Marantz 2238B Distorted Left Channel

[catrafter]

Hi Tony,
I would test every transistor and diode to see if that leads to something.
I wonder if there is +35V getting to the R723-R727 junction. It looks to me like the bias path is from +35V supply, thru R723,R725 To the base of Q707 driver (+0.6V), R729/Q735/R733/R731 feed current through Q741 and back to the base of the other driver Q709 (-0.6V).

Tom

 

[mbz]

point b shows the pulsing in the waveform

Ok, I assume this corresponds with noise. Indicates the problem is before the relay and speaker selector, ie, the power amp.

Please probe the input to the power amp, left channel, either at J701. If there are physical links between pre-out and main in then
you can probe the links however I suspect marantz uses a switch. You will have to reduce the Volts/div.

Maybe post a pic of this pulsing/can you identify if the pulse is positive or negative.

Suggest you have a good look for dry solder joints (google for pics) on the power amp board and check wire wrap at J701.

 

[tonefoto]

Ok, I assume this corresponds with noise. Indicates the problem is before the relay and speaker selector, ie, the power amp.

Please probe the input to the power amp, left channel, either at J701. If there are physical links between pre-out and main in then
you can probe the links however I suspect marantz uses a switch. You will have to reduce the Volts/div.

Maybe post a pic of this pulsing/can you identify if the pulse is positive or negative.

Suggest you have a good look for dry solder joints (google for pics) on the power amp board and check wire wrap at J701.

View attachment 1332334

So far I've found and replaced an open 220Ω resistor R745 in line with the emitters on the left channel. There is no pulsing sound, just distortion at any volume on left side. Right still good. I connected my scope on J701 and J702 and fed a 1khz signal, both sides look identical.

 

[tonefoto]

So far I've found and replaced an open 220Ω resistor R745 in line with the emitters on the left channel. There is no pulsing sound, just distortion at any volume on left side. Right still good. I connected my scope on J701 and J702 and fed a 1khz signal, both sides look identical.

I am able to read bias now on the left channel and adjust it to about 20mv, its very jumpy and won't adjust below 12mv. DC is adjustable to zero V

 

[tonefoto]

I am able to read bias now on the left channel and adjust it to about 20mv, its very jumpy and won't adjust below 12mv. DC is adjustable to zero V

I installed the correct PNP output today which probably helped. Here is a waveform at J701/J702 showing equal 1khz input signal, another off the emitter of Q705 (left) and Q706,

 

[blhagstrom]

I should have mentioned in original post that there is NO bias on left channel. I do have a scope but am not too familiar with tracing signals

You have a bad transistor hiding in there.

I've been down that read. They test good on a meter and with test tools but are not holding up under full power.

Make a list of the small transistors and search for recommended replacements.

 

[eiraved]

I haven't noticed any inspection of solder joints being accomplished by the OP in this thread. Unless the bad solder joint just happens to coincide with a transistor replacement, you won't find the culprit. I highly recommend using a magnifying glass and a strong light to perform a thorough inspection prior to replacing any transistors. Reflow questionable solder joints and retest.

 

[tonefoto]

I haven't noticed any inspection of solder joints being accomplished by the OP in this thread. Unless the bad solder joint just happens to coincide with a transistor replacement, you won't find the culprit. I highly recommend using a magnifying glass and a strong light to perform a thorough inspection prior to replacing any transistors. Reflow questionable solder joints and retest.

Checked for bad or cold solder joints on the P700 entire board, reflowed solder where it looked questionable. I think Blaine might be right

 

[mbz]

So far I've found and replaced an open 220Ω resistor R745

Check R741, 220ohms on the same path, might also be stressed.
Diode test Q727, 729 (amp powered off)

Suggest continue with the 1kHz input and check waveform at
Q707base/probe at R725 (ie, don't probe transistor legs)
Q709b/Q709cathode
Q705b/R707

You will probably end up testing all transistors/diodes as per @catrafter

 

[tonefoto]

Check R741, 220ohms on the same path, might also be stressed.
Diode test Q727, 729 (amp powered off)

Suggest continue with the 1kHz input and check waveform at
Q707base/probe at R725 (ie, don't probe transistor legs)
Q709b/Q709cathode
Q705b/R707

You will probably end up testing all transistors/diodes as per @catrafter

thanks for your help on this, I will keep testing transistors and diodes and report back

 

[tonefoto]

Check R741, 220ohms on the same path, might also be stressed.
Diode test Q727, 729 (amp powered off)

Suggest continue with the 1kHz input and check waveform at
Q707base/probe at R725 (ie, don't probe transistor legs)
Q709b/Q709cathode
Q705b/R707

You will probably end up testing all transistors/diodes as per @catrafter

pulled leg out of Q727/729 tested good, signal at R725 looked good. Do you mean check cathode of diode Q729 for checking Q709B?

 

[tonefoto]

thanks for your help on this, I will keep testing transistors and diodes and report back

probed Q709 b doesn't change waveform, probed Q705 b and saw flat waveform

 

[mbz]

Probing transistor legs is just too dangerous. Easier to probe a resistor, diode,,,
connected to the same node. Q709 cathode and R707 are safer areas to probe.

Traces did not give expected results.

Set spkr selector to "A"/System 1, supply 1kHz sine wave and probe the left speaker
post. This should show the distortion. Typically the distortion will be in one half
(+ve or -ve) of the sine wave.

