• 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

SX-1250 Power Amp VR-2 wont adjust.

Stevesmullen

Stevesmullen

Well-Known Member
Hey Everyone.
I have a SX-1250 that the VR-2 on the left Power Amp is unresponsive to adjustment,the voltage is fixed at 220mv (desired 100mv) when read on pins #7 and #19.
I just recapped both power amp boards,Power Supply and Protection circuit,all other adjustable voltages are good.
I did not check these adjustments before the recap.
The unit sounds good ,but the left amp gets warm much quicker than the right side (obviously).
My first idea would be to slide the Power Amp board out and Ohm check the VR-2 itself.
I will wait for some tech support first.
Thanks Steve
 
Register to hide this ad
Test Q3, Q4, Q5 & Q6. The usual DMM diode test. Check for the off chance of bad resistors if the transistors test ok.

DO NOT under any circumstances allow the circuit between Q4 and Q6 (the two stv-4's and the variable resistor VR-2) to open up, as the voltages will snap to + and - 65v, turn on both outputs hard at the same time and blow the drivers and outputs to kingdom come... This is the worst place in the amp to have a problem due to that damage potential.

Q3 & Q4 are a current mirror feeding the Q5 & Q6 differential amplifier.

The voltage across R15 & R16 in the current mirror should be the same.

Also compare the Q5 & Q6 voltages to the schematic.

Also check (just in case) Q1 & Q2

These are two differential amp stages and they can do funny and non-obvious things to compensate for faults.

Any questions about the circuits, the wikipedia definitions of current mirror and differential amplifier are excellent.
 
Stevesmullen said:
Hey Everyone.
I have a SX-1250 that the VR-2 on the left Power Amp is unresponsive to adjustment,the voltage is fixed at 220mv (desired 100mv) when read on pins #7 and #19.
I just recapped both power amp boards,Power Supply and Protection circuit,all other adjustable voltages are good.
I did not check these adjustments before the recap.
The unit sounds good ,but the left amp gets warm much quicker than the right side (obviously).
My first idea would be to slide the Power Amp board out and Ohm check the VR-2 itself.
I will wait for some tech support first.
Thanks Steve
Click to expand...

Does the voltage remain the same no matter how far you turn the VR (in either direction)? If so, maybe the VR is bad? Just a thought, by no means an expert here.
 
Did you replace the pots? I first received my SX-1250 from ebay about 7 months ago, the left channel was very noticeably warm. It gets very warm very quickly. It gotten so hot until I finally realize it need to be checked. I checked the bias and it was 350mv 3 1/2 times the ideal 100mv. Fortunately it survived. The bias pot was intermittent but wouldn't hold the voltage down. After replacement, it works very well. If you haven't done so, get it fixed. Don't bother saving it with deoxit, replace it.
 
checking / replacing the trimmer isn't a bad idea. but if you can't get the bias down with zero ohms there is another problem.

Here's a parts list:
markthefixer said:
sx-1250 amplifier , "driver cards" board : AWH-048 replacement parts:

Mouser parts numbers, a mix of Fairchild (512-) and ON-Semi (863-)

differential inputs, better off if the hfe is matched
2sa750 to-92 ecb 50v .05a .25w 100mhz hfe:450
twenty: 512-KSA992FBU (ln)to-92 ecb 120v .05a .5w 100mhz 150-800hfe $0.05 ea

bias / vas
2sa858/2sc1439 to-92 ecb 150v 0.5a .5w 100mhz hfe:150
two: 512-KSA910YBU to-92 ecb 150v .05a .8w 100mhz 40-240hfe $0.12 ea
two: 512-KSC2310YBU to-92 ecb 150v .05a .8w 100mhz 40-240hfe $0.11 ea

pre-driver
2sa899/2sc1904 to-126 ecb 150v .05a 1w 100mhz hfe:150
two:x 512-KSA1142YSTU to-126 ecb 180v .1a 1.2/8w 200mhz 100-320hfe $0.21 ea
two: 512-KSC2682YSTU to-126 ecb 180v .1a 1.2/8w 200mhz 100-320hfe $0.31 ea

