Some help repairing a power amp Kenwood M2

[K-amps]

Need some help repairing a power amp Kenwood M2

Got this Kenwood M2 power amp off eBay, hoping for an easy fix. Does not look that way. Hoping to get some helpful tips on this amazing forum.

Its a weird issue that I have not experienced before.

• All outputs are ok. Amplifier powers on fine and relays click in etc.
• At very low volumes there is sound (less than 1vac RMS), then gets distorted in both channels as volume is raised.
• Pushing a 1kHz sinewave yields a distorted waveform that traces all the way back to the upa68h input differential.
• There is No bias in the output stages, little to no DC offset.
• I have replaced the upa68h and it makes no difference.
• The distortion is in both channels (slight variation) not just one.
  
• please see attached pics from my scope. At 1vac the sinewave looks good, and above that starts to clip assymmetrically.
  
• Here is a hint, with NO load, the sinewave looks like a sawtooth, but with an 8 ohm load connected (my speakers), it starts to look like a sinewave.

The service manual is available here: Kenwood Basic M2 Stereo Power Amplifier Manual | HiFi Engine I can't attach it, its too large for this forum.

 

[rayma]

Have you tested the power supply? That should be first.

 

[K-amps]

Thanks, Yes, replaced the PSU sections with spares (Transformers, Bridges, Caps) no difference. Then added power caps onto existing caps, no difference. Inputted directly to input diffs from sig gen, no difference. Core voltages are spot on.

 

[rayma]

Are all the measured power supply and circuit quiescent DC voltages correct, including the input stage DCV
across the Zeners (which are fed by current sources)?

 

[K-amps]

Zener voltage are good. I measured the input diff (upa68) voltage to a good channel and have done variations...

 

[Regenpak]

This is a flakey design. Look at the first page of the manual. If something has to kick in then in your case it prolly doesn't. Your best bet is to check the voltages marked in the schematic. From the bias V(BE) multiplier Q19 through IC1 (that seems to do something magical) to the weird output transistors Q29-35. No idea what those designers have been smoking...


Edit: the fact that it occurs in both channels is an important clue. Look for things that are shared, like some aux voltage that is missing.


Edit2: This procedure requires differential probes. Or scope with math capabilities (Ch1-Ch2).

 

[K-amps]

Vbe voltage is absent.

There are two relays that click in, both work ( I get signal at the outputs)

The design seems ok. These amps have been working for close to 40 yrs. Cant fault it that much. The OP is class G. Dual rails, lightweight. The OP devices are dual BJT’s (2 substrates on single package) custom Sankens.

 

[K-amps]

This is a flakey design.Edit: the fact that it occurs in both channels is an important clue. Look for things that are shared, like some aux voltage that is missing.


Edit2: This procedure requires differential probes. Or scope with math capabilities (Ch1-Ch2).

Exactly, it’s some shared supply... just need to figure out where.

 

[Regenpak]

So you have no bias on both channels? As in base voltages of Q21/23 and Q22/24?

 

[K-amps]

Yes

 

[OldDIY]

Outputs - NLA. These are dual transistors in a single 5-pin package. IIRC, @tarior repaired one of them by creating outputs from two single transistors.
Quite often M2 came with burnt out exits. The M2 has a feature called Sigma Drive. As far as I remember it is a servo feedback system. Many years have passed, so the details are unclear. But if the Sigma Drive disconnects at one end while the amplifier is running, it will pull out a whole bank of output transistors.
The servo works well in terms of sound, but you need to be careful when using it. Make sure the wiring is secure and double or triple check its connections.
Kenwood Basic M2A amp opinions? | Audiokarma Home Audio Stereo Discussion Forums
Connecting sigma cords
file:///home/chronos/u-a99b0deed3b051c0cc81336fc71531c69c5be0b7/MyFiles/Downloads/Kenwood-Basic-M2-Owners-Manual.pdf

 

[K-amps]

The outputs are fine. I have a few of those in stock. Mostly pullouts. New ones ( when they were available) cost $53 a piece.

 

[mjona]

Off the wall, it seems like the speaker load is causing the positive rail to droop at your front end. Perhaps your LTP is in static balance but only to a point and when the tempo increases it runs out of steam and starts operating in an uneven manner.

You have a lot of resistors with triangle warnings in this amplifier. There may not be enough current flow getting to the your front end.

It might pay to measure voltage drops across some of these to check this and to eliminate the possibility of some values having changed.

 

[K-amps]

Interesting observation. They do seem current starved, however this happening to both channels at once is perplexing.... could the use have driven the inputs with a high voltage source and blew something?

 

[OldDIY]

A feature of the M2 is the additional sigma drive feedback. Apart from the usual amplifier.

 

[K-amps]

Indeed. They just loop the feedback resisters to separate binding posts. They can cause issues of misused. I was also wondering if that caused the damage

 

[mjona]

In view of the observation about output stage failure if the Sigma servo disconnects, it is hard to see this failure arising at low power.

I looked at the Service Manual on Hi-Fi Engine by "muff" where the parts list shows a number of Flame Proof resistors in the emitter connections of the output devices themselves, as well as in those of the pre-driver stages.

If the outputs are OK as you have stated these may have survived a fault condition due to fusing of some of these resistors.

 

[U230421]

Exactly, it’s some shared supply... just need to figure out where.

So you have no bias on both channels? As in base voltages of Q21/23 and Q22/24?

Shared are the power stabilizers around Q46 and Q47 for the front end, and the rail smoothers around L1 and L2 for the vas and bias circuits. So check all involved components there thouroughly. As proper bias is missing, I'd check the L1/L2's first.
Less likely are the limiters doing wrong through R55-57 / R56-58 from their own channels, though shared with the protection circuit from Q45.

 

[K-amps]

Never thought the protection circuit could be involved... good idea.

 

[tojoko]

Kenwood basic m2

Hi
Did you ask DRM audio, he knows a lot about these Kenwoods....mine have his upgraded parts is it, since 10 years, still working fine, the original heat-sink with the special transistors worked well for 35 years and probably will still work, I just felt like doing the upgrades, for fun. Still got the original parts, if anybody is interested...PM me.
Cheers, Tojoko.