1. Time for some upgrades in server hardware and software to enhance security and take AK to the next level. Please contribute what you can to sales@audiokarma.org at PayPal.com - Thanks from the AK Team
    Dismiss Notice

Bias problem on a KA-7100, new trimmer does nothing

Discussion in 'Kenwood-Trio/Kensonic-Accuphase' started by tusntuk2, Oct 11, 2018.

  1. tusntuk2

    tusntuk2 AK Subscriber Subscriber

    Messages:
    188
    Location:
    Arkansas, USA
    Can't set Bias on a KA-7100

    Having a problem with an ineffective bias trimmer on a Kenwood KA-7100.

    Got a KA-7100 to restore and after initial DBT test and a little de-oxit I tried setting the trimmers for center voltage and bias before the restoration. The Left channel trimmers would not respond. The amp played, the left side wasn't blown, so I didn't worry about it. I was planning on a full recap, new trimmers, etc. anyway, so I put it in the repair pile.

    Time to rebuild, so I changed all electrolytics and all diodes, put in new transistors for the relay circuit and voltage regulators and the new Bourns trimmers. I measured the old pots and set the new replacements similar. The amp passed every DBT test during rewiring, and the relay would always click on. At full AC power I tried setting the bias and no luck, there was no response at all while turning the Left Bias trimmer. The restoration included changing every transistor in the channel (except the Darlington pair), so I did that, hoping to fix the bias issue with a new driver at Q13. Nope. The center voltage trimmer sets to 0, but the bias pot - nothing. Took it out, tested it, it seems fine. Put it back in, it doesn't change the voltage at the emitter resistors. It measures 0 across the control points. And the Right side works fine.

    Maybe there are clues in these noticeable voltage readings:

    Q13 E -176mv (yes, neg) (s/m= 1.15v)
    C -.89v (s/m= -1.15v)
    B -.68v (yes,neg) (s/m= .6v)

    Q7 E 43v (s/m= 44.5v)
    C 7.47v (s/m= 5.3v)
    B 42.7v (s/m= 44v)

    Q9 E 43v (s/m= 44.5v)
    C -176mv (s/m= 1.15v)
    B 42.9v (s/m= 44v)


    I have checked the outputs, the emitter resistors, the diodes, and other resistors involved in this circuit, even the thermal dot. All of the new left channel transistors were checked, compared, rechecked for orientation, and DBT tested, and remember, the amp plays. Actually sounds great. But I can't set Bias on the left.
    Something's amiss somewhere, but where do I look now?

    Help!?! and thanks...
     

     

    Please register to disable this ad.

  2. hopjohn

    hopjohn Silver Face Subscriber

    Messages:
    1,102
    Location:
    Dallas, TX
    Perhaps you have a bad solder joint at or near one of the test point pins? What voltage do you get when measuring directly across the emiitters (or emitter resistor legs) of the left channel? It should be around 20mV.
     
  3. tusntuk2

    tusntuk2 AK Subscriber Subscriber

    Messages:
    188
    Location:
    Arkansas, USA
    Thanks for the quick reply, hopjohn! I have appreciated your answers in AK. Yes, I checked the resistors at each side (Qe27,29) and over both of them - still 0. Output transistors test OK, and (strangely?) the amp plays and sounds very good, of course I don't want to run it this way. I'm not good at "seeing" where the current goes (or comes from) to power the bias trimmer. I compare voltage readings to right side. I do have a good 'scope, but I haven't learned to use it yet. Maybe this is a good opportunity? I use a decent mm and a cheap but effective transistor tester. The center voltage trimmer seems to work. Every resistor I have tested has been within limits. Maybe there is a bad diode among the newly installed? Still pulling at straws here.
     
  4. hopjohn

    hopjohn Silver Face Subscriber

    Messages:
    1,102
    Location:
    Dallas, TX
    Could be that you have the left trimmer set so low that you're just not seeing a reading. With a multi-turn trimmer you may have to make 4-5 full turns before you have a high enough voltage to see a change. If you have a DBT try adjusting it again to see if that may be the case, remember to turn it well back down before adjusting at full power.
     
  5. tusntuk2

    tusntuk2 AK Subscriber Subscriber

    Messages:
    188
    Location:
    Arkansas, USA
    Good tip about the turns, it was needed on the R center voltage pot. Unfortunately, it does not seem to be the fix for the L bias. You know, if I let my mechanical brain take over I get a suggestion like "it's grounding somewhere". 000.0. That's what the DVM says when I ground the + lead after measuring mV's. My analytical side then says "you can't ground the ouput!" and there you go. I checked and replaced R65-68, per EW's old suggestion, but they all measured good. Same 000. I changed the trimmer again, I stock a few, but the new one shows the same problem, no effect on the bias voltage. What am I missing here? (yes, 500K)

    What path should I follow to measure voltages and compare them to the right side? Signal path? Rails? Anything in the relay circuit?

    Appreciate your helping hand, hopjohn!
     
