Bottlehead Forum
Bottlehead Kits => BeePre => Topic started by: Bryon on March 09, 2018, 11:00:38 AM
-
The left channel on my Beepree is acting oddly. It will often play perfectly for hours then the left channel fades away, then fades back in, then fades away again. It seems different than the sharp sound break that a short makes and there will sometimes be a very quiet signal for a while before and after it cuts out. One of the more interesting aspects that I just noticed is that when I turned on my Crack, which is connected to a second set of outputs on the Beepree and the same power distribution [Furutech e-tp80,] the channel came back - the next time the channel didn't come back until the Crack was turned off. So the power fluctuation seems to influence the fade in/out.
After reading PB's response to James about a similar problem, I reflowed the connections to the MJE5731A without any improvement. Does this sound like it could be the LM431 regulator [another of PB's suggestions to James] or one of the transistors?
It is very frustrating, as when it is working, the BeePree sounds amazing, but it has been on the bench so many times, it's starting to believe that it's at home there.
Any suggestions would be very appreciated.
Bryon
-
The description sounds like the heater or filament of one of the tubes is losing power. Look to see if the tube on that side is glowing when the sound cuts out. If so it may be a bad pin contact with the socket or a loose heater wire connection.
-
If at all possible, taking voltage measurements will be very helpful.
In the other post that you're referencing, a loose wire was responsible for the issue.
-
It was a loose heater wire in the case of my BeePre. But I still have another intermittent fault in that Left channel where the channel would make noise and the two B side LEDs would go out. Yesterday I swapped the two C4S boards and the fault follows the board. So at least I know that the board is as fault, not any of the connections to the board or tubes. I've ordered a new C4S board to get rid of this intermittent issue since I can't find any broken leads or bad solder joints on the bad board.
I can sympathize with the frustration of having an intermittent BeePre as it is really amazing sounding. And bringing it back and forth to the bench is not a great way to treat a preamp.
James
-
Thanks Dan and PB for your suggestions and James for your empathy - I'm not quite at the point of wanting to change out the C4S boards but maybe soon. Hope this works out for you - I feel your pain.
Dan - I've resoldered all the tube connections and checked for continuity a couple of times as that was one of my first thoughts, but to no avail. When I was building the Beepree, I thought I'd optimize the tube sockets so I put in teflon sockets as I had read they reduced vibration. After the build I read a number of posts on the forum about how people were having issues with them, so maybe they are the problem. You're probably right but I'd like to make sure they are the problem before I attack the sockets.
PB- I rechecked the voltages again, but as before, most of them are spot on, with a few out a few volts, but well within 5%. The outliers are 189.5V at terminals 10 and 15 instead of 186V and terminals 22, 25, 40,43,46 at 217V instead of 213V. I was a bit worried about B2 and B3 until I remembered they were reversed.
B side of the preamp connects to the Left Channel - right? It certainly looks like it to me, but I've made bonehead mistakes before and wouldn't want to be messing with the 'good' side and ending up with neither channel working. I haven't checked the A side 4-pin sockets - should I?
thanks again for your input
Bryon
-
If it's a filament issue -
Run the preamp right side up. Check for filament glow. Tilt the tube around in the socket and see if the glow cuts out. Having the preamp upside down or on its side can affect this condition, so you really need to have the preamp in an upright operating orientation.
-
Thanks for the suggestion - I put the preamp back in after doing the voltage tests and, of course, everything is working perfectly. The tubes continue to glow as I tilt the tubes and the channel doesn't cut out. If I remember correctly, the tubes always glowed, even when the channel went out.
Bryon
It worked for about an hour, then the left channel cut out - if it is the teflon sockets, is there one which is more suspect than the other ? Which one should I replace first or are there things to do first?
Bryon
-
Oh, yeah, I don't have a lot of trust for PTFE sockets, at least not till you get to the 4 pin jumbo ones. If you still have the ceramic sockets that came with the kit, I would give them a try.
Also, when the signal drops out, pay special attention to the POS OUT voltage on that side.
You are correct that when the preamp is facing you and right side up, the right channel goes through the right side of the chassis. When you flip the preamp over, then the orientation swaps so that the right channel goes through the left side.
