I have restored a few tube amps in the past. A few weeks ago, I came across the forum and decided to try an SX-626 just to find out how good it is. I got one from eBay for $40. It was well packed and delivered without any problem.

I am sure that I was lucky. The machine had very little dust inside, all lights are working, a few scratches on the outside. All pots seem to be OK. A bit scratchy at the beginning but long gone by now. Everything worked.

It sounded good right out. The FM stereo reception was OK. There is a loud background noise for local public radio due to HD spill over. I have fixed this problem with my Denon receiver by adding a filter after the IF stage. I studied the schematics and did some change. I have not totally removed the white noise yet.

I then decided to recap the power supply board and triple the original PSU cap (3300 uF). I also bypassed the output electrolytic capacitors (2200 uF) with a pair of 7.5 uF PP film caps. There are many other electrolytic caps in the signal path. But I decided not to replace them all at the same time.

After all this, I sat down and listen to it. Boy, was I surprised! This machine sounds close to my restored HK A300, Scott 222C, and Dynakit ST70. The sound stage is big and images very well. The high may need some clean up, particularly during loud passages. I don't know whether I can fix that. But this is a really nice receiver.

I have restored a few tube amps in the past. A few weeks ago, I came across the forum and decided to try an SX-626 just to find out how good it is. I got one from eBay for $40. It was well packed and delivered without any problem.

I am sure that I was lucky. The machine had very little dust inside, all lights are working, a few scratches on the outside. All pots seem to be OK. A bit scratchy at the beginning but long gone by now. Everything worked.

It sounded good right out. The FM stereo reception was OK. There is a loud background noise for local public radio due to HD spill over. I have fixed this problem with my Denon receiver by adding a filter after the IF stage. I studied the schematics and did some change. I have not totally removed the white noise yet.

I then decided to recap the power supply board and triple the original PSU cap (3300 uF). I also bypassed the output electrolytic capacitors (2200 uF) with a pair of 7.5 uF PP film caps. There are many other electrolytic caps in the signal path. But I decided not to replace them all at the same time.

After all this, I sat down and listen to it. Boy, was I surprised! This machine sounds close to my restored HK A300, Scott 222C, and Dynakit ST70. The sound stage is big and images very well. The high may need some clean up, particularly during loud passages. I don't know whether I can fix that. But this is a really nice receiver.

I think you have answered your own questions....

Mark,

One thing I did not mention. The machine as is had significant crosstalk between AUX and Phono. It turned out that Pioneer bundles the signal cables together. I separated the phono cables from the AUX. Problem is gone.

As for recap of the RIAA, preamp, and the amp, what is your suggestion? Should I buy Panasonic or Nichicon eletrolytic to replace them? As the output capacitors seem to be all right. Do you recommend to replace them?

How about bypass electrolytic capacitors in the signal path with film capacitor?

Or do you recommend the ultimate film replacement for all capacitors that are 3.3 or below? Any suggestions are welcome. I just don't want spoil the sound after spending all the money. So far, I used my spare or parts stripped from junks.

Mark,

One thing I did not mention. The machine as is had significant crosstalk between AUX and Phono. It turned out that Pioneer bundles the signal cables together. I separated the phono cables from the AUX. Problem is gone.

As for recap of the RIAA, preamp, and the amp, what is your suggestion? Should I buy Panasonic or Nichicon eletrolytic to replace them? As the output capacitors seem to be all right. Do you recommend to replace them?

How about bypass electrolytic capacitors in the signal path with film capacitor?

Or do you recommend the ultimate film replacement for all capacitors that are 3.3 or below? Any suggestions are welcome. I just don't want spoil the sound after spending all the money. So far, I used my spare or parts stripped from junks.

The recommended replacements are Panasonic FC or FM, and Nichicon PW or HE. All Electrolytics, with new...

Film caps add costs that may not be justified by the rest of the internal circuitry. Also film caps may not fit in the available space.

I have a Pioneer SX-3000 (circa 1973-74) that was a Military sales only version of the SX-626 it has a slightly different fascia and cabinet. It started out my wife's brothers who bought it new in Germany in a US Army PX he gave it to my wife for her 16th birthday. It's a sweet sounding little 27 watt receiver and the reason I now enjoy my Pioneer gear.

I have a 636 that I like very much.Big soundstage on it and very robust for 25 wpc.It was my first reintroduction into vintage and is a keeper.On top of all that it is a very attractive looking unit.