drivers
2sb630/2sd610 to-220 bce 200v 2a 25w 4mhz hfe:80
two: 863-MJE15032G to-220 bce 250v 8a 50w 30mhz hfe:70 $0.87
two: 863-MJE15033G to-220 bce 250v 8a 50w 30mhz hfe:70 $0.87
edit: use the KSA / KSC's not the MJE's - Vbe turnon is a bit lower and results in too much bias at times 4/25/09 mtf
drivers
2sb630/2sd610 to-220 bce 200v 2a 25w 4mhz hfe:80
two: 512-KSA940TU_Q pnp to-220 bce 150v 1.5a 25w 4mhz 40-140hfe $0.54
two: 512-KSC2073TU npn to-220 bce 150v 1.5a 25w 4mhz 40-140hfe $0.54

power outputs
2sb600/2sd555 to-3 200v 10a 200w 14mhz hfe:70
four: 863-MJ21194G to-3 NPN 250v 16A 250W 4mhz 8-75hfe $4.10
four: 863-MJ21193G to-3 PNP 250v 16A 250W 4mhz 8-75hfe $4.10

overcurrent detector
2sc869 to-92 npn bce 50v 30mA .2w 150mhz hfe:35
two: 512-KSC2310YBU to-92 ecb 150v .05a .8w 100mhz 40-240hfe $0.11 ea

here's the insulators and thermal grease:
ten: 534-4662 to-3 mica insulators $0.16 each (two spares)
six: 534-4672 to-220 mica insulators $0.12 each
one: 532-249 1 oz thermal compound $4.71


here's the trimpots:
four: 652-3386H-1-501LF $1.29 ea bourns 3386H 500 ohms

as for the capacitors:
four: 470uF/80v 647-UPW2A471MHD @ 1.92 ea nichicon pw - power filter
two: 10/16 647-UVZ1V100MDD @ $0.06 ea nichicon vz - offset smoothing
two: 220/16 647-UVZ1E221MPH @ $0.35 ea nichicon vz - feedback integrator

at this point, there's not much left original....here's the diodes:

1s1886 = 1amp 200v general purpose, being replaced by uf4004 = 400v 1 a fast recovery diode
eight: 512-UF4004 $0.12 d6, d7, d10 & d11
four: 1s2471 512-1N4454 $0.03 d8, d9
two: 1s2473 512-1N4148 $0.03

This is just so you can plan ahead.
There would be a few more parts for beefing up/renewing the protection circuits...
one: 522-ZTX1056A NPN TO-92E ebc 150v 3a 1w (Vce(sat)=0.3v @ 3a) 120mhz 30-1200 hfe relay driver transistor
one: 100uf/16v (100/35) 647-UPW1V101MPD $0.15 protection circuit timing capacitor
one: 4.7uf/25v (4.7/50) 647-UPW1H4R7MDD $0.23 protection circuit ac power on signal integrator
two: 100uf/50v (100/50) 647-UPW1H101MPD $0.50 protection circuit voltage regulator
two: 470uf/6v (470/16) 647-UPW1C471MPD $0.52 protection audio input caps
Click to expand...
 
Last edited:
Tugboat said:
Does the voltage remain the same no matter how far you turn the VR (in either direction)? If so, maybe the VR is bad? Just a thought, by no means an expert here.
Click to expand...
Yes the voltage remained the same (220mv).I Ohm tested the pot and it does appear to work.
Thanks for the suggestion

ccheath said:
Did you replace the pots? I first received my SX-1250 from ebay about 7 months ago, the left channel was very noticeably warm. It gets very warm very quickly. It gotten so hot until I finally realize it need to be checked. I checked the bias and it was 350mv 3 1/2 times the ideal 100mv. Fortunately it survived. The bias pot was intermittent but wouldn't hold the voltage down. After replacement, it works very well. If you haven't done so, get it fixed. Don't bother saving it with deoxit, replace it.
Click to expand...
I will replace them if they are bad or for preventative measure if I put in another Mouser order soon.
Thanks for the help.
 