Next step is to probe the base of the following transistors,

+ve
Q711b (R757)
Q707b (R725)<<== already checked/ok'd

-ve
Q713b (R759)
Q709b (Q729 cathode)<<== already checked/ok'd

Q715 and Q717 look like some overload protection. If their base voltage increases
past some threshold they will divert audio via Q727 and Q729.

Probe
Q711e R763 or 739
Q713e R765 or 741

Q727, Q729, Q713, Q733, C719, C721 (10uf/16V) should be tested.

 

[tonefoto]

I discovered emitter resistor R765 reads 31kΩ while the other 3 emitters read from ,39Ω-.51Ω

 

[tonefoto]

Probing transistor legs is just too dangerous. Easier to probe a resistor, diode,,,
connected to the same node. Q709 cathode and R707 are safer areas to probe.

Traces did not give expected results.

Set spkr selector to "A"/System 1, supply 1kHz sine wave and probe the left speaker
post. This should show the distortion. Typically the distortion will be in one half
(+ve or -ve) of the sine wave.

Next step is to probe the base of the following transistors,

+ve
Q711b (R757)
Q707b (R725)<<== already checked/ok'd

-ve
Q713b (R759)
Q709b (Q729 cathode)<<== already checked/ok'd

Q715 and Q717 look like some overload protection. If their base voltage increases
past some threshold they will divert audio via Q727 and Q729.

Probe
Q711e R763 or 739
Q713e R765 or 741

Q727, Q729, Q713, Q733, C719, C721 (10uf/16V) should be tested.

Sorry mbz , I have to backup. I discovered I had my scope settings out of whack but have it somewhat under control now. I revisited with a probe on R725 and R726 for comparison, as well as R707/R708. Photos show R725 and R707
Both waveforms show the distortion

 

[tonefoto]

Probing transistor legs is just too dangerous. Easier to probe a resistor, diode,,,
connected to the same node. Q709 cathode and R707 are safer areas to probe.

Traces did not give expected results.

Set spkr selector to "A"/System 1, supply 1kHz sine wave and probe the left speaker
post. This should show the distortion. Typically the distortion will be in one half
(+ve or -ve) of the sine wave.

Next step is to probe the base of the following transistors,

+ve
Q711b (R757)
Q707b (R725)<<== already checked/ok'd

-ve
Q713b (R759)
Q709b (Q729 cathode)<<== already checked/ok'd

Q715 and Q717 look like some overload protection. If their base voltage increases
past some threshold they will divert audio via Q727 and Q729.

Probe
Q711e R763 or 739
Q713e R765 or 741

Q727, Q729, Q713, Q733, C719, C721 (10uf/16V) should be tested.

Distortion shows up in waveform after probing all the resistors starting from R725 through R765

 

[mbz]

Check the waveform at the left speaker post, this should show the distortion.

Keep in mind that the input signal/sine wave is "split" into a positive half
that is handled by Q707, 711, 715 and the negative half Q709, 713 and 717.



Bit unclear what trace is what.
Top trace, left pic appears to show some slight clipping on the positive half.
Check "clipping mechanism" Q727, Q715, Q731 and C719.
This might just be an oscilloscope resolution issue.

Bottom trace, left pic looks like a input signal, looks ok

Top trace, right pic
Hopefully a pic at R725/Q707b

Bottom trace, right pic looks like a supply rail at R725?

I'd certainly change the emitter resistor R765, this is the probable cause
of the distortion.

 

[tonefoto]

Check the waveform at the left speaker post, this should show the distortion.

Keep in mind that the input signal/sine wave is "split" into a positive half
that is handled by Q707, 711, 715 and the negative half Q709, 713 and 717.

View attachment 1333983

Bit unclear what trace is what.
Top trace, left pic appears to show some slight clipping on the positive half.
Check "clipping mechanism" Q727, Q715, Q731 and C719.
This might just be an oscilloscope resolution issue.

Bottom trace, left pic looks like a input signal, looks ok

Top trace, right pic
Hopefully a pic at R725/Q707b

Bottom trace, right pic looks like a supply rail at R725?

I'd certainly change the emitter resistor R765, this is the probable cause
of the distortion.

The image on the left is R727, the trace shows the distortion clearer on the right image which is R707

I'm going to replace the emitter at R765

 

[tonefoto]

Check the waveform at the left speaker post, this should show the distortion.

Keep in mind that the input signal/sine wave is "split" into a positive half
that is handled by Q707, 711, 715 and the negative half Q709, 713 and 717.

View attachment 1333983

Bit unclear what trace is what.
Top trace, left pic appears to show some slight clipping on the positive half.
Check "clipping mechanism" Q727, Q715, Q731 and C719.
This might just be an oscilloscope resolution issue.

Bottom trace, left pic looks like a input signal, looks ok

Top trace, right pic
Hopefully a pic at R725/Q707b

Bottom trace, right pic looks like a supply rail at R725?

I'd certainly change the emitter resistor R765, this is the probable cause
of the distortion.

Is it permissible to use an emitter resistor value close to the .39Ω 3W called for?

 

[mbz]

Is it permissible to use an emitter resistor value close to the .39Ω 3W called for?

I would be against it but you could go to 0.47ohms, would need to replace R763 with same.
The 0.47ohms may affect bias setting.