I like how distancing the signal cables inside the 626 cleared up your crosstalk issue. I'm going to do this next time I'm in my Pioneers.

Before recap, I decided to adjust the amp first. I found the B+ right at 59 volt. However, the voltage right before the coupling caps are all at around 25 volt. Significantly less than the 29.8 V on the schematics. The voltages across the 0.5 ohm are 13 mV and 27 mV. That translates to idle currents of 26 mA and 54 mA.

I could not find instructions on the service manual for bias and mid point adjustments. Can I assume that V1 and V2 are for the mid point (i.e. move 25 to 29.8 volt) and VR3 and VR4 are for the bias?

Could it be possible that C7 and C8 dried out? However, since mid points for both channels shifted, I don't rule out that the C21 is leaking. It puzzles me why Pioneer chose 50V cap for C21. It seems clear to me that rating is lower than the actual voltage it sees.

As recap is a significant amount of work, I want some opinion before proceed. I also do not want to adjust VR1 and VR2 just to find out that action causes other issues that need more effort to fix. My first class A amp kit went up in smoke before I can finish bias adjustment.

Check the Pioneer manual database for your SM. You'll find it as a sticky at the start of the forum page. Just follow the instructions to get there.

I have the SM. It does not say anything on adjustment. However, I found another thread specifically on this subjected posted by Mark. I will do some adjustment first just to find out what happened. It does not seem to be too dangerous.

I have the SM. It does not say anything on adjustment. However, I found another thread specifically on this subjected posted by Mark. I will do some adjustment first just to find out what happened. It does not seem to be too dangerous.


The transistors in the amp may have shifted in operaing points over the years. I hope the procedure you found is the 1/2 vcc setpoint one.

Yes vr1, vr2 are midpoint adjust and vr3, vr4 are bias adjust.

Without digging too deeply into C21, other than it is a turnon delay timer component for the amp, there is a 5.6k resistor but upping the voltage won't hurt. The OLD caps had better margins than the newer ones.

Recapping should be more of a time rather than cash investment, and even IF things went wrong badly, the transistors aren't expensive or hard to find modern replacements for.

Try adjusting the 1/2 vcc setpoints first and see how it behaves. After all this time things shift around. and 29.8v to 25v isn't that bad...

were the 13mv and 27mv readings on different channels? when adjusting idle current just watch the voltage that indicates the bias current carefully for a while, and don't swing it higher. all higher does it make heat. after a "knee" point increasing bias current doesn't lower distortion any further, it just makes heat.

The only clue we have is the total current draw to the amplifiers is 70ma on the schematic (p24 of 49 in the pdf). I would try a bias setting of 30 ma (15mv) each. Thus one is almost bang on, and the other a bit high.

Mark,

I adjusted V1 and V2 by turning both pots counterclockwise until they stopped. Now I got 29.5 V. Not bad. Since I do get the turn-on thump, could a bigger C1 help to reduce the turn-on thump?

Mark,

I adjusted V1 and V2 by turning both pots counterclockwise until they stopped. Now I got 29.5 V. Not bad. Since I do get the turn-on thump, could a bigger C1 help to reduce the turn-on thump?

It's probably lost value..... leave it be, the turnon thump could be something else that has drifted... when they are all new, THEN we worry about that....

Are you in the Continental USA? (I ask in terms of recommending parts suppliers...)

Yes, I live in the US. I am familiar with Mouser, Digikey, Parts Express, and MCM. Electronics is my hobby. I have rebuilt a few tube amps.

SX-626 was my first receiver. Bought it in the 70's, still have it, still sounds good to me, paired with a pr. of HPM100's. Particularly like the looks of the bronze colored face. Very unusual for the time.:thmbsp:

Vintage SS guys & gals, which do YOU think sounds better, Sansui 771, H. H. Scott
stuff, Fisher whatsis-500, or Harman Kardon? All I can say is that when I got my
771 and listened, I sold my Marantz 2235 faster than a speeding bullet!
Does anyone on AK care anything for original Luxman stuff? Just curious...
For the record, I enjoy my 771for 7 years now, on FM radio or with CDs
~ Pioneer PD 5700 or with one of four mid-'80s cassette decks.

Just a few updates:

I recap the whole receiver. The only think that I did not replace are the big power supply cap and two output coupling caps. The improvement in sound is not much. Maybe they need more time to break-in. The only important finding is that when I push in the low filter, the amp does not make big noises any more. I can still see the woofers popping out during turn on. They never made any noise before nor after the recap.

