Bottlehead Forum
Bottlehead Kits => BeePre => Topic started by: Jamus on February 01, 2016, 02:56:21 PM
-
2 terminals have voltages out of spec, not sure if this is something to be concerned about...
Terminal 17: should be 90-110v. Mine is 125v
Terminal B2: should be 90-110v. Mine is 135v
All others terminals were good
Thanks!
Jamus
-
17 and B2 are connected by a piece of wire and cannot be different by 10V.
If you swap 300B's between channels, does this voltage move to A2?
-
Before switching tubes I rechecked voltages and this time they were equal (142v each).
I swapped tubes and the error followed. Now terminal 2 and A2 are both 135v. Terminal 17 and B2 are both 99v. Bad tube?
-
Before switching tubes I rechecked voltages and this time they were equal (142v each).
That's a 300B that's not conducting.
I swapped tubes and the error followed. Now terminal 2 and A2 are both 135v. Terminal 17 and B2 are both 99v. Bad tube?
I would take your soldering iron and reheat the pins on the 300B that's acting up, with a special emphasis on pin 2. If you have enough heat to get the solder to flow, there's a small hole on the end of each pin, and a little extra solder can be added.
If these are old 300B's with years and years of use, this may not do you any good.
-
Thanks PB, I'll try that tonight. This is a brand new Sophia Electric mesh plate, still under warranty. In my shoes would you just send it back or is this a pretty minor but permanent fix? I just don't want to end up right back here after warranty lapses in 30 days.
-
Oh, I would send it back and not solder on the pins. If the factory has any questions about the operating point of the 300B, let us know and we can help with that.
-
Thanks again, this is helpful information.
Jamus
-
I got replacement tubes today and rechecked voltages. Now all 4 relevant terminals (2, A2, 17, B2) all measure 121v. This is only 10% over spec (90-110v) so I assume this is okay (?). All other voltages were within spec.
Next I hooked up CD player and BH crack to listen. All sounded great at first, listened for about 5 minutes. I left it playing for about 30 minutes to continue warming up. When I came back to listen there was almost no volume.
I took the BeePre out of the chain and the Crack works fine. I let everything cool down for a few minutes (top plate on BeePre was very hot). When I fired everything back up it sounded normal for about 5 seconds, then volume dropped down again.
Any suggestions on where to start with troubleshooting?
-
Recheck all the voltages after the sound drops out.
-PB
-
Now those same four terminals (2, B2, 17, A2) measure 142v each. All other spots are still in spec.
For what it's worth when I sent the last tubes back to Sophia, they said they remeasured both and they worked fine.
-
You have 142V and still ~5V DC or so on the ungrounded end of each parallel pair of 8 Ohm resistors?
-
Correct. 5.11 and 5.16
-
Can you very, very carefully take a measurement of the DC voltage at pin 3 on each 4 pin socket?
This is sounding a lot like a missed or flaky solder joint somewhere, but it's strange that it happens in both channels.
You have plate voltage, heater voltage, and bias voltage, so it is likely a mechanical connection that isn't allowing current to flow properly.
-
0V for both
-
I reheated all terminals around this area (1-20, 4 pin sockets, C4S boards). No change.
Going back to the resistance checks these 4 terminals only reached 75k ohms. Could that be a clue to what is going on here?
-
As long as you have 0V on pin 3 of the 4 pin sockets, then there isn't any concern for unusual operation conditions due to grid voltage variations.
It might also be good to ensure that the tubes are making good contact with the socket pins.
-PB
-
Tubes do seem to be making good contact but I'm not really sure how to confirm it.
I noticed a new problem that wasn't there yesterday. The two LEDs on the C4S board, B side, near the front of the chassis light up on start up but then quickly go out. (Edit: perhaps not completely out but very faint)
-
You might try reheating the center legs of all the MJE5731A's. These can be tough to get a good solder joint on, and you practically cannot overheat these transistors when soldering them, especially on the center legs.
-
Now that you mention it a few of those did look a little suspect. I reflowed them all but no change in the voltages or the LEDs dimming on the B side.
-
You can also try poking around with a wooden chopstick to see if you can find the suspect connection that is causing this problem.
-PB
-
Just another observation in case it helps in diagnosing this problem. The voltage at these four terminals in question spikes to 215v when the unit is powered on, then they all settle at 142v within a few seconds.
Also, the issue with the B channel LEDs dimming once the tubes heat up moved to the A channel when I switched tubes. This is still on the "B side" of the C4S board and they are still lit, just faintly.
-
Just another observation in case it helps in diagnosing this problem. The voltage at these four terminals in question spikes to 215v when the unit is powered on, then they all settle at 142v within a few seconds.
This is the EL84/TL31 regulator powering on and doing its job.
Just out of curiosity, did this BeePre ever work properly? (I had assumed it did, but I'm realizing that this might be an error)
-
I got replacement tubes today and rechecked voltages. Now all 4 relevant terminals (2, A2, 17, B2) all measure 121v. This is only 10% over spec (90-110v) so I assume this is okay (?). All other voltages were within spec.