Okay.
I was able to get the Power Amp and cooling fins seperated and insulated enough to do some live testing of Q1,2,3,4,5 and 6 and R15,16,17,18 and 19.
The Only 2 transistors worth reporting from live testing (I think) are Q3 and Q4.
Q3
Base-61.7
colle+62.7
Emit-62.3
Q4
Base-61.7
Colle-1.3
Emit-62
I thought that the voltage on Q3 Collector looked strange so I pulled Q3 after studing the schematic and ruling out C6 and R17 (my own little goose chase) Q3 tested good,which I still dont understand how?
So I fiqured at this point I would reference marks exact instructions that I have printed out and all of my Goose chase voltages written on.
DMM test on diode of following transistors.
Q3 Pass Original C1439
Q4 Failed B-E open and it is a NTE 194
Q5 Pass Original A858
Q6 Pass Original A858
I thought that the voltages looked proper on Q4 live testing.
Obviously Q3 and Q4 need to be a matched set.
I thought that I should mention that D2 has a broken leg on the black body,I dont know if this is a problem It came that way to me.
I dont understand what the expected voltages should be between the collectors of Q3 and Q5?
What next O wise one??
Thanks Steve
 
Waitaminnit........ D2, the stv3h?????? On the power transistor heatsink?

THAT's the worst case scenario....... This was playing music?? :YIKES:

Differential amps do strange things when one of the pair is bad, and we have two in a row, the second with a current mirror.

The collector voltage of Q4 will be about -1.8v and the collector of Q3 should be close to -1.8v .

I don't think Pioneer matched the second stage differential. Just replace both with a pair of ksc2310's.

On D2, try to whittle down the case a bit and see if you can get a prermanent connection, or if you can, just replace it. Maybe swap it for the stv3h on the on-board heatsink where it won't get mechanical stress.

I could try to analyze (guess) why things didn't go up in smoke, and it sort of resolves to the point that Q4 wasn't conducting, but even then I wonder if the stv3h let loose after your last powerdown?
 
Hey Mark or EW?
What can I do to prevent any issues when I plug this unit back in after I have replaced Q3 and Q4 and tested D2 stv3h.
Can I leave the Output Transistors unplugged from the Board and check the voltages somewhere first?
Steve
 
Since the driver transistors are referenced through their emitter resistors to the output of the amp, you can safely power up with the output transistors removed (but of course you cannot measure or set the bias).

You can still get DTV-3H diodes through Partstore: http://www.partstore.com/Part/Pioneer+Electronics/Pioneer/STV3HYG/New.aspx

There are a lot of resistors on the 1250 driver board covered in a fabric weave...these can burn up with no outward indication since the resistor is covered up. They all need to be checked.
 
EchoWars said:
You can still get DTV-3H diodes through Partstore: http://www.partstore.com/Part/Pioneer+Electronics/Pioneer/STV3HYG/New.aspx

There are a lot of resistors on the 1250 driver board covered in a fabric weave...these can burn up with no outward indication since the resistor is covered up. They all need to be checked.
Click to expand...
Thanks for the link for DTV-3H Diodes,I was curious if they were still available.
I swapped and Diode tested both STV3H diodes and then installed new ksc2310's (Q3,4).
Powered up the unit and now I am getting 186mv to 220mv adjustment on VR2 with voltage being read at pins#7 and #19.
I will pull the board and check out all the resistors covered in fabric ,and the other resistors in that circuit?
Anything else given the new readings??
Thanks Again. Steve
 
Stevesmullen said:
Thanks for the link for DTV-3H Diodes,I was curious if they were still available.
I swapped and Diode tested both STV3H diodes and then installed new ksc2310's (Q3,4).
Powered up the unit and now I am getting 186mv to 220mv adjustment on VR2 with voltage being read at pins#7 and #19.
I will pull the board and check out all the resistors covered in fabric ,and the other resistors in that circuit?
Anything else given the new readings??
Thanks Again. Steve
Click to expand...

Yes, establish a baseline.... FIRST ........

The BIAS VOLTAGE at the bases of the output transistors adjusts the IDLE CURRENT.

go to the GOOD channel, and disconnect the outputs, write down the voltage between the base connections of the outputs, (this is to be able to return it to the same setting) then see what the voltage is at the minimum and maximum trimmer adjustments.

The outputs are disconnected to protect them from the high idle current that would result at the top end of the bias voltage adjustment range, this way the adjustments are harmless, and you have plenty of time to take the readings.

THIS is what counts, the CONTROLLED bias voltage of 1.2v (+0.6v plus -0.6v) on the output transistor bases that adjusts the idle current flow with the SMALL change in that bias voltage.

You want to see that (JUST for EXAMPLE - you will get the real numbers) the bias voltage can adjust from -0.55v and +0.55v resulting in a reading of 1.10v across the bases of the output transistors to -0.65v and +0.65v resulting in a reading of 1.30v.

