• 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

Matching hFE for Differential Pair

gort69

gort69

AK Subscriber
Subscriber
Apologies if this has been asked before...

Which pair would likely achieve the lower DC offset - a pair that measures dead nuts at 550, or a pair that measures 605/600?

The meter I'm using a cheapo ($30 ish) HF cat. no. P3772 meter. Working on a KR-9600, using KSC1845FTA, NPN.

Thanks in advance.
 
Register to hide this ad
I think it will be almost the same , actually the real conditions on the circuit will differ from the multimeter test.

I'd use the 550 pair, and swap them if you have more than 20 mV offset. Sometimes swapping them the result is different. I guess the rest of the components in the circuit makes that difference...
 
Were they at the same temperature when tested? If so, Elnaldo has you covered. There are several differential pairs in the power amps, but no offset adjustment other than fiddling with the differential pairs between Q4/5 and 6/7 - Interesting.
 
Thanks guys -

Yes - I tested them all this afternoon - 120 of 'em! I've been told Qe1, 2 are the pairs I should match?
 
I thought offset was determined by Vbe match? I'm sure I read that somewhere. Gain match is more about distortion.
 
The surrounding resistors need to be matched on the second pair. The first aren't balanced to begin with. I'd probably install some SIP sockets for experimentation. Uses power packs so no mistakes!
 
Thanks guys - completed the left channel this morning, and by hFE matching Qe1 & 2 reduced the DC offset in it from 198mv to -1.6mv. Not too shabby! I used a pair at 594/592
 
Impressive. I still think it was a little weak of them not to incorporate a trimmer, but whatever floats the boat, I guess.
 
Well I overstated it a bit. I had the selector in AUX - which I hadn't tried before - when I made the test. Turns out the selector switch was filthy and wasn't passing much of a signal to that channel. Once all cleaned up and working correctly DC measures 7.5mv on that channel all warmed up. I'll take it!
 
DRecovery said:
I thought offset was determined by Vbe match? I'm sure I read that somewhere. Gain match is more about distortion.
Click to expand...
True this. For differential pairs, Vbe (beta) is a better reference. And for drivers, finals, etc... hfe is a better reference...
 
If you look at a true matched bjt pair spec, they usually state Vbe matching as the criteria.
Some as I refer too below, match too as good as 1mV.
http://linearsystems.com/assets/media/file/datasheets/LSIT120-2.pdf
You can make a simple differential pair circuit and measure the voltage across the collectors to select for a near perfect match.
Better use some precision resistors like 0.1%, or hand select matched R's as you do not want the resistors to affect the readings.
 
gort69 said:
Well I overstated it a bit. I had the selector in AUX - which I hadn't tried before - when I made the test. Turns out the selector switch was filthy and wasn't passing much of a signal to that channel. Once all cleaned up and working correctly DC measures 7.5mv on that channel all warmed up. I'll take it!
Click to expand...

You're not supposed to have any signal at all from the preamp when checking the offset. You should have the volume set to zero and the selector set to aux with no signal going in the jacks.
 
I'm still pretty green at this stuff. Before I even connected speakers to see if the amp board I rebuilt worked, I tested for DC offset. I didn't have a source connected to the AUX input jacks for the first test. I did have the selector set to AUX, volume at min, tone defeated. The first test I did gave me the -1.6mv reading. When I was done, I connected a CD player to the aux inputs and got no sound out of the left channel. Scratched my head and muttered a few curse words, so tried the CD player on a tape loop and the left channel worked fine. So i cleaned the selector switch and got the channel back when set to AUX. Disconnected the CD player, selector set to AUX, volume at min, tone controls defeated, and got 7.5mv. Valid test?

Another question - would the test be valid if I were to remove the pre/power jumpers?

Thanks for the input guys
 
I don't think they tell you to remove the jumpers to check the offset in most service manuals, just to have no input and volume at zero (and sometimes an 8 ohm dummy load but that's not really necessary).

Now if you have a noisy transistor or other component in the preamp section that happens to be after the volume control you'll get all kinds of screwy offset readings because it is introducing noise into the power amp. In a case like that I suppose you could decouple the pre and power sections to check offset but you'd have to address the noisy components in the preamp section eventually.

The other side of that coin is the possibility that there may be noise introduced by the power amp input connections being left open with nothing connected to them, it depends on the particular amp.
 
You must log in or register to reply here.
Back
Top Bottom