[Novice Project] Kenwood KR-9940: '70s TOTL Quad Stereo Receiver

[guy232]

After finishing my past project here (tek 2213a) I decided to start documenting my next one.

A KR-9940, quad receiver in excellent condition it seems. I'm a novice but have most/all of the tools needed. I want to do the opposite to everything an audiowanker would do, so before I hook it up to give it a listen I think I should inspect it and gather numbers. Maybe this won't be a long project if everything checks out?



The interior was quite clean, hooked up to DBT w/ 150w bulb, powered it up, bulb dimmed as it should.

Next things to do will be to check DC offset, PS voltages & bias.

The only thing i'm not 100% on yet is the bias. I believe this is a quasi-complementary NPN setup and I think that I have to do the steps outlined below, still looking into whether or not I can find more accessible points to connect my leads to.

Service .pdf attached to this post.
Pics w/ bias excerpts & overview shots of the heatsink:
http://imgur.com/a/8aSgtXK

• Set amp to idle/no load
• connect dmm pos clip to emitter lug of output transistor Q1/Q3/Q5/Q7. I have some long IC clip leads that I can reach the lugs down in the heatsink with.
• connect dmm neg clip to the collector of amp Q2/Q4/Q6/Q8. Since the collector is the case, I can't clip onto that. Am I right that I could instead clip to thruR" (red) wire connects to?
• Connect leads across each emitter resistor, will require moving some wires and pulling the resistors into position to clip hooks onto
• adjust corresponding VRe1~4 to about 15-20mV
• let idle 30 minutes, check the 4 bias voltages again and adjust to 25mV

Will update with more pics & data once I do a bit more reading.

Any insight, comments, or notes pertaining to anything about the KR-9940 are definitely welcome!

[mzacharias]

Since you have already powered it up without problems, as part of the initial checkout I would recommend just putting on some load resistors and looking at test tones on a scope. Doesn't hurt to assess the condition of the controls, switches, relays etc and points to possible problem areas to address later.

(I'm retired and don't have that much to do, so one thing I do is watch Youtube repair videos. I see so many who hook up a couple of junk bench speakers as if that will tell them how a unit is really performing.)

[guy232]

Cheers, good idea on using the scope, will be breaking out the dummy load and scope next to do the final checks like you mention.

I figured out what my issue was, I think I was being too delicate when moving the emitter resistors. I thought that bending wires/components would likely break them, looks like I was wrong about that. With a little pulling and bending the 4 emitter resistors were positioned and clipped onto. Idle bias is now set at 25mV.

Uploaded all the hires pics to imgur if anyone finds it helpful in the future https://imgur.com/a/3ID7n1P

[mzacharias]

Remember to hook the scope ground lead(s) directly to chassis when checking output across a load. The 9940 is bridgeable and if the 2/4 switch is in the wrong position an amp channel could be damaged if the scope is hooked up wrong.

Sorry if this is just stating the obvious.

[guy232]

Definitely appreciate the info! Wasn't sure about that but was going to be looking into it heavily before I did any connections due to what i've earlier read, some or which has been conflicting.

Will do some more reading into bridged style amp that you mention.

Not sure if it this effects anything and my terminology may be incorrect: I read that the 9940 supposedly doesn't combine all 4 amps if you have it set to 2 ch mode, only is 50w channel, but the lower model (8840/8040 supposedly "straps" 2 amps per channel together in 2ch mode, similar early version model 7340/8340 does as well I think, maybe 9340 does too)

I've not tried using the dummy load i've built yet but I believe it will work for 2 channel @ 8ohm non-precision/non-max power testing when the scope is properly hooked up. Resistors are cheap ones from amazon, 8ohm 100w wirewound w/ Al body.

[guy232]

Note: I think I may type too much, will try to cut it down. Tried designing my pics in hopes of relaying more info with less paragraphs. Theyre also hi-res so can zoom in if needed. If theyre confusing/useless feel free to let me know.

Setting aside the scope until done checking voltages. With the unit powered on I checked all of the voltages on power supply board X00-1590-10.

Voltages listed below, main things that stand out are:
#14 &16: rapid changing mV from approx -23mV to +12mV. Also #17 & 19: I believe B- B+ are possibly incorrect.

