• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

KA-7100, ridiculously high DC offset

R

rjsilva

Active Member
Today I recapped the main and front boards of my recently aquired KA-7100 and replaced the relay (I replaced the relay driver last week). It was getting late and knew I wouldn't finish the rear board so I decided to test the unit out. The relay didn't click when I turned the 7100 on, but then it clicked the second time I turned it on. There was a definite "pop" when using the speaker selector knob.

So, I tested the dc offset - it was >2V! I turned the corresponding adjustment pot and it would stop turning before getting below 1V. I checked the bias and it was up to 26mV, so I adjusted that to 20mV and went back to the dc offset and no change. Both channels are equally high.

Any help would be greatly appreciated!
 
Last edited:
Register to hide this ad
Did you replace the trim pots as well? They can and do go bad. If you do this, pull the signal input wires off the power amp board and then try setting the offset. If it still won't go below 1v, there's a problem on the power amp board. If it does, but then goes high when you reconnect the inputs, something on the preamp board(s) is causing it. An electrolytic capacitor installed the wrong way will do the, ask how I know.

Lee.
 
Thanks Lee. I didn't replace the trim pots but I had no problems setting the dc offset prior to recapping. Are the signal input wires the blue and yellow ones?

Last night I double-checked for caps being installed the wrong way on the main board, but I (stupidly) already put the amp back together and wasn't able to see all the caps on the control board. I'll check that out this morning.

Not sure if this matters, but I installed 6 film caps on the control board - the four 3.3uf (with Wima) and the two .33uf tantalums (with Nichicon). I did lift one of the solder pads trying to get out a tantalum (argh), but I think it survived. I should probably test that...
 
Last edited:
Update:

- Caps all seem to be installed correctly on control and power boards. The one lifted solder pad looks and tests ok.
- R76 looks toast, and my multimeter measured it at over 20kohm and continued to climb. I don't know why it got toasted, and I don't think the damage is new, so I'm going to replace it and see what happens (maybe that's the cause of my other intermittent static/no audio issue?).

What's the best way to remove the blue and yellow wires? Do you cut them, unwind them, desolder the terminals from the board, etc.?
 
Last edited:
Also, is anyone able to verify that R56 is soldered to the negative terminal of C22 on the control board? I was sure last night when I was doing it, but in the spirit of being careful I thought I'd double check it is correct (the schematic looks to me like they're connected, but I'm a schematic novice).
 
rjsilva said:
What's the best way to remove the blue and yellow wires? Do you cut them, unwind them, desolder the terminals from the board, etc.?
Click to expand...
Normally don't have to mess with 'em, but if you really need to disconnect them, desolder the terminals. Leave the wirewrap alone.
rjsilva said:
Also, is anyone able to verify that R56 is soldered to the negative terminal of C22 on the control board? I was sure last night when I was doing it, but in the spirit of being careful I thought I'd double check it is correct (the schematic looks to me like they're connected, but I'm a schematic novice).
Click to expand...
One side to Ci22 neg, the other to the collector of Qi8. If that resistor is burned, Qi8 and/or Qi10 is toast.
 
Thanks for verifying the connection EchoWars. To clarify, the toasted resistor was R76 and not R56. I just wanted to verify that I didn't solder C22 to R56 inadvertently.

Here's the latest:

I replaced R76 and recapped the preamp board (except for the one cap underneath the RCA connectors - I didn't have time to do that). Same deal. The DC offset for both channels is about 1.5V. And I still have my no-audio-but-get-static-when-volume-knob-is-up-high issue (that should be another thread, though, I think).

What can I do next? Does the burned R76 mean some other component(s) are likely toast? Though I'm fairly certain R76 was burned prior to recapping.
 
Re76? 1K, 1/4W? If so, that's the current limit resistor for the 28V zener diode. In stock form, Kenwood has that resistor cooking with about 200mW of power. A 1/4W resistor will eventually give up with this power, so replace it with a 1/2W resistor.
 
I did use a 1/4W on R76 for the replacement, but I'll go ahead and replace that with a 1/2W soon.

Though ... what about the high dc offset issue?
 
Offset trim pots are VRe1 and VRe2, should be located close to the dual Nchannel FET (uPA63H)

This Kenwood amp is biased to 40mA. Bias trim pots are VRe3 and VRe4. Adjust for 40mV between the emitters of the power transistors, or 20mV across each 0.47Ohm emitter resistor. Do this adjustment with NO load, volume control to minimum.
Click to expand...

http://mail.audiokarma.org/forums/showthread.php?t=314642
 
Last edited by a moderator:
rjsilva said:
Though ... what about the high dc offset issue?
Click to expand...
The input differential of this amp is a dual FET. Unusual for these to have problems.

The preamp is cap-coupled to the amp, so there should be no DC being through to the power amp. Just for shits and giggles, hit the Subsonic Filter and/or Tone Bypass. This changes the coupling config a bit. Then check the offset again.

Both channels having the same offset issues tends to point toward a grounding problem or power supply issues. The input stage is grounded solely through the wire that runs from Pin (1,3, or 5) of the preamp board to Pin (3, 4, or 5) of the amp board. With power on, you can measure voltage from, say, Pin 5 of the power amp board to Pin 9 (between the two signal wires) at the top of the preamp board. It should be 0.00V. If it is anything else, you have a grounding problem and need to ferret it out.
 