-PB
-
Thanks for the response PB - I was afraid I might have to change out the sockets - something I am trying to avoid.
The problem with getting a diagnosis is that when the preamp is turned off and restarted both channels work again for a while before the left signal disappears. I guess I could connect the Beepree to my system, innards up, and wait until it malfunctions to check the POS out but when it is in position to connect to a source and an amp in order to know when the left channel drops, it really isn't accessible.
I don't understand why the left channel reappears when I switch my Crack on or off. Doesn't that suggest an electrical problem, maybe caused by a voltage surge, rather than a mechanical issue, like the sockets? [You might notice, I don't really want to change the sockets and hope it is something else.]
thanks again for your suggestion - I guess I'm going to have to bit the bullet - any suggestions about which socket to change first?
Bryon
-
If you could get a voltage measurement when things drop out, that would be a piece of information that would let us know what you're actually experiencing.
You can hook your phone up to the BeePre and a cheap pair of headphones to the output with clip leads (it will sound terrible, but pass signal), then wait for the dropout, then check voltages.
I would make it a plan to change out the sockets anyway.
-PB
-
Thanks for the suggestion - it sounds like a good plan.
I should have time tomorrow to give it a shot.
I'll let you know what happens -
Can you tell me why turning off and on the Crack brings the channel back for a while?
Bryon
-
Can you tell me why turning off and on the Crack brings the channel back for a while?
If they are close together, then fiddling with the Crack may jostle a loose connection in the BeePre which causes it to run properly for some time.
-
Well, after 21/2 hours of playing tubes down, the left channel was still playing fine and pos out hadn't changed significantly. Pos in was down a few volts on both sides after it had played for an hour - should I be concerned? When I put the amp back into the system, tubes up, it sounded great for a couple of albums and then the left channel went out again. Does this suggest the problem could be heat related?
I tapped the chassis lightly to see if it would jostle the left channel back into action, which it didn't, and then flipped the power switch to the Crack carefully so as not to cause any vibration transfer to the Beepree, and the channel reappeared.
I do have the luxury of having another properly functioning Beepree I got for a second system and I'm thinking of transplanting the C4S and/or Filament regulator boards from the left channel of the working Beepree to the faulty Beepree, before I tackle the tube sockets.
My guess is that playing upside down shouldn't stop the dropout, if the problem was the sockets but it might affect any components that are heat sensitive, given that the 'guts' are exposed, would avoid overheating. I think I'll try raising the Beepree up a few more inches [It is already an extra inch off the shelf] and see if that helps, then perform the transplant.
Would you do the Filament regulator or C4S board first?
thanks for your time and attention to my problem
Bryon
-
I agree that you should try raising it and cooling it before you change out any boards. The problem is that running it tubes down could indicate a heat issue, but could also indicate a tube internal structure or socket contact issue that only sags in such a way as to cut out when the tube is upright and hot.
-
Yeah, I'm still looking for ways to avoid changing out the tube sockets. I've tried a few tubes in each position and none of them affect the cut out problem. I lifted the Beepree and the left channel is now playing but at about half the volume as the right channel through the Crack. It seems related but it isn't acting the same as it was before so there may be something else. I'll try it through the Paramounts and see if that changes things.
If I do have to change the sockets, is there any way to know which ones to try first - the 4 pin won't be as difficult to change as the 9-pin so I'm tempted to start there, but if the facts point to the 9-pin being at fault, then I don't want to mess with the 4-pin.
thanks for your input -
Bryon
-
A lot of the process involves checking voltages when the problem occurs. When you hear the signal playing half as loud check the plate voltage of the 300B at pin 2. I wonder if there might be an issue with the output capacitor on that channel.
-
Can you tell us what your line voltage is? If it's borderline on the low side, turning on the Crack may be temporarily aiding in a low line voltage situation.
-
Both channels have been playing without interruption for about 6 hours - Yay!!!
I decided to transplant the C4S board from my working Beepree first. It played perfectly for about an hour and then the left channel went missing again. Next transplant was the Filament Regulator board and miracle of miracles, it worked.
