• 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-1010 Bias Current too high?

jackgiff

jackgiff

Well-Known Member
I just purchased a Pioneer SX-1010 from an AK'er (thanks fdejong) and it had what I assumed was a power supply problem. It would only play for a few minutes before it starting going in and out of protection.

When I received the unit, the wood cover and bottom were the first things to come off. There was no obvious damage to the power supply board. I had sort of expected to see it in pieces.

The problem was traced to the protection circuit, and one of the 330 uF caps was the culprit. All of the electrolytics on this board were replaced, except for the .22 uF's, which I didn't have. No .22 uF mylars or foils either. I will pick up some at my son's company, and replace the electrolytics with mylar.

Now the unit plays for hours on my bench, but WOW! The two amplifier heat sinks got amazingly hot. I checked the offset and bias, which were both within spec. But the bias current specified ends up consuming 22.4 watts in biasing the output stages. Seems excessive to me.

The SX-1080, with similar power amp and current sensing technology, uses only 30 milliamps instead of the the 50 milliamps specified for the SX-1010. I set the bias current in the SX-1010 at 30 MA, and can't tell the difference just by listening. But it sure cuts back on the heat from the output heat sinks.

The distortion may be increased by cutting the bias current, but I have no distortion measuring equipment. Maybe one of the Pioneer gurus will pop in here, and give me a lesson on how that is wrong. I am certainly no audio amplifier designer.

The parts for the power supply board re-cap are on order, and the performance of the 1010 is really very impressive.
 
Register to hide this ad
There a real nice sounding unit once recapped and all the noisy transistors are replaced.
 
I get 10.8 watts..

54v x 2 = 108 volts rail to rail

50mA idle current per pair x 2 pairs = 100mA idle current (confirmed by schematic) edit: this 100mA specification is for BOTH CHANNELS, thus 25mA per pair x 2 pair x 2 channels.

108 x 0.1 = 10.8 watts...

yes, you could send a sine wave through, measure the distortion, reduce the current, again measure the distortion, graph it and find the "knee"...

But that is why it is called a class AB amplifier, and the idle current sets the point for the class a to class b transition.

since the 1010 was top of the line, maybe they figured a bit more heat for extended class a operation was worth it.

I haven't graphed that, (distortion versus power / voltage level) but now I will be thinking of it the next time a 1010 is on the bench, but you have to remember that the noise floor now becomes a significant contributor to distortion. distortion is ANYTHING that isn't part of the sine wave at the appointed frequency.

edit: see my next post, this has a basic error in the initial assumption.
 
Last edited:
Thanks for the reply, Mark. I greatly appreciate it. Ask for a guru, and there you are. Your calculations are correct. Using +/- 54 volts, the wattage would be 10.8 Watts per channel, but there are two channels, which yields 21.6 Watts total. I used the spec'ed voltage of +/- 56 volts for my calculations.

Since I don't have any distortion measuring equipment, at least that I know of, it would be neat if you or someone could measure the distortion at both bias current levels. That is asking a lot, but would be greatly beneficial to some of us guys who don't like to feel his equipment near the boil. The schematic in the service manual on AK for the SX-1080 is so blurry, I can't make it out for sure, but the output looks to be close to the SX-1010. It seems if the higher powered unit can use less bias, and still have lower distortion numbers, maybe the SX-1010 can as well.

I hadn't considered that the bias current will set the point for class a to class b. I will go back and research that some. Thanks for the pointers.
 
Those heatsinks should NOT get hot unless you are trying to blast the roof off. Under normal operating conditions it should get 'maybe' lukewarm.

If your heatsinks are getting hot, something is very wrong...either with the unit or your measuring procedure.
 
I just realized my error!!

The schematic shows 100mA current, but that is for BOTH channels.

50mA per channel

which means 25mA per PAIR.

The measuring procedure is WRONG.

Pins 14 to 16 should show 25mV, which is more in line with other models.
 
EchoWars said:
Those heatsinks should NOT get hot unless you are trying to blast the roof off. Under normal operating conditions it should get 'maybe' lukewarm.

If your heatsinks are getting hot, something is very wrong...either with the unit or your measuring procedure.
Click to expand...

Thanks for showing up, EchoWars. I always value your opinions. My wife won't let me blast the roof off any more. And I don't need music that loud. The unit is setting on a flat table, with nothing on top of it. And it gets too hot for me when bias is set per the service manual.