  6. dr*audio

    dr*audio Fish fingers and custard! Subscriber

    Messages:
    18,772
    Location:
    Cleveland, Ohio
    Have you replaced Q5 and Q9? Have you checked R31 and R27? What are the voltages on Q5? You are missing drive on the positive half of the amp.
     

     

    Please register to disable this ad.

  7. tusntuk2

    tusntuk2 AK Subscriber Subscriber

    Messages:
    188
    Location:
    Arkansas, USA
    Hello dr, thanks for your help. (Nice Ric!) Yes, I have replaced Qe5 ( KSC1845) and Q9 (KSA1220ays.) Both R27 and R31 test right at spec values on s/m.

    Voltages - Qe5 E 20.0v C 42.7v B 20.5v
    Qe9 E 43.3v C -131mv B 42.9v

    I double-triple checked the Q9 collector, yes, it's a negative 131 millivolts. The original transistor had measured 284 millivolts. That and a couple more noted above were reason to change to new transistors.

    Re27 120 ohm 44.9v (rail) / 42.8v
    Re31 1.97K ohm 44.9v (rail) / 43.4v

    An interesting thing happened when I measured the rail voltage. At first both sides saw 50.2v, then after warming a minute the left side dropped to 44.9v. Right side stayed at 50.2v. I guess that's close enough for the relay to open, but I just noticed the drop while testing. I had changed both regulators Qe31,32 with the MJE150xx pair as noted in previous AK posts. I've used them several times with good results. This was the first time I saw the drop to 44.9v.
     
  8. dr*audio

    dr*audio Fish fingers and custard! Subscriber

    Messages:
    18,772
    Location:
    Cleveland, Ohio
    Check the voltage on the cathode of De1. Should be 24V.
     
  9. dr*audio

    dr*audio Fish fingers and custard! Subscriber

    Messages:
    18,772
    Location:
    Cleveland, Ohio
    Also check the voltage on the anode of De3, should be -28V
     
  10. tusntuk2

    tusntuk2 AK Subscriber Subscriber

    Messages:
    188
    Location:
    Arkansas, USA
    De1 = 24.8v cathode; Zener diode 1N5252B confirmed
    De3 = 29.2v anode; Zener diode 1N5255B confirmed

    I'm concerned about variations in what I think is rail voltage, the +/- 47v from the rectifiers. I've noticed it changes a bit as the amp warms up. I have never concentrated on measuring this before, so I don't know that it might be perfectly normal, maybe adjusted by the regulators. This time I measured it as +50.1v and -50.6v. Not a big change, for sure, but I don't know 1) it should be this high (not 47v) and 2) if it should change at all, and 3) if the +/- discrepancy is important.

    I had also already changed Re17, 18, beefing them up to 1w per a suggestion from leesonic.
     
  11. dr*audio

    dr*audio Fish fingers and custard! Subscriber

    Messages:
    18,772
    Location:
    Cleveland, Ohio
    Rail voltages are unregulated. Shouldn't be a problem. What are the voltages on Q3, and check Q5 again at the same time and post them both in case something drifted.
     

     

    Please register to disable this ad.

  12. tusntuk2

    tusntuk2 AK Subscriber Subscriber

    Messages:
    188
    Location:
    Arkansas, USA
    Thank you dr. I studied the s/m some more and see what you mean (unregulated). A better word may have been "unaffected". Anyway - here are voltages for Q3 and Q5 (full AC)

    Qe3 E 20.2v C 42.9v B 20.6v
    Qe5 E 20.2v C 43v B 20.8v

    They all seemed close to specs on the s/m. Then I measured around the power supply, and found +/- 31.1v at the 28v zeners D17,18, the same voltage present at the base of the regulators Q31,32. I believe that's the B+/- for the control board. Now the amp has been on about 10 minutes, and I noticed the main heat sink getting very warm, so I shut it down. The heat did seem to originate on the left side. My testing was all done with no load on the outputs. I'm expecting the heat comes from the L output transistors with no bias (?). I sure don't want to burn the L outputs.

    It's a great puzzle, and I appreciate the help!
     
  13. dr*audio

    dr*audio Fish fingers and custard! Subscriber

    Messages:
    18,772
    Location:
    Cleveland, Ohio
    It seems like you might have a measurement error and you really are getting bias on the left channel. Take a dental pick or exacto knife and clean the leads on the emitter resistors on the left channel, then connect your meter across one of them with the power off and measure it to make sure you are making good contact. You should be using mini clip leads to do this, not the needle type probes. Then put your meter back on volts and make the bias measurement again. If it's heating up there must be current through the output transistors.
     
  14. tusntuk2

    tusntuk2 AK Subscriber Subscriber

    Messages:
    188
    Location:
    Arkansas, USA
    I agree, some current must be going through the outputs because the relay closes and the amp plays, and if left on very long its heating up. I don't have an understanding of how the electricity flows, I'm an analytical mechanic. So I "see" (measure) the function of things, and where they are not working. Fix that. My puzzle is that it does seem to work and I can't find something tangible out of place.

