Mark Levinson No 27 Problem

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I am having intermittant problem with a No27 Levinson.
Breaker trips on front and emits a buzz in speaker on one channel.
However, if I ramp up with variac to 50v one channel has alot of DC offset (300mv) and after about a minute it drops to 10mv. At that time I can continue to ramp voltage up to 120v without tripping breaker. The amp plays fine after doing this startup procedure. When power is removed it goes back to the same problem on restart.
Could this be a bad cap? Its difficult to trace because using the variac seems to temporarily cure the problem before I can start troubleshooting.
 
It is a terrible amplifier.
Looking at the schematics in service manual.

Shows the cascode is 'roender-style' with current source to bias.
 

Attachments

  • marklev_input.png
    marklev_input.png
    40.2 KB · Views: 1,048
It was a cold solder joint on the daughter board.
There is a couple resistors on that board that get really hot which is why it was intermittent. This is a difficult amp to work on since it really cant be tested with power on due to the construction. You really need to inspect and test all devices that are suspect. Leaky transistors are the worst ones to find. But start out at the heat sources and change any caps located near these sources or test them with an ESR.
Did you bypass the input board and inject a signal? Do the speaker relays open?
I'll send you the schematics when I get home tonight.
 
Yes, I injected a signal and can read it at the input to the main PCB. I do not get the signal out for the unbalanced input, but I do get a 50 Hz hum out when I inject a signal at the balanced input. Also the DC voltage at output is quite high (around 120 mV). The speaker relays open.

However the negative rail voltage of the input and voltage gain stages are 20 V less than should be (instead of +- 93 V, it is + 93 V and -70 V). So, the failure may be located in the VB-5PC board (although probably not very likely, as I assume that there is just the rectifier function located for this power supply) or it may be located on the AC8 PCB. there the cause could be in fact primarily a bad cap or a cold solder point or also leaky transistors all located where thermal cycles occur. However before going down this route, I want to make sure that it is not something on the VB-5PC board. BTW, I understand that to disassemble the AC8 PCB the big smoothing caps need to be removed. Or is there another way of disassembling?
 
Hello John
I have a trouble that LED at the front panel drinking. My friend say that indicate trouble happens in No.27.5 amplifier. I would like to know what happens. I read you have schematics, operation manual and procedure. Could you send me those by e-mail kindly?
My e-mail address: sasai@sa2.so-net.ne.jp
/Best regards, Toshi

http://www.munsonandbryan.com/downloads/No27_5_test_procedure.PDF
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.