I am still left with a Beepree [the transplant donor not the recipient] that isn't going to work consistently on the left channel, but at least now I have a general idea of what is causing the problem.
Now that it is narrowed down to the Filament Regulator board, is the most likely candidate for replacement the LM1085 [as it is the only thing on the board that needs a heat sink] or one of the other transistors or regulators?
thanks PB, Dan and James for your support and suggestions - as always you guys rock.
Bryon
I
-
It's hard to say what might not be working without knowing what kind of voltages you were seeing.
When I have servied BeePres with filament voltage issues, more often than not it has been a flaky solder joint on the 6 lug strip under the filament regulator board where the two 10,000uF caps hang out.
I have never had to replace a 1085 regulator in anything I have ever worked on.
-
Given that after the board was exchanged, the problem disappeared, the problem seems to have been a component on the board so probably not an issue on the lug board. Are any of the other components on the board likely to be heat sensitive? Dan mentioned a possible cap issue - so maybe the .22uf?
What voltages would help you diagnose the culprit. My line voltage is high - 121V. I did all the voltages a week ago so I can give you any relevant numbers from then. I did notice a change in Pos out and Pos in. Voltages were Pos In - initially 140V settling to 137V and 9.98 on the Pos Out rather than the 8.6 that I had measured before. That seems high but the channel was working at that point.
thanks again
Bryon
-
-
Do you have the feet on the base, to allow some air flow under the chassis?
-
The base is on sorbothane feet and then elevated a further 3/4" on anti-vibration pads so there should be lots of ventilation. Earlier in the process, I elevated them another inch or so to make sure there was enough air flow but it did nothing to solve the problem. The Beepree also has room on each side and the top for air flow. The only thing I can think of is that maybe I overheated one of the components on the board when I was soldering it, not enough to kill it dead, but just enough to injure it to the point of exhaustion after an hour
Bryon
-
Given that after the board was exchanged, the problem disappeared, the problem seems to have been a component on the board so probably not an issue on the lug board.
The connections on the terminal strip will get nudged when swapping out boards, so this is still a real possibility.
Are any of the other components on the board likely to be heat sensitive? Dan mentioned a possible cap issue - so maybe the .22uf?
The 0.22uF cap is in the high voltage regulator circuit, so no issues there.
My line voltage is high - 121V.
Nah, that's perfect.
I did notice a change in Pos out and Pos in. Voltages were Pos In - initially 140V settling to 137V and 9.98 on the Pos Out rather than the 8.6 that I had measured before.
Pos In should be 10-15V. The voltage that's approximately 150V will develop on the other half of the board that holds the TL431. Having 8.6V on Pos Out would indicate that something in the low voltage circuit needs some help.
-
I seem to recall the prototype needed to have its 33µF and 150µF caps changed after a coupla thousand running hours. You could start there...
-
The bad news - the left channel has disappeared again after 3 hours play time today.
PB - I missed seeing a decimal point on the Pos In - I just remeasured and it is actually 14V on both channels. The Pos Out is now 9.975 or thereabouts on each side-
You may be right about the connections on the terminal strip - I'll try resoldering them and see.
Josh - thanks for the suggestion to change the 33uf and 150uf caps -it's worth a try
Bryon
-
The 10,000µF 10V were also replaced at some point. Might be more useful since it's now showing to not be the board.
-
Something does not add up. You say the tubes stay lit but the filament supply is dropping out.
-
It doesn't make sense to me either - maybe the fact that the channel didn't drop out had nothing to do with changing the filament board and was just coincidental? Would the tubes stay lit if it was the sockets? maybe we're back there.
Bryon
-
Ha, yeah, those teflon sockets... It's really impossible to tell if they are the issue without yanking them out and replacing them, and I'll certainly feel bad if they aren't the cause of the issue!
9.975V and 14V are dead nuts on.
If the tube socket connection gets flaky, you should lose the 5V reading across the paralleled 10W cathode resistors. I would be probing there next.
-
I resoldered the connections to the lug board under the C4S board and the both sides worked for about 5 hours last night so I'm hoping you were right PB and this will be a permanent recovery. I'm gone for a few days, so I'll update when I get a chance to play it for a few days.
thanks again for all your help
Bryon