• 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

Spec 4 Surprise Overhaul and restoration

Yes, I did...when I discovered the idle was close to behaving, I took it out. At the same time I put the power limiter transistors back in. Removing the 10k resistor still only got me to 47mV.
 
Sp33ls, I dont have all the necessary test equipment to calibrated the power limiting (clipping). I do have a portable fluke oscilloscope, but no signal gen, or distortion meter. So I have a fellow AKer in the next town over who can do this.
 
One more question...I got my new output installed and everything works fine however I can only achieve 24mV when adjusting idle...is this bad to let go?...it calls for 50mV but I am considering letting it slide unless told otherwise. Will not having 50mV damage my channel in any way?...it sounds awesome and outputs never got hot after 1/2 hour of mid volume. Right channel is perfectly within specs.
 
as a general rule, increasing idle current reduces distortion UNTIL a knee point in the graph is reached, after which distortion does not change and only product of increased idle current past that point is increased heat input to the output transistors that has to be dissipated.

that said, i find your situation continues to spark my concern.

i would look at every transistor on both amps, comparing the Vbe at every transistor position amp channel to amp channel in a no-sound state.

something is still afoot.
 
I agree Mark, its weird how changing the drivers made the idle go in the other direction. Maybe because the Vbe is so different?...the other amp with older mje15030 and 31 worked perfectly. I am still looking for better matched pairs. The only things left are transistors and polyester caps that havent been changed. All resistors have been lifted and checked. Gotta be the Vbe's. Can the stabistors cause this, because those havent changed?...
 
You dont suppose the power limiter pots have anything to do with this do you?...this channel clips off about half volume, the other one doesnt.
 
clips?!

ok, this is for diagnosis:
get the clipping condition occurring
turn the vr4 (or 5) to stop the clipping
continue turning the pot in that direction, until the pot stops at the end. what direction was it?

power off, dmm to ohms
dmm black to slider connection of pot.
dmm red probe th1,R31 & R44 to determine which end of pot it is at (zero ohms).

results could be like: VR4 CCW goes to (th1, R31) or (R44) end of pot to eliminate clipping.
 
Okay Mark, will try this after dinner. Thanks for all your help. Will find a way to thank you properly!
 
Hi all, Back to the spec problem after a sanity break...so I am still trying to diagnose why one amp wont idle to 50mv..so I checked and rechecked my transistors placement and function. I made an observation right off...Q8 and Q10 get extremely hot right off after power on. Q9 and Q11 just get warm. Checked all the resistors in this path...what is left? Any ideas?...I am almost there!
 
Mark, I found the problem. R43 (4.7) was bad and I replaced that. Now sort of back in business except I am in the same boat as before. I am getting a steady 88 mv for idle...better than 120!...I put the 10k resistor back in R13 like before and it went up to 90mv...larger resistance maybe?...I am getting there. One more push and I owe you a cold one!
 
Q8 & Q10 don't get inordinately hot now?

88 mV is at VR2 = 0 ohms?

and putting 10k across (NOT IN PLACE OF!!!) R13 (620) raises the idle current?
 
Nope, just warm like the other side. I did put across, not in place of. I turned VR2 all they way clockwise and 88 is as low as it would go.
 
It raised it just a few MV, but that could have been warm up time. We would still like to get VR2 somewhere in the middle wouldnt we?
 
Ok, we're gonna need a lot of voltage measurements. From the trouble channel and from the ok channel.

There is a quick check that can be done - VERY CAREFULLY!!!
(red to base, black to emitter - use clip leads!!!)
measure the emitter to base voltages of Q8 & Q10 & npn output transistors (pin 2724b-26e, 23b-25e)
measure the emitter to base voltages of Q9 & Q11 & pnp output transistors (pin 19b-17e, 18b-16e)
on the bad (hot) channel and then the good channel for comparison.
8 readings hot, 8 readings good

now, for the other readings:

ok, on power off,
check R20 & R21 for 33 ohms.
Check R33, R34 for 220 ohms,
Check R35 & R36 for 100 ohms
Check R42 & R43 for 4.7 ohms - I'm sorry that they are all fabric covered. Don't lift a leg this test...

I have lost count - are Q6, Q7, Q13 & Q14 in circuit ? or not? Just for my mental model.

first, your dmm black probe "ground reference" is going to be AND STAY PIN 11


Ok - power on and take voltage readings of:

One side of R20, then the other side of R20 (same as Q8 base)
One side of R21, then the other side of R21 (same as Q9 base)
Q10 base, Q11 base
Q10 emitter, Q11 emitter
One side of R42 (same as Q10 emitter), then the other side of R42 (bases of npn output transistors)
One side of R43 (same as Q11 emitter), then the other side of R43 (bases of pnp output transistors)

If there are any significant differences in voltage between sides of R20, R21, R42, R43:
now is the time to disconnect the black dmm probe from pin 11.
read the voltage ACROSS that resistor.
black dmm one side of resistor,
red dmm the other side of that resistor at the same time.

edit: pin 27 was wrong, pin 24 is correct. 4-26-2022 @ 5:45 pm
 
Last edited:
Mark, I just discovered something. I have the two new pots VR1 and 2 reversed. This wouldn't make a difference would it?. They are both 470 ohms and I did the same thing on the working board, so I would image not. But you never know.
 
FYI
I have found that mje15030,31 require more Vbe than mje15032,33 and even ksa940,ksc2073
Imo
You are putting way to much effort in to matching devices in a 3 emitter follower design, I only care about the input stage for Vbe match and that’s usually good enough, you do not have the test equipment to measure what the results are anyways and if so we’re talking about small amounts
 
You must log in or register to reply here.
Back
Top Bottom