This voltage STARTS OUT as 2.6v (+1.8v and -1.8v) coming from the two stv-3h diodes and VR2, then drops to 2.4v (+1.2v and -1.2v) on the other side of the pre-drivers, the 2sa899/2sc1904 pair, THEN drops to 1.2v (+0.6v and -0.6v) on the other side of the drivers, the 2sb630/2sd610 pair.

so: we are looking for
the range of the voltage across the bases of the output transistors 1.1 to 1.3v to see if the outputs are being fed properly, or if your high idle current reading is due to a fault in the output transistors, there are a LOT of unknowns in between the pot and the emitter resistors where you are measuring the idle current.

1) the 3.6v (+1.8 plus -1.8) it starts with and how IT adjusts, example: 3.41 to 3.82

2) the 2.4v (+1.2 plus -1.2) coming out of the predrivers and how IT adjusts, example: 2.30 to 2.50

3) the 1.2v (+0.6 plus -0.6) coming out of the drivers and how it adjusts, example: 1.10 to 1.30

I hope this was clear enough for you to see how the bias voltage adjustment is dependent upon operation of the intervening stages and that the idle current measurement could be affected by some fault beyond the Q3, Q4 stage you were looking at.

So we are looking for something like:
(remember these are not measured to ground, but base to base !!)
(first number is pot at min, then second number is pot at max)

good channel 3.41 to 3.82 , 2.30 to 2.50 , 1.10 to 1.30

bad channel 3.61 to 3.92 , 2.50 to 2.70 , 1.30 to 1.50 (bias stage problem, everything else garbage-in / garbage-out)
or
bad channel 3.41 to 3.82 , 2.50 to 2.70 , 1.30 to 1.50 (bias stage OK , pre driver problem, everything else garbage-in / garbage-out)
or
bad channel 3.41 to 3.82 , 2.30 to 2.50 , 1.30 to 1.50 ( bias stage OK, pre-driver stage OK, driver problem,)
or
bad channel 3.41 to 3.82 , 2.30 to 2.50 , 1.10 to 1.30 (bias stage OK, pre-driver stage OK, driver stage OK, bad output(s) )

notice the numbers get off course later and later in the circuit, with the lower number NOT getting down low enough!!
 
Hey Mark.
Thanks for the Data and instructions,I will print and follow them when I can make time to be focused and thorough.
Thanks.
 
Hey Mark.
With the outputs diconnected there isnt any voltage between the Output Transistors PNP and NPN Base.
There also isnt any voltage at Pins #7 and #19.
Am I doing it wrong or did this not work the way you had calculated?
Steve
 
Stevesmullen said:
Hey Mark.
With the outputs diconnected there isnt any voltage between the Output Transistors PNP and NPN Base.
There also isnt any voltage at Pins #7 and #19.
Am I doing it wrong or did this not work the way you had calculated?
Steve
Click to expand...

You have to read the base voltages where they originate, the emitters of the driver transistors on the board or the pins going OFF the board TO the bases, not on the disconnected transistors.

AND pins 7 & 19 are SUPPOSED to be cold. That's the WHOLE IDEA... to not burn up your outputs while testing the all important bias voltage drives.

attachment.php
 

Attachments

  • sx-1250_amp_measuring_points.jpg
    sx-1250_amp_measuring_points.jpg
    93.9 KB · Views: 247
Thanks for the clarification,I thought that we were talking about the Outputs on the heat sinks.But this makes sense as it pertains to your instructions.
Im not sure if I have time tonight,but I will report when I do.
Thanks Steve
 
Hey Mark.
Well I finally made some time to get the readings that you wanted.
Both sides read equally at first with the OPs unplugged.
Q4 to Q6 Collector (1.895)
Q9 to Q10 Base (305mv)
Q9 to Q10 Emiter ( 0 )
So then I plugged the OPs back in on both sides and the Idle current was at 100mv on both sides then the left side started to climb up to 300mv (as read on pins#7 and #19 on the power amp boards.
The Voltage readings with the OPs plugged in are as follows for the bad side when the MV reading was at 300mv and climbing.
Q4 to Q6 Collector (3vdc)
Q9 to Q10 Base (3vdc)
Q9 to Q10 Emiter (1.64)
So then I swapped sides for the Power Amp Boards to eliminate the OPs and the problem also switched sides so the OPs appear to be good.
I quickly tested the voltages at Q5 and Q6 Base,Emiter and Collector and the voltages were normal,at this point I could feel the OPs starting to get warm so I quit.
Seems that when the unit is stone cold the MV reading is okay but then quickly starts climbing as the unit warms up.
Thanks Steve
 
I can't use those readings.