    This is my third 7100, and about the tenth restore I have done. I try to double/triple check everything I do simply because I don't know how it works. Yeah, I'm crazy, but I love the old amps! had my fingers in them since early Fenders in the '60's.

    I have researched lots of old threads about the 7100 and have incorporated many of the tips. EW suggested to change the output drivers if ever the outputs were blown, and to change the v/l circuit transistors too. Although the output was not blown, I changed the drivers Q23-26, all four. The relay driver commonly got weak or failed, so this (pair) was also changed. The bias wasn't working so I changed Q13. I changed all the diodes. Eventually I have changed every transistor in the left channel except the differential input and the main outs. I changed the trimmers and the relay, and all electrolytics.

    I had planned to rebuild this amp for my personal use, and I had stocked the equivalent transistors, as noted in many AK notes. I wanted to see just what this amp could do, so I wanted to change all the transistors. Each time (sub-circuit) I compared it to the right channel - only the R output drivers were new - and powered up with DBT. Each time the relay clicked and bulb dimmed. Then i found differences when comparing voltages, and the L bias was always 0. The right side seems to work fine.

    I will test the emitter resistors, I'm away right now, but I wanted to fill in some blanks about the unit.

    When I got the amp it was described as "wouldn't power up". But it did on a DBT, except there was no LED and no relay click. Of course it "looked" dead. As soon as I replaced the LED the relay clicked and the amp played. A 7100 idiosyncracy, no?

    I next tried to set center and bias voltages, and did some de-oxing. The right side set OK, the left did not, it would not set bias or center offset. The amp played, so I knew the output was not blown, and since I was planning on a full restore I let it go until the repair. With new transistors, electrolytics, relay, trimmers, etc. it should probably work fine.

    I assumed! WRONG!

    Now I'm stuck with a puzzle, and don't know where to look for the next piece. I have noticed voltages differences in the v/l circuit, so that is my next area of focus. (Oh, after changing the L center trimmer I was able to adjust the offset to 0. But the bias trimmer = 0.)

    Thanks again for your kind and patient help, dr and Hopjohn. We will solve the mystery! That's the fun of it, no?

    I'll remeasure the emitter resistors soon.
     
  15. dr*audio

    dr*audio Fish fingers and custard! Subscriber

    Messages:
    18,772
    Location:
    Cleveland, Ohio
    I mentioned that because I have had oxidation on the resistor leads causing no bias reading, especially the combination resistors with short test pins on the top.
     
  16. tusntuk2

    tusntuk2 AK Subscriber Subscriber

    Messages:
    188
    Location:
    Arkansas, USA
    I know you're right, i must sound as exasperated as I am! Sorry, don't mean to, and you are great to help me. I took the resistors out and scraped them, then measured them and compared to another pair I had in another parts amp. When I put on the test leads the values wander from .7 to 1.1 ohms. Both the R emitter res's are the same. The other res's from the parts amp do exactly the same thing - they wander from about .7 to 1.1ohm. None of them measured at .47ohm, and none were stable during the reading.
     

     

    Please register to disable this ad.

  17. tusntuk2

    tusntuk2 AK Subscriber Subscriber

    Messages:
    188
    Location:
    Arkansas, USA
    Reinstalled the emitter resistors, when I measure it (R bias) starts with some millivolts and quickly and steadily drops to 00. Measured the cp pins and measured with direct contact underneath on the new solder points.
     
    Last edited: Oct 14, 2018
  18. tusntuk2

    tusntuk2 AK Subscriber Subscriber

    Messages:
    188
    Location:
    Arkansas, USA
    Odd measurements in Left side Qe 15, Qe21, Qe17 and Qe19. I'll show comparison to Right side, Qe16, Qe22, Qe20, and Qe18.

    Qe15 E 363mv C -10.4mv B Fluctuates +1.xv to -.8v (?) Rapid bouncing around, impossible to read
    Qe16 E 1.22v C -.5mv B .87v

    Qe21 E -10.2mv C Rapid fluctuation (like qe15 B) B -10.5mv
    Qe22 E -1.6mv C .88v B 24mv

    Qe19 E -384mv C -10.8mv B -.362v
    Qe20 E -1.18v C 2.3mv B -.80v

    Qe17 E -10mv C -10mv B -.37v
    Qe18 E 3.5mv C -.79v B -17mv


    Checked pin orientation and numbers against old transistors, schematic and cheap tester. I used the suggested KSA945C for the NPN's and KSA733c for the PNP's. Confirmed the physical layout too.
     
  19. dr*audio

    dr*audio Fish fingers and custard! Subscriber

    Messages:
    18,772
    Location:
    Cleveland, Ohio
    Make sure your DMM is on DC volts. Check the other channel to make sure it is measuring correctly.
     
  20. dr*audio

    dr*audio Fish fingers and custard! Subscriber

    Messages:
    18,772
    Location:
    Cleveland, Ohio
    Do you have an oscilloscope? The amp may be oscillating.
     

Share This Page