• 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

HK DC Balance Problem

R

resoundpdx

New Member
I have been working on an HK6650R amp which is now working perfectly - almost. Both channels of the power amp showed distortion and the DC balance was WAY out on both (multiple volts). I reheated all solder connections on the power amp board and was then able to dial in the DC balance on one channel perfectly. On the other, the output is now clean and undistorted, but about 1.5 volts is the best DC balance I can achieve. I have measured all resistors (all good and same between the channels) and all transistors test good. The power rails are steady and identical between the two channels. The adjustment pots are also good. I have read through every thread on this problem on AK that I can find, but am pretty much out of ideas. Suggestions?
 
Register to hide this ad
You don't say what channel the problem is so check the voltage on the emitter of Q449 or Q450. Should be about -15V.
 
Thanks for the quick response. Emitter of Q449 (bad channel) measures -13.85v. Emitter of Q450 (good channel) measures -13.88v.
 
Ok, your supplies are good. If you have +1.5V on the output you may be missing drive on the negative half of the amplifier.
Start measuring the voltages on each transistor; E C B and record them here with the reference designators.
 
Will do. It should be instructive. I have been puzzling over why the feedback to the differential pair would not "self correct" the offset. Going through methodically will help to answer that.
One nice thing about working on stereos I have found repeatedly is that when there is one channel working you have a basis for comparison.
I appreciate your help, my friend. Stay tuned!

david
aka reSound Audio
Portland, Oregon
 
OK. Here are the measurements. Tolerance is 0.1 or 0.2v.
Power rails on output drive transistors = +42.8, -42.7
In all of the following, the order of numbers is E, C, B:
Q403 -0.6, 41.5, 0.0
Q405 -0.7, 41.0, 0.1
Q407 -13.4, -11.3, -12.7
Q409 -14.0, -12.7, -13.4
Q449 -14.0, -26.2, -14.6
Q411 41.3, -40.6, 40.5
Q413 40.3, 42.3, 41.1
Q415 -40.4, -42.2, -41.0
Q417 0.0, -0.1, -0.7 [remeasured several times]
Q419 -40.1, -40.1, -39.5 [remeasured several times]
Q421 41.1, -0.3, 40.6
Q423 -40.8, -2.6, -40.2
Q425 41.0, -0.3, 40.4
Q427 -40.6, -2.6, -40.1
Q429 -2.6, -0.2, -2.0
Q433 -0.8, 43.6, -0.3
Q435 -2.1, -42.0, -2.6
Q437 42.9, -0.8, 42.3
Q439 -42.6, -2.0, -42.0
Q9 -1.4, -42.8, -1.4
I did not check the output drivers but can do so if of interest.

david
 
I guess I could compare all of these measurements with the other/good channel. But seems like focus should be on the feedback path - for some reason the offset at the output is not being presented back to the differential pair for correction.
 
Diode test Q419, looking for c-e short.
Check fusibles FR443, FR451 both 180 ohms, still in spec?
Have you replaced Q801 or Q429, replacements may have slightly
different (Vbe) properties.

-ve half voltages swamping +ve half of amp so Q433b is lower than
expected (+1.2Vdc?)
 
I have not replaced any transistors on the amplifier board so far. I did replace C411 on a hunch, but that made no difference. And R27, which was visibly charred from some prior mishap.
Q419 diode tests fine (I had the same suspicion). FR443 and FR451 both measure 179 ohms.
In case it helps, the idle dials right in very smoothly.
Since overall this channel of the amp functions nicely (good power output and low distortion with sweep input) I keep wondering if there is just something out of whack in the feedback path. But that is really just a chain of resistors so not much to go wrong. I know that is simpleminded but will double check all resistances just to satisfy myself.
More insightful suggestions definitely appreciated!

david
 
I agree with mbz, additionally I think you have some voltage measurement errors: You have the base of Q421 as 40.6V but you have Q413 emitter as 40.3V and they are connected, so which is correct? It makes a difference as to whether Q421 is turned on. Please remeasure the voltages on Q421 and Q413. Also check R435, could be open.
 
I was watching for measurement consistency in cases like that during the process, and went back to check several measurements that did not agree. That one escaped me. I also noticed some slight shifts as the amp reached stable temperature, which is doubtless the cause of some minor discrepancies. I'll recheck Q421 and Q413, as well as R435 and R437.
 
Well, no smoking gun:
R435 and R437 are spot on @ 12K
rechecked these others while at it:
R419 = 39.4K
R423 = 390
R425 = 22
R433 = 20.1K
Remeasurement of Q421 and Q413 shows consistency:
Q413 = 40.3, 42.3, 41.1
Q421 = 40.8, -0.3, 40.3
Offset on output still ~ 1.5v min :-(
 
R435 = 12.0K
R45 = 22.3K
R441 = 2.19K
R445 = 179
R459 = 131
But.......
R429 = 720 BINGO
Replaced with 560 ohm resistor and DC offset dials right in.
(I did not have a fusible resistor so used normal one).
Looking at schematic this makes sense to me now. But I don't think I ever would have found it without your patient help.
Can't thank you enough!

By the way, why does the design have R435 & R437 in series with no intermediate connection? Same for R441 and R459? Why not use a single resistor? Is it because the exact value can only be achieved with a pair?

david
 
Glad I could help. Usually putting 2 resistors in series is so you can use lower wattage resistors, or because they needed an odd value.
 
You must log in or register to reply here.
Back
Top Bottom