TU-8200 Auto-biasing issue

[jimpudar]

Hello,

First of all, let me mention this is cross-posted to Reddit AskElectronics.

I just finished putting together a Elekit TU-8200 stereo hi-fi amplifier.

I hooked it up and it sounds fantastic, but I noticed that the 6L6GC power tube on the left channel gets about 100 degrees F hotter than the right channel. Using an IR thermometer I found the left channel is about 280F and the right channel is 180F.

I tried swapping the two power tubes to see if there was a tube issue, but again the left side got much hotter. Based on the measurements I have taken, it seems that the left side is biased properly and the right side is being biased far too cold.

I have the jumpers on both channels set to "ultralinear" mode. I haven't tried triode or pentode mode.

I took some measurements at the tube pins:

Left Channel
plate: 250 V
grid: -13.5 V
cathode: 417 mV
calculated plate current: 70 mA

Right Channel
plate: 257 V
grid: -30 V
cathode: 75 mV
calculated plate current: 12.5 mA

I double checked all the resistors in the right channel auto-biasing circuit and I also desoldered and tested Q2 (the PNP transistor) with a multimeter and it is working as expected.

Can anyone suggest any further troubleshooting steps around the right channel auto-biasing circuitry?

Thanks!

Jim

 

[jimpudar]

I just measured the voltage at the base of each PNP transistor Q1 and Q2:

R (Q2): 828 mV

L (Q1): 543 mV

Here are some more measurements:

Opamp voltages:

L - (pin 2): 600 mV
L + (pin 3): 314 mV
L out (pin 1): 26 mV

R - (pin 6): 600 mV
R + (pin 5): 79 mV
R out (pin 7): 0 mV

Transistor voltages:

L C: -7 V
L E: 1.16 V
L B: 543 mV

R C: -32 V
R E: 1.16 V
R B: 828 mV

 

[jimpudar]

Since there was so much trouble getting the cathode voltage to the base of the PNP transistor in the right channel, I decided to try swapping the transistors around.

The problem did indeed follow the transistor, so now I need to find a replacement.

Can anyone suggest a replacement transistor which will work the same as the KTA1266Y?

Thanks!

 

[jimpudar]

I had some A1015 on hand, and they seem to work perfectly as a replacement for anyone reading this in the future.

Good luck!

 

[jimpudar]

Since this post has been getting a lot of views, I figured I would post some photographs I took while testing the transistors with a curve tracer.

I compare both the good and bad A1266-Y transistors to a known good A1015.

The bad one is very obviously bad 😀

Comparing Elekit TU-8200 transistors on a curve tracer - Album on Imgur

(Please let me know if uploading the images here is preferable to posting a link to an Imgur album and I will do so.)