HarmAn Kardon 930 Bias Adjustment Help

[mcaddie7]

I want to preface this with I'm a pretty big nob and have never touched a transistor or biased anything. I have however recapped my Harmon Kardon 930 to what seems great success. There are several guides on this site for biasing this receiver but non include pictures and thus I really can't follow them. Some say have a load on the speaker terminals, some say don't (capacitive coupled?). I'd like to get a definitive guide for this receiver.

So two things:

1) I checked the DC offset on each speaker terminal and go about 120mV on each one. Is this good, is this bad, or should there be some load attached (and if so how) and how to adjust this?

2) for biasing the Output the manual states the following:

The test points are on the schematic here:


My question is, where the hell is TP1 on the actual board? Is it the yellow and purple wire that seem to serve no purpose? When adjusting this should there be a load or not?


Any help would be greatly appreciated for a begininer such as myself, as the other guides I just can't follow and want to be a sure (plus it should help future noobs when fixing this receiver).

Thanks so much!

 

[jheu02]

I will preface this with I don’t have a 930, and did not stay at a Holiday Inn Express last night. The instructions show that test point 1 equates to R 617. Looking at the scat, I circled in green the two resistors which equate to the test points for each channel. Scat also shows two circles labeled TP1 and TP2. Since the board isn’t labeled, and I can’t see the trace side, I can’t be positive, but one of the two in each pair of cement resistors the green arrows point to will be the correct resistor for each channel respectively. There are also two pins sticking out from the board, which I circled. Again, can’t see the traces and if they lead to one of those cement resistors, but those pins MIGHT be the test points shown in the scat, one pair for each channel. The bias trimmers are circled in yellow.

120mV DC Offset is pretty high, but not to the damaging your speakers level. That's mainly due to the differential pair of transistors or other component in that portion of the circuit. I believe the diff pair will be what I circled in blue, but others more familiar with the 930 will likely chime in.

 

[Warren-Oates]

There should be no DC offset on a cap coupled design, unless the coupling caps have failed. You will get an erroneous reading of DC offset if you tes with no speakers hooked up. If you test the DC offset with some junk speakers hooked up it should be about zero.

Test Point one is shown on your diagram, it is the leg of the emitter resistor R617, 5 Watt 1/2 ohm. To attach there you will need a mini J hook, be careful not to short something when you test, or you can destroy parts.

I would watch some videos on testing bias voltage across emitter resistors so you are more familiar with what you are attempting.

 

[dlucy]

There should be no DC offset on a cap coupled design, unless the coupling caps have failed. You will get an erroneous reading of DC offset if you tes with no speakers hooked up. If you test the DC offset with some junk speakers hooked up it should be about zero.

Test Point one is shown on your diagram, it is the leg of the emitter resistor R617, 5 Watt 1/2 ohm. To attach there you will need a mini J hook, be careful not to short something when you test, or you can destroy parts.

I would watch some videos on testing bias voltage across emitter resistors so you are more familiar with what you are attempting.

The Harman Kardon twin power series, e.g. 430 630 730 and 930, are direct-coupled instead of cap-coupled (like the prior everything). These twin-power-transformer receivers are more powerful, more modern, but also more neutral-sounding as there are no capacitors immediately before the speakers imparting some tonal or sonic changes. Not better or worse; just different.

 

[Warren-Oates]

Ah, I just assumed they were old enough to be cap coupled. To the original question then 120MV is quite high. If there is no adjustment for it then likely the differential pair transistors have drifted too far apart and will need replacement.

 

[dlucy]

So, several things:

1. Disconnect your speakers. No reason to risk them, no reason to keep them connected.
2. Use a 10 watt or higher 8 ohm resistor instead of a speaker. Connect each of these load resistors to the speaker terminals as if they were the speakers.
3. As @jheu02 points out above, attach your multimeter to each side of R617. If the two sticking-up metal rods at TP1 do, indeed, connect directly to each side of R617 (you can easily test with your multimeter's continuity setting or 0 ohms if you don't have that setting), set your multimeter to volts DC
4. Follow step #4 to get as close to 15mV as you can. If you can't get there exactly, that's OK

 

[dlucy]

Ah, I just assumed they were old enough to be cap coupled. To the original question then 120MV is quite high. If there is no adjustment for it then likely the differential pair transistors have drifted too far apart and will need replacement.

Agreed.

If you can't get the mV down into the 50mV range, then you may need to either:

• replace some components (hard to do)

or

• increase the bias resistance by adding a resistor to the trimpot (thereby increasing the range of resistance it can apply)

 

[Warren-Oates]

For convenience you can use JUNK speakers as your dummy load, by junk ones I mean that still measure about 8 ohms but you don't care about them like they have no real value.

 

[Warren-Oates]

The 120mv reference was DC Offset, not bias, if I read correct in post 1.

 

[jheu02]

The bias should have no effect on the dc offset, so adding a resistor to the bias pot will mess things up. He’s got two issues Doug…

120mV dc offset
and not knowing where to connect to set the bias

That’s why I tried showing pictorially what we were talking about, but not owning a 930, I was making an inference about the two pins and which of the two in each pair of emitter resistors was the correct one to use for the test point.

