Harman Kardon HK 930 popping noise troubleshooting

[dbxdx5]

The 930 is known for a loud pop after turn on, however, the particular unit on my bench keeps on popping for a few minutes after.

I first pulled the Main-In/Pre-Out jumpers and fed a signal directly into the amp. Other than the typical HK initial pop, there were no further eruptions. So the amp section would seem to be ok.

With the function switch set to Aux 1, I measured the voltage at the output (R) of the Tone Control PCB while also measuring it at the speaker output (R). I chose Aux 1 because this bypasses the Preamp board, making the Tone Control board the only board between the RCA input and the Preamp Out. Here's what I measured:

Initial Turn On

Tone Control: quickly fluctuating voltage as high as 1.4V DC.
Speaker: quickly fluctuating voltage as high as 19V DC.

First Three Minutes
Tone Control: voltage settling down to between 600mV and 700mV
Speaker: voltage still going as high as 2 - 3V, but beginning to drop to 150mV in tandem with the output of the Tone Control board.

After 10 Minutes
Tone Control: voltage fairly steady at 465mV
Speaker: voltage staying at around 115mV (still high for DC offset, so I'll likely replace the differential pairs to see if I can lower it)

Power Down
Speaker: spike of 16V or so, gradually dissipating.

I'm thinking there's either a leaky cap or a bad transistor on the Tone Control board. Maybe more likely a transistor since the issue seems to disappear after warm up. Unfortunately, the service manual doesn't include any reference voltages to measure against, which is how I would usually track down the offending component. While I can hit each transistor with a quick shot of freeze spray, I'd like to hear how others would suggest narrowing down the possibilities.

 

[Osvaldo de Banfield]

It seems to me the amp is conditionally stable, say, at the border of oscillation. Try a recap as a first tentative.

 

[sgrossklass]

Probably it's C515 and C565 in particular that are trying to compete with a proverbial sieve. A recap of the tone board as suggested is probably in order. Stock electrolytic voltage ratings seem a bit skimpy - I would go with whatever modern capacitor (e.g. Panasonic FC) is the same capacitance and physical size, which is likely to get you 2 or 3 steps higher in voltage.

 

[rayma]

It seems to me the amp is conditionally stable, say, at the border of oscillation.
Try a recap as a first tentative.

With this age of amp, need for a recap is expected anyway.

 

[wiseoldtech]

The complaint isn't clear enough.
The OP did measurements on the Right channel.
But....
Does this popping only come from the right channel?
Please clarify if it also comes from the Left as well.


Otherwise, we've no real basis to troubleshoot things in a proper way other than speculation and endless posts.

 

[dbxdx5]

It seems to me the amp is conditionally stable, say, at the border of oscillation. Try a recap as a first tentative.

Probably it's C515 and C565 in particular that are trying to compete with a proverbial sieve. A recap of the tone board as suggested is probably in order. Stock electrolytic voltage ratings seem a bit skimpy - I would go with whatever modern capacitor (e.g. Panasonic FC) is the same capacitance and physical size, which is likely to get you 2 or 3 steps higher in voltage.

With this age of amp, need for a recap is expected anyway.

Agreed, a recap of the tone board makes sense.


And the wiseoldtech is living up to his name (the wise part, not the old) : I had thought it was just the right channel, but I checked the left, and the symptoms are there too.

 

[rayma]

The old HK receivers of this era have problems with the preamp sections due to caps, switches.
The tuner and power amp sections often still work great, though.

 

[wiseoldtech]

Agreed, a recap of the tone board makes sense.


And the wiseoldtech is living up to his name (the wise part, not the old) : I had thought it was just the right channel, but I checked the left, and the symptoms are there too.

Oh, I'm old, for sure (66).
And since the problem is coming through BOTH channels, the issue is with something common (power supply?) that concerns both channels.
The left channel power supply also feeds both channels of preamplification. (see service manual)
That common supply (12V) should be investigated. (TR905/906 and associated components)

 

[dbxdx5]

So coming out of the power board at pin 3 and into the preamp and tone boards through B3 is 20V. I believe that's about right, no?



However, I am not seeing 12V at B6, but rather ~1V.



TR904
E: 20.7V
B: 21.2V
C: 40.2V


TR905
E: 20.1V

B: 20.7V

C: 40.4V

 

[latala]

I have had problems with the small signal transistors in lots of these the early plastic types
were prone to go noisey with popping sounds I would blanket change them all as they will all eventually fail
I have a large hk collection loved em when they were current
Trev

 

[wiseoldtech]

So coming out of the power board at pin 3 and into the preamp and tone boards through B3 is 20V. I believe that's about right, no?



However, I am not seeing 12V at B6, but rather ~1V.



