Bottlehead Forum
Bottlehead Kits => Eros Phono => Topic started by: tintin on January 02, 2023, 04:00:09 PM
-
I have just completed an Eros 2 build that did not pass the final voltage checks on the C4S board and the D1/2 D LEDs do not light- going back, the voltages at the MJE5731A on the shunt board are different. Previously at the check point before installing audio circuit, they were both reading 210V, now one MJE5731A (side A) reads 150V while the other reads 210V as before.
IA=IB=270V
OB= Breg side B 210V
OA=Breg side A 150V
Kreg side A = 3.5V
Kreg side B= 7.1V
On the C4S board
IA=220V
IB= 150V
OA= 155V
OB= 145V
OC= 95V
OD= 145V
OkA= 95V
OkB= 145V
OkC= 1.4V
OkD= 0.2V
All tubes are lit fine and I tried another set. I have checked the connections on the PCB board and surrounding 12AU7 socket and everything is ok- am wondering which component or area to check next. Am wondering, given the MJE5731A were correct before installing the audio/RIAA circuit and C4S board, if anything there could affect the readings upstream.
Thank you
-
Solder joints, solder joints, and solder joints.
If you unhook the wire going into IB on the forward C4S board, does OA on the D socket PC board pop back up to around 220V? If so, you should be carefully checking for problematic solder joints on that forward C4S board, as there are likely issues there.
The 0V at OKD with 145V at OD makes me think that the shielded twisted pair wiring feeding pins 4 and 5 on the EF86 socket on that side may not be well connected and allowing that tube to conduct.
-
Hi Paul thank you for the quick response- yes disconnecting IB makes them return to 220V, so the issue is downstream from the shunt. I will keep looking. Thanks
-
Solder joints, solder joints, and solder joints.
If you unhook the wire going into IB on the forward C4S board, does OA on the D socket PC board pop back up to around 220V? If so, you should be carefully checking for problematic solder joints on that forward C4S board, as there are likely issues there.
The 0V at OKD with 145V at OD makes me think that the shielded twisted pair wiring feeding pins 4 and 5 on the EF86 socket on that side may not be well connected and allowing that tube to conduct.
Have just checked the wiring for pins 4 and 5, they are well connected on both EF86. Is there any other point which is likely to cause this, or should I suspect the circuit board or the pins? Trying to narrow down on where I may need to look. Thanks
-
I have just reflown all the solder joints and connections of the lower 3 sockets, as well as the C4S board. I checked all connections comparing the incorrect side with the correct one, with a multimeter and nothing is detached or wrongly connected. The problem still persists. I notice that the voltages are equal bilaterally if I disconnect the red wire to OC and OD, if that means anything. Are there any further tests to narrow down the problem?
-
Images of the board and affected EF86 side
-
Was this build completed with lead free solder?
-
The solder joints on the board generally look good, but in a few spots I see a joint that has either a small gap between the lead and the pad or a solder fillet that doesn't completely surround the lead and cover the pad evenly. Those could be cold joints so it might be worth reflowing them one more time.
-
Thanks for the ongoing assistance- Paul I have used the same lead solder as in previous projects, definitely not lead-free.
Doc thanks, I have gone back over the board, and added solder, my board now reads
IB= 220V (correct)
OB= 210V (OB=OD=OkB)
OD= 210V
OkB= 210V
OkD= 3V
The left side remains within the correct limits. I connected the right side of the board to the other EF86 and 6922 triode, and the voltages read the exact same as the above points. Since the board is symmetrical, am I correct in concluding that this completely localises the issue to something on the B side of the C4S board then, or could it still be something off the board? Is there anything I can test to further isolate or rule out a faulty component, bad joint or overheated pad? The corresponding MJE350s read the same as above so could it be those?
Thanks again
Edit- I began tracing the voltages and comparing the 2 sides through R1, PN4250 and the MJE350- all legs read 220V until the middle leg of the A side MJE350 becomes 155V, while the middle leg of the MJE350 remains 220V. Does this suggest a faulty MJE350 or could it still be a problem elsewhere? It is a little difficult for me to source parts is why I haven't tried installing another immediately to test.
-
Could you post some photos of the C socket?
-
Apologies forgot to post these before- I have since fixed up the excess solder on pin 4 and there is no contact with the adjacent pins, but the problem remains the same. I used a multimeter and checked connections from pins to their attachment points and they are also well connected.
-
If you have an adjustable temperature soldering iron, I would turn it all the way up. A lot of those joints don't look like they have had the chance to flow out properly, which will happen with insufficient heating.
-
After taking a break from the project, I have returned to it and reflowed all the solder joints with the maximum temperature on my iron. Unfortunately the problem still remains- are there any further resistance or voltage checks I could do to try and narrow down on the issue? Otherwise I may need to send it in to you guys for repair. I have checked the R1 resistors to ensure they are the correct 34.7 ohm ones as well. Thank you
-
Can you post some fresh pics of your resoldered kit?
-
Here are photos of the EF86 and 6922 sockets, as well as the boards
-
I would pop a fresh 2N2222 in on the offending side.
-
Thanks Paul part is on order- unfortunately I think I may have created another problem while testing; I accidentally shorted the blue wire from terminal 21 to the tab on terminal 28, there was a small spark and now the corresponding MJE5731A reads 170V. Is it possible I have damaged the transistor, or another component? Will wait for your guidance on any further tests and if I should order another part too. Thanks
-
There is no blue wire that goes to terminal 21, do you mean the blue wire on 28 touched something it shouldn't?
I would advise disconnecting the blue wire that leaves that half of the shunt regulator and report the OB and KregB voltages.
-
Apologies, yes the blue wire from terminal 28 touched terminal 21.
With the blue wire disconnected, OB = 175V, KregB = 0.11V. OA= 220V and KregA=1.9V within range I believe.
-
The 431 regulator on that side appears to be blown.
-
Thank you for the speedy diagnosis, will order another 431 regulator as well and work from there. Is there reasonable suspicion for any other parts to be ordered in case they may have been affected too? Just trying to save on shipping costs in case.
-
Just to confirm, will any of these be suitable? The only difference I can see in specs is the operating temperature.
https://www.digikey.com.au/en/products/detail/texas-instruments/LM431AIZ-LFT1/3901932
https://www.digikey.com.au/en/products/detail/texas-instruments/LM431BCZ-NOPB/364136
https://www.digikey.com.au/en/products/detail/texas-instruments/LM431CIZ-NOPB/364143
Thanks again
-
https://www.digikey.com.au/en/products/detail/onsemi/LM431ACZ/1049094
-
Got it, will update with results when parts arrive. Thanks again for the ongoing support, much appreciated
-
I have put in a fresh 2N2222A and still have incorrect readings
With the blue wire disconnected, MJE5731A reads 220V correctly, with it connected, it reads 150V. Further:
IB= 150V
OB= 150V
OD= 145V
OkB= 145V
OkD= 0V
I am sure the 2N2222A was installed correctly, as the legs are 0 resistance to other components connected via the PCB.
Am waiting on the LM431 part which should resolve the shunt board, but that is a problem in the other channel.
-
The 0V at 0KD would still make the 2N2222 suspect. I believe we have also seen similar issues to yours when a trace on that PC board up front has been damaged.