 

[dlucy]

The bias should have no effect on the dc offset, so adding a resistor to the bias post will mess things up. He’s got two issues Doug…

120mV dc offset
and not knowing where to connect to set the bias

That’s why I tried showing pictorially what we were talking about, but not owning a 930, I was making an inference about the two pins and which of the two in each pair of emitter resistors was the correct one to use for the test point.

you guys are pushing my OCD hard... I'm >this< close to buying a 930 just to figure it out....

 

[Goldie99]

This might help - it's page 11 of the SM....

 

[Warren-Oates]

Two R616? I think HK is known to have a lot of typos in their manuals during this era. It appears the first R616 is indeed R617.

 

[jheu02]

Two R616? I think HK is known to have a lot of typos in their manuals during this era. It appears the first R616 is indeed R617.

And if you compare the above sm amp board to the earlier posted scat, the board shows TP 1 and 2 on the same channel and TP 3, 4 on the other channel. The scat clearly shows TP 2 is for the R channel, TP1 for the L channel. However, IF the traces as depicted are correct, the lower pin on each channel connects to the lower lead of R617, 637, and the upper pin is tied to the large pads that the other leads of each resistor tie to. So, you would attach your DMM leads to the two pins and adjust the corresponding trimmer to set the bias for each channel.

 

[Mumbles]

These are the pins you want to clip onto for setting bias. If some of the components are in the way, they can also be accessed from the other side of the board.

Note this board is from a 630 Twin, which is very similar to your 930 board.

 

[mcaddie7]

you guys are pushing my OCD hard... I'm >this< close to buying a 930 just to figure it out....

Okay so if I’m reading all these posts correctly, my DC offset I shouldn’t worry about?

And my bias I’ll connect to tp1 and tp2 on the board and adjust each variable resistor while having an 8 ohm resistor in parallel with each speaker?

(I did put an 8ohm resistor on each speaker terminal but they got really hot so I removed them quickly)

Thanks for any help!

 

[Mumbles]

but they got really hot

You might have had the volume turned up with selector set to FM or phono?

To set bias, no input signal or speakers used, just two 8 ohm 50 watt minimum resistors hooked to one channel in place of speakers. The other speaker channel remains turned off. Selector set to AUX. Stereo switch set to stereo. Bass, treble, and balance set midway with volume turned right down. The resistors should remain cool.

The resistors will get hot though when using a scope to set DC balance on units which are adjustable as the amp is being driven into clipping, at or near max volume.

What kind of resistors are you using? I find the cheap wire wounds found on Amazon are OK for occasional use.

 

[mcaddie7]

You might have had the volume turned up with selector set to FM or phono?

To set bias, no input signal or speakers used, just two 8 ohm 50 watt minimum resistors hooked to one channel in place of speakers. The other speaker channel remains turned off. Selector set to AUX. Stereo switch set to stereo. Bass, treble, and balance set midway with volume turned right down. The resistors should remain cool.

The resistors will get hot though when using a scope to set DC balance on units which are adjustable as the amp is being driven into clipping, at or near max volume.

What kind of resistors are you using? I find the cheap wire wounds found on Amazon are OK for occasional use.

View attachment 3181732

That'll do it, I wasn't using high power resistors, just 1/4 W. Woops. I'll order some heavier duty ones. Okay, and in terms of DC bias, I'm guessing I can't really adjust that on this model but am hoping once everything else is being adjusted that the DC offset is low? If not guessing that means I need to check back in about possibly new transistors?

 

[jheu02]

You're confusing some terms, and unfortunately, it does matter. I hope the following explanation/analogiy helps you understand it better…

DC Offset is the 120mV you’re seeing at the speaker terminals. It's basically the difference in work being done by the transistors in the circuit. When you have DC offset, one side of the circuit is working harder than the other and thus not balanced with the other side. That difference “gets through” to the speakers. IDEALLY it should be 0mV….perfectly balanced, and thus the sides cancel each other sort of like a math equation. So no, your DC offset of 120mV is not good. It will probably not kill your speakers, but will likely present itself as distortion to the sound. Speakers are driven with AC signal, not DC. High enough DC and you can set your speakers on fire.

Bias current that we're talking about here is related to the output transistors. When setting it, it's similar to having your car started, but at idle…it's ready to go, but not doing anything yet. You are measuring it in DC, as the components run on DC and pass an AC music signal. The bias is what you CAN adjust on this model, hooking your meter to the two pins and trying to get it to read 15mV or close.

 

[mcaddie7]

So thank you all for the great knowledge, I hope this helps someone else. But I ran into an issue.....

Bought some 8ohm, 50W resistors and hooked them up to the speaker A terminals. With volume down, knob on Aux, and only the Speaker A selection switch activate I was able to use the two test pins on the board. The right side was at 17mV so I turned the pot to until 14mV was had. Yay!

However once I tried the left side, The potentiometer didn't seem to change much. It was mostly 0mV but it did have a small spot that let it be adjusted up to 3mV where it is currently at. For reference the receiver sounds beautiful, and the channel isn't dead or really noisy. But I did think it was kind of quiet.

Is this a transistor issue / short issue from when I was working on this board? (again it sounds great, just lacking a little bass if anything) Should I replace those two potentiometers with something more modern (if so what?).

Thanks for the guidance!