1: 50v... 56V
2. GND
3: 14V... 15.4V
4: 24V... 23V
5. GND
6. 10V... 9V
7. 10v... 9V
8. gnd
9. 24v... 23.7V
10. 14v.... 15.3V
11.gnd
12. 40v... 51.5V
13. 30v... 35.9V
(?) 14. UNSTABLE, rapid fluctuation +3mV to -16mV
15. GND
(?) 16. UNSTABLE rapid fluctuation ~+20mV to -28.3 mV
17. +b... 10.8V
18. gnd
19. -b... 11.3V

I also inspected the board, the transistors mounted to small heatsinks all measured at 60-70°C after idling for 20min minutes.

There was a bit of a dark area and what appears to be a slightly yellowed resistor and some smaller resistors covered by cooked brown glue. It looks pretty bad but there is no corrosion and all of the resistors measure the proper resistance.

I checked the small Protection & PS board, it's heatsinked transistor measured at 60°C after idling 20min. There was some browning of the board by Q1 but no other visual clues.




I believe I have gathered all of the numbers I need in order to continue troubleshooting. Now I need to do some research to find out exactly what to do with that data. Will update.

Any advice or critiques always appreciated.

[mzacharias]

Points 12 and 13 being high suggest regulator transistor QK1 may be shorted. IF this is the case. other associated parts might also be bad, for example QK2. In fact, a defective QK2 could alone be bad and produce this symptom.

[guy232]

Much appreciated! Will go ahead and remove them and test them. Will also read up on those to better understand why they cause the higher voltage. Will report back.

[guy232]

Ok, removed Qk1 and Qk2

Qk1 (2sc789) reads a short on Collector to emitter, meaning it is failed I believe.

Qk2 2sc1000(BL) passes all multimeter checks.

I see in the manual that Qk1 should be a 2sc789 O or R. I notice the one in my unit is a 2sc789 Y pictured in previous post. I wonder if it is factory, I notice another 2sc789 on the relay board is identical to this one.

I've searched to find a proper replacement may be a TIP31B. Very new to transistors so going to research this a bit more.

[mzacharias]

In practice there would be quite a few transistor subs for the 789. As long as the package style is convenient to use, and the basic specs such as VCBO, collector current, power dissipation etc are OK, you're good to go.

Replace the smaller Darlington driver as well.

Good work!

[guy232]

Cheers, wasn't sure what a darlington driver was but based on some google it seems you mean Qk2 (based on how they are arranged) Will do.

Do you happen to know if this type of PS topography has a common name so that I can research and figure out why everything is where it is? Or is this just something that requires me to dig into a text book for awhile to understand? Was hoping this service manual would be like the one for my tektronix scope and have in-depth theory of operation for all PWB but alas it does not.

Many thanks for your insight, I have actually learned quite a bit here! Will definitely keep things up to date just need to place a couple orders in case the local shop doesn't have a suitable replacement.

[mzacharias]

I'm not an engineer, but this is a simple series regulator circuit, and a Darlington configuration is very common.

[guy232]

Just updating, voltages are closer to spec now thanks again! Just spending some time looking at the schematic to let in sink in. Wondering if the original Q1 died a natural death or if something may have caused it. I read that the ones used in the PS are often pushed hard.

Replaced:
Q1: 2SC789 > TIP31B
Q2-3: 2SC1000 > KSC1845
Q4: 2SC1419 > MJE15032G
Q5: 2SA755 > MJE15033G

pin 12 (40 V) was 51.5 and is now 42.5 V
pin 13 (30 V) was 35.9  and is now 29.6 V

Q1 heatsink still rapidly heats to 70C just like before, maybe this is normal.



All other voltages are the same as before:

   1: 50v... 56V
3: 14V... 15.4V
4: 24V... 23V
6. 10V... 9V
7. 10v... 9V
9. 24v... 23.7V
10. 14v.... 15.3V
12. 40v... 42.5 V
13. 30v... 29.6 V
(?) 14. UNSTABLE, rapid fluctuation +3mV to -16mV
(?) 16. UNSTABLE rapid fluctuation ~+20mV to -28.3 mV
17. +b... 10.8V
19. -b... 11.3V

[mzacharias]

I think so far as the voltages are concerned - you're good. You can see from the browning of the board that Q1 normally runs hot. It might be possible to arrange for better heat sinking (I'm not above cheating a bit myself).