The latest:

1. This morning the subsonic filter and tone bypass both changed the DC offset from 1.3V to -1.8V (approx.).

2. From pin 5 (black on amp board) to pin 9 (between yellow and blue on control board) I am reading 4.4mVDC and 1.7mVAC. If this matters, the relay decided to start switching off immediately after switching on, so it was off when I took these readings.

Are those readings high enough to warrant a potential grounding issue? I'll go ahead and look around for broken/chaffed/etc. wires and connections, but is there something else I should be looking for?

If it's not a grounding issue and instead a power supply issue, what am I looking for?
 
More:

I tried setting the DC offset with the subsonic filter on and could easily get it down to a few millivolts. Not sure what that means...but for someone more knowledgeable than me, I assume this can help narrow down the issue.
 
Last edited:
rjsilva said:
1. This morning the subsonic filter and tone bypass both changed the DC offset from 1.3V to -1.8V (approx.).
Click to expand...
We need logical troubleshooting! Try one, then the other, then both. Document the results of each.

The sub-5mV reading for the grounding seems to indicate ground is fine.
 
rjsilva said:
More:

I tried setting the DC offset with the subsonic filter on and could easily get it down to a few millivolts. Not sure what that means...but for someone more knowledgeable than me, I assume this can help narrow down the issue.
Click to expand...

The subsonic filter is a 0.33uF tantalum cap, unless you replaced it with something else. By switching it on you are blocking DC from upstream in the pre-amp. I bet if you remove the blue and yellow wires from the power amp board, you will be able to set the bias and offset correctly. There is enough wire to cut and re-wrap them, or you can remove the pins as Glen (echowars) said.

I would go back to the preamp boards and check for solder bridges or caps installed the wrong way. It could also be a bad transistor. Most, if not all of these small signal transistors in this amp can be replaced with 992 or 1845s.

Lee.
 
Thanks guys for the replies!

EchoWars said:
We need logical troubleshooting! Try one, then the other, then both. Document the results of each.
Click to expand...

Here are averaged readings (very wobbly). As I previously mentioned, I adjusted the dc offset with the subsonic filter on.

Subsonic - R: 2mV, L: 9.5mV
Tone defeat - R: 75mV, L: 80mV
Both - R: 4mV, L: 7.5mV

The relay clicked off as soon as I switched off subsonic and tone defeat, but I presume the dc offset would have measured somewhere in the ballpark of 3V or higher (based on the "normal" 1.2V-1.5V when the trimmer pots are at their lowest plus the 1.8V adjustment I did with the subsonic filter switched on).

leesonic said:
The subsonic filter is a 0.33uF tantalum cap, unless you replaced it with something else.
Click to expand...

I replaced them with 50V 0.33uf Nichicon film caps. I also used 50V 3.3uf Wima film caps on the four 25V 3.3uf electrolytics.

I'm assuming the problem is something I did since it only started after I recapped it. I'll check again for solder bridges and caps installed incorrectly...
 
I was hoping for:
Subsonic on: V=?
Subsonic off: V=?

etc...

But if the relay is kicking in, then you'd have to measure somewhere other than the speaker relays. And, BTW, there's nothing keeping you from measuring the DC being sent to the amp...right there at those terminals at the top of the preamp board. Try this with the filters off and the tone on.

I might mention that, because of the large DC being sent to the amp, make switch changes with the power OFF. No reason to complicate matters by toasting the amp.

In essence, it looks like DC is being fed to the amp from the tone circuit. This means:
1. Ci23 and Ci24 are installed backwards.
2. Both the Qi7/Qi9 and Qi8/Qi10 display the same fault that is causing Ci23 and Ci24 to be reverse biased.
3. One of the power supplies to this whole section is bOrked, causing the output stage to shift its quiescent operating point and reverse-biasing the output capacitor. The board receives +/-28V coming from the regulated supply on the main board, and this is dropped further with a couple of resistors (Ri65 and Ri66) and a couple of WZ182 (18.2V) zener diodes. Here's where the failure probably is (as long as you installed Ci23 and Ci24 properly, and you should verify that this is not the case with a meter): either the 28V supply on the main board (which I think you fixed) or the zener regulator on the control board (resistors or the zener or burnt traces or broken wire or something shorted that's pulling the voltage down).
 
Thanks Echowars for continuing to help! I'll get the measurements tomorrow.

Regarding Ci23/24 - I replaced those with these Wima film caps (also Ci19/20):

http://www.mouser.com/ProductDetail...ULWPdT5TCakaZYaOjv2LT6e2va%2bpOh28XDLXjc8zLm2

Perhaps the circuit wants a polarised cap? Are you able to verify? It seems unlikely to me that the other potential problems you brought up (Qi7/8/9/10, etc.) surfaced coincidentally when I recapped the board.

I have some polarised caps at that rating so maybe I'll stick them in - a bit of a waste of money for those Wima's if it fixes it!
 
You must log in or register to reply here.
Back
Top Bottom