeed correct. And if so, my sincere apologies for the wild goose chase......
-
Went back with tubes out and measured across the transformer t4 and t5 (the 6.3V heater circuit). I measured 6.1ACV
And yes, I did use AC previously and it was across the correct terminals T4 and T5. But I used a different meter which perhaps is a bit off calibration. It still measures around 5.3. But the other one measured 6.1.
I am off to add the small board back to the unit and recheck voltages in DC.
-
OK, that's great. This makes for a useful lesson that used to be pounded into our heads when I was a physics major - a number is worthless without units. I will remember to remind about the proper meter setting up front on these threads.
-
So amazingly enough, this has been fixed since I swapped the transistor....many days ago......Lesson learned also.
Put the board back in. It passed all checks. And I listened to a song or two on my HD650s. Marvelous improvement just with the input stage mods.
Well, at least I had time to put together the big board already (along with extremely careful construction of the heat sinks per the directions), so I ought to be able to complete the rest of the kit in a few hours.
I cannot begin to express my appreciation for your hanging in there with my wanderings when all it required was a rotation of my meter. Your patience and forbearance was beyond kind, and you have my eternal thanks.
And I will be certain to continue to shop here and recommend it heartily to my friends!!
Now, off to complete the kit....with much lighter spirits!
-
Just to close this out. I installed the large board and completed tests with no further issues. Thanks for all the help. It sounds great.....
Mark