TR904
E: 20.7V
B: 21.2V
C: 40.2V


TR905
E: 20.1V

B: 20.7V

C: 40.4V

Your input of 40V to the collectors is right, it's fed by the L ch power supply.
But the zener diodes D901 and D902 clamp the voltage down to 12V for B3 and B5 - B6/A/F feeds to the rest of the circuitry.
So something's "wrong" along there.
Either the diodes are open/defective, or the regulating transistors TR904/905 are defective.

 

[dbxdx5]

You're right. I checked both diodes, and D902 is open. I need to see if I have a suitable sub.

 

[wiseoldtech]

You're right. I checked both diodes, and D902 is open. I need to see if I have a suitable sub.

You must mean D901, which biases the base voltage of the regulator transistor.
Because if you had 20 volts on the base, it's defective.
I hope you're using the same service manual that I have.

 

[dbxdx5]

I checked D901, and it was fine. D902 is the 12V zener that's connected in parallel with the .01uF C911, and it's bad.
D901 is a 21V zener. Or at least that's what I'm seeing in the service manual.

 

[wiseoldtech]

I checked D901, and it was fine. D902 is the 12V zener that's connected in parallel with the .01uF C911, and it's bad.
D901 is a 21V zener. Or at least that's what I'm seeing in the service manual.

That is correct, my mistake, the print isn't clear.
HK schematics are sometimes blurry hand-drawn slop messes.

 

[dbxdx5]

I would definitely call this one a blurry, hand-drawn slop mess.

I've replaced D902, and now the wildly fluctuating DC voltages at the speaker outputs are gone. I still need to do more testing, but looks promising. Thanks wise one.

 

[wiseoldtech]

I would definitely call this one a blurry, hand-drawn slop mess.

I've replaced D902, and now the wildly fluctuating DC voltages at the speaker outputs are gone. I still need to do more testing, but looks promising. Thanks wise one.

You're welcome.
After decades of having serviced thousands of video/audio products, I can almost troubleshoot them blindfolded by now.
HAHAHAHA!


Don't mind my humor, I'm just permanantly stoned from all those years of soldering iron fumes.

 

[sgrossklass]

Ah, the bad old days, when having decent PSRR on your amplifier circuits was being considered "nice to have" rather than essential. Otherwise this issue would not nearly have been as visible on the output. I can only presume that D902 was intermittently shorting out, overloading the regulator.

The preamp circuitry in this HK (particularly the 2-transistor npn "opamp") is similar to the Sansui AU-5900's. What I learned from that one:
1. Regulator noise is critical. HK eliminated most of zener noise "brute-force" with a 1000µ in parallel, but zeners have made a lot of progress since then. If you can find a matching low-noise type, use that for D901. Also, C909 may be worth swapping out for a lowish-ESR type (e.g. Panasonic FC). I wouldn't overdo it with the low ESR types either as I'm not convinced about long-term life expectancy with these... basically, for that you want low leakage, which is just about fundamentally opposed from low ESR.
2. Preamp PSRR can be increased substantially by splitting part off from the 68k collector resistors for some RC filtering, similar to what the phono section already does. Maybe +B3 --> 6k8, 22µ to ground*, 62k (or 12k - 10µ - 56k), or if you want to combine two RCs for left and right, use one 3k3 and 47µ (or 6k2 - 22µ) and then branch off with both 62ks (56ks) there. Use metal film Rs for the 62k/56ks in particular. I wouldn't go any lower than 56k, at some point it shows up in loop gain and hence distortion performance. I would prefer not combining RC filtering for all 4 stages for fear of instability, though even that may not be an actual problem.
3. (*) Grounding can be dodgy in these, as in not very star-like at all, and you should be careful where you connect your extra filter caps to avoid injecting power supply noise into your ground. (The lower the Rs in your RCs, the more critical it is. Guess that's why they only have 22k in the phono section.) I'd look for the spot on the board that promises lowest ground return impedance to the power supply - which on the tone board might be towards the output side. If needed, an extra power ground return wire runnning alongside +B3 all the way back to the power supply may help out.

 

[dbxdx5]

I spoke too soon. While the extreme voltage fluctuations seem to have settled a bit, I'm still getting lower level pops, on the order of a 1-2V DC, both channels.

Backing up to where we were before, what voltage should the tone and preamp boards be seeing at B3?

 

[wiseoldtech]

I spoke too soon. While the extreme voltage fluctuations seem to have settled a bit, I'm still getting lower level pops, on the order of a 1-2V DC, both channels.

Backing up to where we were before, what voltage should the tone and preamp boards be seeing at B3?

Intermittent issues such as that require focused troubleshooting, requiring an O'scope, signal tracer, dvm's.
Since the issue is common to both channels, as previously mentioned, the power supply is suspect, and should be throughly inspected.
Caps, resistors, even the transistors in that area could be the cause.