Oscillations in output waveform

[svkr2k]

Dear forum members,
I'm repairing old Kenwood Amplifier KA-32. I have replaced the output power transistors and driver components. For testing, I generate sine wave and feed to the amp.

The speaker output waveform was looking nice and clean during the initial no-load tests. After some time, when I connected the output to the speakers, I could see high frequency oscillations in the speaker output waveform. This results in the output transistors becoming very hot.
Can anyone help me understand the reasons for this?

Thanks and regards,
Raj.

 

[djoffe]

a possibility: The zobel network may have opened...usually 5-10 Ohms in series with 0.1 uF...It controls the gain of the output stage, which increases under load, especially if it's a quasi comp amp

 

[orbanp]

You did not mention, but did you replace those components with the same ones or with newer semiconductors?
The newer semiconductors could be faster (higher bandwidth), and they might need different compensation in the amp.

Good luck, Peter

 

[jerluwoo]

Here is the schematic Kenwood KA32 Service Manual free download,schematics,datasheets,eeprom bins,pcb,repair info for test equipment and electronics

 

[wahab]

This amp has a CFP power stage , that s extremely uncommon
on mass produced amps.

This kind of output stage is quite prone to instability ,
so as pointed above it s necessary to use the same devices
otherwise compensation must be adapted.

You can try to increase the 22pF compensation caps to 47pF.
(C37 and C3?).

 

[diy didi]

I have found that sometimes it can be caused by a grounding issue, not within the amp, but by the earth lead of your scope. try running your scope from your isolation traffo, or try disconnecting its earth. This method especially helps when probing "internally set to bridged" amplifiers.
If this doesnt work, check and try previous posts wrt caps etc....

 

[svkr2k]

I shall try the proposed changes.

Also, here are some more information that I missed in my original post:
1. I replaced the transistors, drivers with new but same part number.
2. I have NOT replaced the old electrolytic capacitors.
3. I have not replaced the emitter feedback resistors (0.22ohm, 5W). I noticed a thin line in ONE of them that appeared like a "hair-line" crack, but not sure.

 

[sjdemark]

Have you measured the emitter resistors with an ohm meter? Do they measure around .22 ohms?

 

[svkr2k]

Hi all,

I haven't measured resistance for .22ohm resistors. I Need to get a better multimeter to do that. I will try.

Today, I did:
1. replaced all electrolytic capacities with new ones.
2. replaces C37 & C38 (22pf) with new 47pf ceramic caps.

But, the problem of high frequency oscillations + very loud "hum" is still present.

here are some more observations:
1. when I use only one channel (either Left only or Right only using Balance pot), then the oscillations/hum goes away.
2. sometimes if I increase Treble to maximum, I get oscillations/hum again while running only one channel.

 

[ontoaba]

looks like cable feedback, gained phase lagged positive feedback that went through the cable of output to cable of input because they are long, may be it need RC low pass filter at the inputs.

 

[svkr2k]

Hi all,

Oops ! I forgot to ground the heatsinks !!!

Today, I connected the heatsinks to ground and everything works fine !
Thanks for everyone who gave me ideas.

I still have one issue: Low frequency "Hum" is audible when I set the volume to higher side. (Probably, mains frequency???) Any ideas to adoid this?

Regards,
Raj.

 

[sgrossklass]

See whether it's still there with the inputs shorted.

I never noticed this thing only has a specified damping factor of 20 at 8 ohms (0.4 ohms R_out). Funny little amp.

 

[gootee]

Hi all,

Oops ! I forgot to ground the heatsinks !!!

Today, I connected the heatsinks to ground and everything works fine !
Thanks for everyone who gave me ideas.

I still have one issue: Low frequency "Hum" is audible when I set the volume to higher side. (Probably, mains frequency???) Any ideas to adoid this?

Regards,
Raj.

See if the hum is still there with nothing connected to the inputs.

If not, then it could be that the input signal/gnd pairs have too much enclosed loop area. And it could also be that some conductor pair with mains frequency has too much enclosed loop area.

 

[svkr2k]

See whether it's still there with the inputs shorted.

I never noticed this thing only has a specified damping factor of 20 at 8 ohms (0.4 ohms R_out). Funny little amp.

Hi sgrossklass,
I couldnt understand what you mentioned on damping factor, could you please explain. Thanks a lot !


All,
I believe I should mention this ... I replaced all electrolytics with common grade (probably industrial) and not "audio grade" capacitors. Is this okay? Or how will this affect the performance of the amp?

Thanks !!!
Raj.

 

[gootee]

The electrolytic caps will probably be fine. I would worry more about the matching of the transistors you replaced.

Did you try testing for hum with the inputs open (not connected) and with the inputs shorted, as we asked?