Maybe my definition of "hot" would be like your definition of "lukewarm." After all, those terms are relative, not absolute. But compared to two Kenwood units, a KR-7600 and KR-9400, the Pioneer is much hotter than either. A Marantz 2275 sits in a cabinet in my bedroom, and runs much cooler than the Pioneer. Even a Nikko NR-1019 runs cooler.

Since the heat appears to be generated in the power amplifier heat sinks, and there seems to be no imbalance in the bias current of the two transistors in each pair, what fault could possibly cause the extra heat?

Do you agree with Mark, that the manual could be in error?

markthefixer said:
I just realized my error!!

The schematic shows 100mA current, but that is for BOTH channels.

50mA per channel

which means 25mA per PAIR.

The measuring procedure is WRONG.

Pins 14 to 16 should show 25mV, which is more in line with other models.
Click to expand...

Mark, the manual has an error on pages 14 and 15, showing the locations for the protection circuit assembly and the power supply circuit assembly being swapped. Perhaps another mistake in the manual is possible, and 25 MA per stage would set the operating power at a much more tolerable level than the 50 MA per stage that the manual calls for.

I will set them at 25 MA, and see what happens.

And I will attempt to get a thermometer reading of the heat sink temperature so we are talking absolutes rather than relative.
 
Last edited:
I borrowed Mama's kitchen thermometer, and held it between the fins in the middle of the heatsinks to get these readings. The wood cover was left on while everything stabilized, and then removed to allow access to the heat sinks for measurement.

At a bias setting of 50 MA, the heatsinks were measured at 130 degrees and 132 degrees. The bias setting was then checked, and read only 42 MA on one channel, and 44 MA on the other. Had they been re-adjusted to the called for 50 MA, the temps would have obviously been even higher.

At a setting of 25 MA, which was checked before measuring the temperature, the heat sinks were measured at 105 and 106 degrees. Much closer to what I would consider ideal.

My background is not audio (darn it), but rather design and manufacturing of metal detectors, which are battery powered. Having to operate for hours on a 9 volt battery or a pack of AA batteries means not wasting any power. If you can feel from the case temperature that the unit is turned on, you are using way too much power. Maybe that background has made me overly sensitive to heat produced by units that have a power cord to draw from, without limits.

The unit runs a bit warmer now than my Kenwoods, but close enough. I will leave it set at 25 MA, and hopefully someone will measure the distortion at both bias settings, and report it here.
 
Mark, that's a great catch.

My SX-1010 also runs also very warm at idle (no load), and it has been completely rebuilt. Also, judging from the typical values one sees for Pioneer receivers, the service manual spec of 50mv bias is a misprint. Found the same 50mv spec in the SX-939 service manual.

In any case, I will be setting my SX-1010 to 20mv.

Rich W
 
Last edited:
This is very interesting info. I've got a working 1010 that I set per the manual at 50 and it runs every day as cool as a cucumber. I am presently re-doing the amp boards and protection boards on another one so is the recommendation to set it at 25? I also have a 939 that does not seem to be overly hot but I don't think I have checked it. Should I go back and change the 50 back to 25 on the working one? If the one I'm working on turns out o.k. I'll probably go back and re-build the power supply in my good 1010. Does anyone know if the 939 is as prone to power supply problems as the 1010.
Thanks,
Duggey
 
Good question on the 939. I also set my old 1010 to 50mv, per MTF's recommendation and the manual specs back in 2011. I sold it shortly there after. I now have a 939 that I've set to the factory spec. of 50mv. Wonder if it should be lower also. Both units had MTF approved recaps using his parts lists, and troublesome transistors replaced.

BTW neither of the units get or got any more than lukewarm at low volume settings. I could put my hand directly on the heatsinks fairly comfortably for over a minute. But push the 939 hard and it does get hot. Not fried egg hot, but uncomfortably warm after 15 seconds directly on the heatsink.

Larry

Larry
 
I've just finished completely rebuilding my 1010 (shotgun'd everything on MTF's parts list plus the big PS storage caps and all output transistors). I set my bias at 25mv for each channel per this thread. It sounds great and when the 1010 is playing at conversation level volume, the heat sinks are just barely warm enough that you can tell the thing is turned on.
 
You must log in or register to reply here.
Back
Top Bottom