• 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

Pioneer SX-950 HELP!!!

Ray122

Ray122

Member
Hey everyone!

In desperate need of some help here. I'm working on an SX950 for a friend, and for the life of me, I can't seem to keep the output transistors on the right channel from frying. I've gone through this a few times now and it's getting very frustrating. I had everything working just fine a few weeks ago and naturally, after giving it back to its owner for a half hour it blows up again. So after being fairly embarrassed by that, I replaced all the semiconductors in the right channel. Did a DBT with a 75 watt light bulb, everything looked good. Even played audio through it, and got the protection circuit to kick in and out under the DBT when I turned the volume up too high. Adjusted the bias levels according to the service manual, again under DBT, and everything was looking good. Came out of protection mode every time I powered it up under the DBT. Took it off the DBT figuring I was in great shape, and next thing I know I hear a little ping and a little smoke came out of the output transistor.

Believe me when I say I've checked everything. Also, keep in mind the left channel is all original and works great. All my voltages are good, from the power supply to the amp. All my voltages in the amp were good under the DBT. I'm really at a loss. The only thing I can think of is that my replacement output transistors are knockoffs. I did buy them from eBay. Everything Else I get from mouser. I've got everything pretty well documented so please whatever you're thinking, ask away!

Please help!
 
Register to hide this ad
Ray122 said:
The only thing I can think of is that my replacement output transistors are knockoffs. I did buy them from eBay.
Click to expand...
There is a good place to start, if it is what you think then we can use this thread as an example of the reasons why not to buy from unknown sources.

Adjust the bias and offset when off the DBT. You only power up on DBT with bias set at minimum.

BTW what was it you buy as fire works?

Use the MJ21193/MJ21194 or MJ21194/MJ21195 from a authorized distributor. Suggest to do both channels (esp if you are going to blast it), we blew the first set of original Toshiba's in the 80's, under warranty, finally put in 250W Motorola devices (MJ15003/4), been in there ever since.
 
We use the DBT to check the idle trimmer for response and smooth operation before bringing the amps up to full house power. After checking the idle adjustment we back it off to min voltage and reset it to the proper specs on house power.
You had the idles set at the specs under the DBT which was limiting the current. Once the set was on full house power, the idles were way too high and hence blew the outputs after a short time. Sorry for your loss.
Rick mentioned the same explanation but I embellished for the layman like myself. :biggrin:
Once the DC Balance is set either on DBT or full power, it remains good and really doesn't need another look, unless you had to go back into the amp for whatever reason.
 
Last edited:
Thanks for the quick reply!
Now that you both mention it, that makes a lot of sense... This isn't my first repair by a longshot, but I'll admit I've learned a thing or two from this project. Although it's not like I had the idle adjustment cranked clockwise, and it was an immediate blow, rather than a short time... but maybe that in conjunction with crappy parts was all it took.

To answer rcs16, I bought a set of 2SB557s and 2SD427s. Specifically from this guy.
https://www.ebay.com/itm/2SB557-2SD...bevHP:sc:USPSFirstClass!11779!US!-1:rk:2:pf:0

I'll see if Mouser has the MJ21193/MJ21194 or MJ21194/MJ21195, and order those for sure, but in the meantime, I've got another set of outputs... Do you guys think I should try bringing down the idle current and giving it another go off the DBT, even if it's just for another fireworks show?
 
Well, I brought the idle current all the way down, threw the outputs in and gave it a whirl. It worked for a little while. Probably had a good 20 minutes before I tried cranking the volume and I noticed the 150-ohm emitter resistors on the drivers going up in smoke. I should probably mention I've upgraded those from a half watt to 2-watt resistors.

I'm going to go ahead and order a set of MJ21193/MJ21194s, but does anyone have any additional thoughts? If I have all the right voltages according to the schematic, and if it's matching the working channel, does that really point to crappy parts? I'd hate to blow up any more outputs, especially if I'm buying good ones from Mouser. Still at a bit of loss.

Thanks for all the insight this far!
 
It almost sounds like a thermal runaway problem, as if the bias circuits did not have any compensating reference junctions thermally tied to the output devices.

Then again, if they are mis-labeled imitation output or driver devices, all bets are off regarding their true performance characteristics.
 
Watthour said:
It almost sounds like a thermal runaway problem, as if the bias circuits did not have any compensating reference junctions thermally tied to the output devices.

Then again, if they are mis-labeled imitation output or driver devices, all bets are off regarding their true performance characteristics.
Click to expand...
I've tested the stacked diodes that mounts to the heat sink using the diode check on the multimeter as well as the ohm meter. The one in the non-working channel matches the one in the working channel. Also the voltages around the stacked diodes matched on both channel. Those would be some tricky parts to replace...
 
On mine, the ink colour is red not black. Hand written Toshiba logo. Those look suspicious. Funny how none of them both to read the original Toshiba datasheet which states they are 80W devices.
So possibilies are, junk parts or the bias generator is going out of wack over biasing the o/p.
monitor the V across the emitter resistors to watch what it is doing, shutdown/investigate if it goes out of wack.
 
Ray122 said:
...I'm going to go ahead and order a set of MJ21193/MJ21194s, but does anyone have any additional thoughts?
Click to expand...

When I rebuilt my SX-1250 the I used MJ21194G and MJ21195G outputs, and it has not had a problem since 2014 running that way. The absolute maximum ratings across the 21193-21196 parts are identical, but the 21194 and 21195 seem to have closer current gain curves than the recommended 93/94 pair or 95/96 pair. Either way, you can't lose.
 
Okay, glad the general consensus is fake news... I mean parts. Thanks, everyone, this has been very helpful. And good advice from rcs16! When I get the new parts, I'll hook up a set of multimeters across the emitter resistors on each channel so I can see if the right channel starts throttling compared to the left, and with any luck be able to shut it down before I lose the outputs.

tsd71, thanks for posting that article! That sounds exactly like my problem. Had no idea fake parts were this much of a problem. Hard lesson learned.

Also, much thanks for the part recommendations. I'm going to pull the trigger on the MJ21196/5 from Arrow on account that free next day shipping can save me $10 and get it to me sooner then Mouser. Haven't bought from there before, but I'll trust they're just as reputable as Mouser.

I'll keep you guys posted! If I have some time today and dig through the trash to see if I still have the original fried 2SB557s and 2SD427s, crack them open and compare to the alleged fakes.
 
Popped open one of the outputs that I've previously fried. I'm no expert by a long shot, but I'm going to go ahead and say that those tiny charred remains were never an 80W device... Unfortunately, the originals are long gone by now, so I have nothing to compare it to.
 

Attachments

  • IMG_0117.JPG
    IMG_0117.JPG
    48.3 KB · Views: 45
  • IMG_0118.JPG
    IMG_0118.JPG
    36.4 KB · Views: 42
Ray122 said:
Popped open one of the outputs that I've previously fried. I'm no expert by a long shot, but I'm going to go ahead and say that those tiny charred remains were never an 80W device... Unfortunately, the originals are long gone by now, so I have nothing to compare it to.
Click to expand...
Fake as fake can be! House power zapped that tiny little die.
 
Time to go back and leave some backside-flaming negative feedback. Those are beyond a joke.
 
You must log in or register to reply here.
Back
Top Bottom