Marantz 18 receiver needs help

[nickb]

Removed R14 stuff...oopps!

OK R42.... If R42 is burning up then the voltages are not what was in post #73. At that time the collector of Q10 was -42.5 i.e assumed close to the supply. I don't see the supply volts in the list of measurements ( i.e Q13 and Q14 emitters). It needs to be known to give a complete picture. It may simply be a case of trimmer R27 being too low a resistance - set it to max. IF still cooking redo the voltages.


On the missing signal to the right channel you need to trace down thru the preamp to find where it disappears using your scope.

I just noticed this thing has a CRT display! Kewl!

[bigdrums925]

Not R14 but R42 across the output Q14

[nickb]

I removed the reference to R14 and updated post #76. I misread....

[bigdrums925]

Got the R42 (& R43) burning solved by turning the balance pot R27

new list of voltages I hope are correct at a little less than full variac:

b c e
Q14 -37.9 0 -37.8
Q13 38 0 38
Q10 0 -37.9 0
Q9 0 38 0
Q6 0 0 0
Q5 0 0 0
Q2 0 0 0
Q1002 36.6 1.5 37.8
Q1003 0 .9 1.6
Q1004 .9 -34 -34
Q1001 25.6 1.6 25.9
Q1005 .8 -33.7 1.6
Q1006 -33.7 -1.4 -34

There are probably discrepancies here but lets see if I can make any sense of it with your help
Thanks

[nickb]

Given that the output node is correct at 0V I suspect that Q1004 collector at -34V is likely a typo and that things are basically OK with the power amp DC. Can you work thru the preamp so find where the signal disappears?

[bigdrums925]

IT WORKS! Thank you so much. and especially for the drawing. How do you do these so quickly?
I'm gonna put all the pieces and parts back together
I'll report back one way or another
BTW what should I set the idle current at? And where should I measure from?

[nickb]

Yea! If you have the new reference diodes fitted follow the procedure on pg 14 of the schematics, basically 80mV across R35 and R36.

That is an unusually high bias current given there are no ballast resistors in the output IMHO. If you have the old diodes fitted I would be more conservative. Hook up a dummy load and drive and a few Vpp into it. Gradually increase the bias current until the crossover notch just disappears and checking (with zero signal out) that the drop across R35 or R36 is no more than 30mV. This is a rare case where a distortion analyzer would be the ideal tool.

I've been drawing schematics for (many) years so have had lots of practice I guess. Final version M18-8.pdf

[tonebells]

Great job getting it going.

Tony

[bigdrums925]

Once again thanks for all the help. Pretty sure I could not have done this without, err I'm sure I could not.
The client was ecstatic about getting his nearly 50 year old receiver going again. He bought it new 1969!
You guys are the best

[bigdrums925]

This receiver has come back more than once since it was finished. In each case the output transistors eventually had to be replaced as well as the current limiters. At least once it was because my probe shorted out due to the very compact space the amp components are arranged. I would like to know what would happen if the PNP/NPN output transistors were reversed?

[The Dude]

It wouldn't work and would blow a fuse almost immediately.

[nickb]

Ideally, you will have to figure out why. It's either too much current, too much voltage, or secondary breakdown which is a bit of both.

My money is on current due to thermal issues and that's easy enough to test for. Run it into a dummy load at half power while monitoring the drop across R35 (80mV at idle). Do it for say one minute and then check the drop across R35 again and check it has not increased. Repeat several time to get the amp hot, checking each time and stop if the voltage gets any higher than (say) 100mV.

[52 Bill]

And another thing to ask the customer about are the speakers and wiring that this is being connected to at his location. Fixing the amp is only part of the equation. If the amp is being hooked up incorrectly or to a load that is too low for the amp to handle, it can cause damage to an otherwise working unit.

[bigdrums925]

What do you think about using a variac with a light bulb limiter in an effort to prevent destroying any more components in an amp?
Reason I ask is.....I often find myself ramping up the variac carefully monitoring the amp meter I added to it to prevent such things.
It has a 1.5 amp scale. Usually I see the current rise slightly before settling down after a peak of about .7 amps to settle down to maybe .3 -.4 amps or less depending on the design of the amp.
However tube amps often will need a bit more and idle at a higher level I have noticed.
This particular Marantz unit has the tube scope display built into its circuit and idles around .8 amps if I recall correctly.
So I am leery to push the current because I've blown so many of the output transistors and had to replace them at $10 ea.
Obviously I don't want to blow any more of them.
When I remove the output transistors (on the right channel) the current draw is very low all the way up to 100% variac.
I've A-B DCR'ed to ground all the points comparing the L&R amps which reveals no real differences.
The only thing left to do is remove the output transistors and power up to check the voltage maps again.
Comments?

[olddawg]

And another thing to ask the customer about are the speakers and wiring that this is being connected to at his location. Fixing the amp is only part of the equation. If the amp is being hooked up incorrectly or to a load that is too low for the amp to handle, it can cause damage to an otherwise working unit.

In most shops I worked in we had printed notes that were taped to the unit after an output failure callback that said, "No further repairs would be done without the inspection of customers speakers and wiring.." In old days it was especially problematic with screw type terminals. People tend to strip wire back a couple of inches and they touch.