Next I hooked up CD player and BH crack to listen. All sounded great at first, listened for about 5 minutes. I left it playing for about 30 minutes to continue warming up. When I came back to listen there was almost no volume.
I took the BeePre out of the chain and the Crack works fine. I let everything cool down for a few minutes (top plate on BeePre was very hot). When I fired everything back up it sounded normal for about 5 seconds, then volume dropped down again.
Any suggestions on where to start with troubleshooting?
Only for less than 30 minutes :'(
-
Just going back over the manual step by step looking for any possible errors on my part. On the C4S boards does it matter which 2 out of 3 bB terminal holes are used for black wires?
-
Ah, OK, I should've picked up on that before I had you send off the tubes. This is almost certainly flaky solder joints in the build, I would just reheat absolutely all of them. Also be sure that all the heasink hardware is on tightly.
-PB
-
Welp, I just finished reflowing every joint in the entire kit and I still get the same 142v at those 4 terminals. I also went step by step through the manual to be sure I didn't miswire anything. Any other ideas?
-
Just going back over the manual step by step looking for any possible errors on my part. On the C4S boards does it matter which 2 out of 3 bB terminal holes are used for black wires?
Just want to rule this out as a possible source of the problem. Are there 3 bB terminal holes on the C4S boards and we are using 2 of them or am I misunderstanding the manual on that? If so, does it matter which 2 we use (adjacent holes) or are they all connected anyways?
-
The three bB pads are connected.
Some photos of your build may help, another pair of eyes can often spot a mistake that you missed.
-
Thanks Chris!
My next plan of attack was to poke around with a wood chopstick while measuring DC volts at the offending terminals. I figure if I see any improvement in the conduction I'll know where to concentrate my efforts.
It was such a downer after reflowing and rechecking every single terminal to see no improvement whatsoever but it helps to know there is such a supportive community on here ready to help.
Thanks,
Jamus
-
I tried probing most terminals while measuring volts at both terminal 2 and 17. Neither budged from 142v while I probed away. I don't think that rules out a flakey connection but after reflowing all terminals I'm starting to lean away from that as the cause (I could be wrong of course!).
I did notice that on the C4S boards, both inboard heatsinks were very hot to the touch while both outboard heatsinks were room temp. Is that normal or could that be a clue to the source of the problem?
Since both sides of the unit are affected I'm wondering if there are any fragile transistors or resistors on either PC boards I could have fried during the build which I could imagine doing twice if I'm using the same bad technique.
Any ideas for next steps? I'll upload some photos if that would help, any particular areas to focus on?
-
One more observation that might help. Now the LEDs on the B sides of both C4S boards are very dim compared to the A sides of each board (when I first turn it on all 8 lights are strong, then the B sides dim).
When I first completed the build a few weeks ago I'm certain all 8 lights remained bright. Then a few days ago I noticed just 2 lights would dim after the unit warmed up. Now it's both channels, 4 LEDs. "A" side LEDs remain bright.
-
How much would it cost to replace both PC boards? :-[
-
Cool heatsinks tend to indicate that the 300B's aren't drawing any current.
When you have 5-ish volts on the non-ground end of the 10W resistors by each 300B socket, that let's us know that the 300B's are drawing current and heating up. This should be sufficient to draw plate current, but your high voltage at OB and cool heatsinks tell us that no plate current is being drawn by the 300B.
This is further complicated by the fact that both channels exhibit the same problem.
Rather than buying new boards, it might be a good idea to have a second set of eyes look over your build. This can be accomplished through the repair service, or by posting your approximate location and hooking up with a fellow Bottlehead builder who can have a look under the hood. A few people have even had their significant others check their work with success.
-PB
-
Thanks PB. I was reflowing all terminals on the boards and ended up with a solder blob bridging across 2 leads on a PN2907A transistor. After trying unsuccessfully to remove the blob from this tight spot I ended up pulling the transistor off the board.
After cleaning it up I proceeded to break it trying to reinstall it. The problem is there was still residual solder in each hole that was a complete PIA to remove. I'm afraid the board may be shot because of those efforts. Picture below, notice you don't see those nice shiny metal collars around the 3 holes where the transistor belongs. Will those still conduct?
-
Yeah, that board is kaput. You can contact replacementparts(at)bottlehead(dot)com about replacement boards/parts.
-PB
-
To avoid this in the future pick up a vacuum solder pump, the type with a plunger. They are inexpensive and work well for clearing liquefied solder from pad holes so you can more easily remove and replace components without board damage.
-
This stuff works great too, it will suck up solder like a sponge http://www.mouser.ie/ProductDetail/Plato/1810-5F/?qs=%2fha2pyFadugWAcSf2iQEFtymdo91GCxIBSLwI45ystI%3d (http://www.mouser.ie/ProductDetail/Plato/1810-5F/?qs=%2fha2pyFadugWAcSf2iQEFtymdo91GCxIBSLwI45ystI%3d)
-
Thanks all. The thing that kills me about this is there was nothing wrong with this transistor until I started mucking around looking for the bad connection. I can fix this but I'm still no closer to fixing the voltage issue. With a Kaiju on order I'm feeling the pressure to get this finished. I'm thinking I should just send this in for service. Should I rebuild the board first or send as is?