I took a lot of time on those last HUGE posts, getting ALL the example numbers consistent, and making sure ALL the language was consistent, leaving nothing open to misinterpretation. A careful examination of the posts should have left nothing to chance...

Then, I posted a picture of what to measure, right down to the lead color.

THERE IS NO GROUND REFERENCE FOR THESE READINGS, and 1.895 means you are measuring something wrong. A reading of 1.895 between Q4's collector and Q6's collector would NEVER allow ANY bias curent to flow, as well as indicating very serious problems with D1, D2 & VR2.
Also the readings were expected in pairs:
markthefixer said:
3.41 to 3.82
Click to expand...

markthefixer said:
(first number is pot at min, then second number is pot at max)
Click to expand...

1. NO OUTPUT TRANSISTORS CONNECTED UNTIL I SAY SO, AND IT WON'T BE IN THIS POST.
2. crank VR2 to MINIMUM OHMS
3. put one voltmeter lead ON Q4's collector and put one voltmeter lead ON Q6's collector, TAKE READING (3.41V)
4. put one voltmeter lead ON Q9's base and put one voltmeter lead ON Q10's base, TAKE READING (2.30V)
5. put one voltmeter lead ON Q9's emitter and put one voltmeter lead ON Q10's emitter, TAKE READING (1.110V)
4. crank VR2 to MAXIMUM OHMS
3. put one voltmeter lead ON Q4's collector and put one voltmeter lead ON Q6's collector, TAKE READING (3.82V)
4. put one voltmeter lead ON Q9's base and put one voltmeter lead ON Q10's base, TAKE READING (2.50V)
5. put one voltmeter lead ON Q9's emitter and put one voltmeter lead ON Q10's emitter, TAKE READING (1.300V)

NOW assemble the report:
"number one channel 3.41 to 3.82 , 2.30 to 2.50 , 1.10 to 1.30 "

NOW GO TO THE OTHER CHANNEL:

2. crank VR2 to MINIMUM OHMS
3. put one voltmeter lead ON Q4's collector and put one voltmeter lead ON Q6's collector, TAKE READING (3.41V)
4. put one voltmeter lead ON Q9's base and put one voltmeter lead ON Q10's base, TAKE READING (2.30V)
5. put one voltmeter lead ON Q9's emitter and put one voltmeter lead ON Q10's emitter, TAKE READING (1.110V)
4. crank VR2 to MAXIMUM OHMS
3. put one voltmeter lead ON Q4's collector and put one voltmeter lead ON Q6's collector, TAKE READING (3.82V)
4. put one voltmeter lead ON Q9's base and put one voltmeter lead ON Q10's base, TAKE READING (2.50V)
5. put one voltmeter lead ON Q9's emitter and put one voltmeter lead ON Q10's emitter, TAKE READING (1.300V)

NOW assemble your SECOND report:
"number two channel 3.41 to 3.82 , 2.30 to 2.50 , 1.10 to 1.30 "
 
Hey Mark.
Those readings are with the leads as you illustrated but I didnt adjust the trim pots to min/max.I will go back through and test again.The readings did seem odd.
Steve
 
Stevesmullen said:
Hey Mark.
Those readings are with the leads as you illustrated but I didnt adjust the trim pots to min/max.I will go back through and test again.The readings did seem odd.
Steve
Click to expand...

Don't forget, we want to do it to BOTH boards, to get a baseline comparison from the good board.

The fundamental theorem of information engineering is that the value of information is inversely proportional to it's improbability...

In otherwords, when you look at a group of readings and say WTF?????????? it better be trusted information.... that's why I'm comparing the good and bad...

BTW, I'm trying to force myself off of vampire type third shift hours, because I have all this building code crap to contend with, and can't be asleep during the daylight...
So no more late nite stuff. 6am - 10pm iiac....
 
You must log in or register to reply here.
Back
Top Bottom