KLH Model 20 recap project..germanium what??

darkhosis

Active Member
Hey there! I'm finally caught up and ready to start restoring this TT system that belongs to my neighbor. Didnt quite know where this thread belongs but I have finished fixing up the turntable and I got the speakers handled.

This KLH amp is something older and different - germanium transistors. So I'm trying to make sure I do this right: capacitor polarity goes backwards??

Also, I ordered all the electrolytics I might need but I may have made a mistake on this one for the main power cap?

https://www.mouser.com/ProductDetail/661-EGPD500E302MM35H
 
I don't think the type of transistor has much to do with cap polarity, but if they designed it with a negative voltage supply they'd go "backwards". Basically just match it to the existing setup and it should be fine.

whats the original cap, and why do you think the one you picked won't work?
 
I don't think the type of transistor has much to do with cap polarity, but if they designed it with a negative voltage supply they'd go "backwards". Basically just match it to the existing setup and it should be fine.

whats the original cap, and why do you think the one you picked won't work?
3000uf 40V. The value of the caps are the same I just upped the voltage rating to 50V and I do think it will work. It's a cap designed for new car electronic systems - "high ripple" tolerance of up to 6A. I picked it because it was high temp. Too industrial, maybe?

Do they still design negative voltage supply systems? If I hadn't read about it here, I wouldn't have known. I'm just getting into recognizing the varieties of design, something I'm going to learn more about in class. When I'm recap-ing boards I try to double or triple check values and placement on the board but negative voltage is a new one for me.
 
First step, rebuilding the main filter cap section. The substructure comes off the wooden frame OK but the knobs won't budge so the plinth stays for now.

In this case, someone had replaced the 2000uf capacitor with a CDE axial that reminds me of ones from the 80's I've seen. When you take off the carboard cover and disconnect the 9 pin connector it becomes more accessible. The question is how to make adjustments for the different gauge of the newer caps and how they connect to the PCB. I would prefer not to totally rewired the PCB it looks haggard.
 

Attachments

  • IMG_7105.JPG
    IMG_7105.JPG
    60 KB · Views: 82
  • IMG_7106.JPG
    IMG_7106.JPG
    57.8 KB · Views: 74
Last edited:
Finished replacing the caps in the power supply and the power amp boards. I need to learn to take measurements pre-recap, The power supply looks like it's outputting at -35v DC but when I get to the power amp one channel is a good 2 volts off from the other one. I can't figure out how the trim pots work from the schematic.

And I'm getting -180mV DC on the left and right channel

Also, I just discovered the output transistors were changed to Motorola SJ3464 not RCA -
 
Last edited:
OK, so I've recapped all of the boards correctly with no major muck-ups. Also discovered that the output transistors were changed in the past to silicon.

But now I'm at the limit of my understanding of how to adjust the 4 trim pots and what the output voltage should look like. I've measured the outputs under an 8 ohm load and it's in the -180mA DC range. This seem excessively high. When I move the trim pot on the output boards the reading does go down, but I don't know how I should balance the transistor trimpots vs. the output trim pots. I need help !
 
do you have a service manual ? all i can find is a schematic .
are you measuring dc without speakers ? this is cap coupled output so should drop to 0v with loads at speakers . see what voltages are right where the 0.27 ohms resistors join together .
 
No service manual that I am aware of. Part of the problem is I've been used to better service manuals and this schematic is hard for me to understand.

I have an 8 ohm dummy load hooked up, was measuring the 180mV DC across those.

Where the .27 resistors join
L = -13.99V DC
R= -13.74V DC

Left side output transistors are noticeably hotter than the right side, too. I'm going to test them in a minute
 
they should be about half the rail voltage . adjustable with r309 left channel r359 right channel . proper way is with a scope and set the clipping symmetrical . no big deal just setting to half rail voltage really . probably never get turned up loud to matter .
no idea what the idle currents should be set too . or if they can even be adjusted right with silicon transistors unless the circuits have been modified . does it look like any resistors have been changed ?
the other 2 trim pots are for bias .
one thing i find with cap coupled outputs is one setting can affect the other so i use 2 meters to make it easier to set up .
did you change the coupling caps c306 c356 ? might be leaky .
i would check and record voltages right across the 0.27 ohm resistors first ..would expect anything from about 10mv to about 20mv there .. i am not very good with maths so find it hard to work out current from the readings .
 
I did replace all electrolytic caps, C306/356 are brand new nichicon UPW series.

So, I take it the rail voltage is the -35V DC I've been measuring? I admit I've been messing around with the trim pots. I changed the ones at those output transistors to be closer so now they're reading around -14.5V DC.

The other trim pots I adjusted so the voltage at the speaker outputs (with load) is -50mV DC and -19mV DC. That seems more acceptable.

I will bring in my oscilloscope tomorrow but that hot Left side concerns me. What could that be??
 
you need to measure across those 0.27 resistors to set the bias at idle . vol down no inputs . adjustable with r368 and r318 .
dont worry about voltages at speaker outs just yet .
 
no i dont think i am explaining it properly ..
leave the bias pots as they are or put them back to where they were in the first place .
then measure dc mv across both 0.27 resistors same connections as between q325 E and q324 C .
 
Ah, I mucked it up.. measuring on the + side when I should be looking on the - side.

-12mV DC and -14mV DC
 
Im sorry, Pete. Could you treat me like complete idiot for a second. When you say "measure across" I interpret that as one probe on the chassis and the red probe sticking on the thing I'm measuring.
 
KLH Model 20 voltage.jpg
I've included the measurement "across the resistor" which was one probe on one side of the resistor and the other probe on the other side. Tell me which one is right :)
 
Nevermind all that other stuff I wrote. I'm sorry it took me awhile to understand your simple instruction.

Across the Right side it is -30mV DC and the Left side is -305mV DC.
 
OK, Pete I think I'm getting where you're going with this. I hooked up multimeters to each side. It's been running for about 10 minutes now (aux, dummy load, vol down) and I've got both sides running at -27mV DC.
 
Back
Top Bottom