For the phono section, I replaced the output capacitors with Wima PET film capacitors and replaced the RIAA 0.033uF caps with a PS caps. The phono section sounds much better than before.

I also added one 12K resistor before VR1 and VR2 so that I can adjust the output voltage to 1/2 of the B+.

I also tried to parallel the output capacitor with an additional 2200 uF, 50V capacitor. The sound turned worse. As the original capacitors have different pin out, I did not replace them to check the sound of the new capacitors. DC offset for both channels are around 0.3 mV. I think the original output capacitors should be still good.

I also removed the 7 uF metallized PP film capacitor I added before. I want to listen to the receiver more before I put them back to compare the sound.

After listening to the receiver for one day, I decided to put back the 7.5 uF bypass capacitors (metalized PP film by Parts Express) for output coupling capacitors. The receiver sounds warm and mellow without them. Some may like it that way. But the film capacitors really bring out the transient and impact of some music.

Now that my SX-626 is fully recapped, output voltage adjusted, and bias adjusted, it sounds good. However, if I turn the volume past 12 o'clock position, the music becomes distorted. Granted it is pretty loud at that volume, it seems to be a little bit shy of the 27 watts claimed.

I tried to connect the preamp out with a DIY amp and confirmed that the distortion is from the amp section. As my low power (+-27 V supply) amp has no problem handling the same volume and louder.

Could there be something else (e.g. driver transistors out of spec)? Or is this very typical for capacitor coupled amplifier with low feedback? Will it be better if I increase the feedback ratio? Or change the output transistors to 2N3055?

At this point I would be looking at the output on BOTH channels on a scope to see where it is clipping, and whether or not the distortion being heard IS clipping or something else.

I think the output cap should be upgrated. You may take a look on post 1900 here:
http://www.audiokarma.org/forums/showthread.php?t=5634&page=127

there is part of a test report for the SX-626 that I used to illustrate a problem on another cap coupled amplifier.

You will find this thread interresting about the output caps in your amp.

http://www.audiokarma.org/forums/showthread.php?t=157951

Thanks for the suggestion from ecluser, I decided to replace the output capacitors. However, the only capacitors I have are a pair of Nichicon rated at 2200 uF, 50V, 105C. I removed the 7.5 metalized PP film bypass capacitors.

The sound of the amplifier is a bit better. Compared with the old capacitor, the new ones have more open sound. I don't feel the need for film bypass capacitor. However, distortion at high level does not improve a lot. Maybe the capacitors are too small. I will order some bigger ones in the future. Maybe 10,000 uF, 63V ones.

This whole thing reminds me when I was playing around with the feedback loop of my HK A250 tube amplifier. The sound of the amplifier moves from shouting to dry when I increased the amount of negative feedback. There is a sweet spot at which the amp sounds nice and dynamic. As this receiver sounds really loud at 12 o'clock, a bit more negative feedback should be OK.

I would keep the 7.5 uF PP cap.

Look at the feedback connection in this amp, around the output cap. The 8.2k // 47pF has a cut-off frequency of ... 400kHz. That means the 15k and 8.2k resistors are effectively in parallel within the audio band (because the 2200 uF is "supposed" to be a short at those frequencies). This is the main feedback resistor which, in combination with the 150 Ohms resistor in the first stage, gives a net voltage gain of 35 for this amplifier. But this doesn't hold at very low frequencies because the voltage on the - side of the cap is not in phase with the voltage on his + side for any frequencies where the impedance of this cap is comparable to the impedance of the speaker. In fact, the voltage on the - side of this cap is very unpredictable when you take into account the reactive load of the speaker at low frequencies.

This feedback topology is load dependent and, in my opinion, is not a bright idea. I would connect the 15k on the other side of the output cap. A drastic change, I know! This way, the output cap would be completely inside the open loop of the amp, and the feedback would compensate for distortion from the cap. This would change the polarisation point of the transistor on the first stage, but not by a great amount.

You may try it in one channel only, and compare the sound between both channels with a mono source. Does it make sense to you?

At very high frequencies, the 47pF cap shunts the feedback resistors to reduce the gain, and improve the stability of the amp. I think any large cap like the one on the output are more inductive than capacitive at 400kHz. It is a good idea to shunt them.

After studying the circuit and a few other capacitor coupled amplifiers, I decided to change the feedback resistor after the output capacitor from 8.7K to 4.7K. The sound becomes tighter. I will listen to it more to determine the long term merit.