Rebuilding a realistic SA-100B HELP NEEDED!!

[KimLaroux]

Quote:

Yes there are 2 R43-just shows you what Japanese early circuit layout was like -As far as R43[at the top] this is standard means of lowering both current and voltage to the input pair of transistors to isolate noise from the output circuit as the output active devices in most amps are not low noise.
This doesnt matter so much at the output because the ratio of signal to noise is very low. But it does matter at the input as even the tiniest will be amplified and heard at the output.
Look at Hi-Fi power amps and voltage circuit feeds between input and output .You will find this is the simplest approach a more intelligent approach is to have separate power supplies to each half.[stabilized] The more money you spend the higher quality of circuit[usually]
So even R43-2 isnt a filter of any sort that we are talking about -only a cheap way of stopping noise from the output.

 
Isn't this the very definition of a filter? 

 

[duncan1]

By noise I mean hum 50/60HZ+ harmonics . The real definition of a filter is a resistance/ capacitance/ inductance connected in series/parallel to reduce or cut -off  a specific frequency so that that frequency is lowered to a very low level.There are low-pass/high -pass filters in just about every branch of electronics.What that 500uf capacitor does is lower the ripple voltage from the power supply in other words as I said 50/60HZ -HUM .It  will NOT lower higher frequency noise[meaning above low audio frequency] What he complains about is usually brought on by noisy active devices and the era that circuit relates to 1966-onwards they were noisy so were the capacitors and carbon resistances no amount of love for carbon resistances from that period will remove the noise.
So the filter he requires is a series/parallel  combination of LOWER values to remove/lower a higher  frequency .It is NOT inserted onto the power rail but in contact with the music signal path.This can be adjusted for the amount removed and frequency by altering the values of the components used for the filter.

 

[scootsit]

Hey guys, thanks for all the insight, and sorry for the lack of update. I've been busy at work. I'm going to try to find some time to play later this week. I'm thinking those output caps (500uF) may be my next candidates for replacement. I have some 1000uF caps that may work perfectly, I'm assuming bigger is better fi they're filter caps.

 

[scootsit]

I took a look, C23 and C24 actually don't match. C23 (I think, the board is not labelled at all) is 200uF, while C24 is 470uF. I'll replace both with ~500uF or higher ASAP and report back.
 
I also tried bridging the input to ground, there was no change.

 

[duncan1]

If the noise wasnt reduced by shorting the input to ground proves it is in the circuit the noise is coming from
With the different value  found for C23 its probable it has been a replacement.which given the age is no big surprise and yes your right replace them with 500uf or higher just make sure the working voltage on the ones you buy is at least 20V above the voltage rail and when connecting to a negative voltage rail that the neg. of the capacitor is taken to that and the positive is -grounded/earthed.
 

 

[scootsit]

Quote:

If the noise wasnt reduced by shorting the input to ground proves it is in the circuit the noise is coming from
With the different value  found for C23 its probable it has been a replacement.which given the age is no big surprise and yes your right replace them with 500uf or higher just make sure the working voltage on the ones you buy is at least 20V above the voltage rail and when connecting to a negative voltage rail that the neg. of the capacitor is taken to that and the positive is -grounded/earthed.
 

Got it.
I'll probably recap the out put, which is currently at 500uF. Any reason I couldn't bump them up to 1000uF if I bipass them? I should get better response, right?

 

[duncan1]

Your thinking is right. Of all the electrolytic   capacitors fitted your your amp and for that matter ANY amp that has a capacitor output changing them will make a major difference to the fidelity of the amp and yes bypass with as good as you can get poly-prop -100NF-film[not metallic coated] and definitely not those X2/Y1 etc mains suppression types audio quality only.
The history   behind  this is that in those days -split-rail power supplies were not used yet and the output had to be isolated from the earth. The Quad 303 used the same. This was THE amp for the Quad 57 electrostatics which as you know present a "hard" varying and capacitance impedance for a power  amp. In other words it was "indestructible" But dont buy one as you would "fall asleep" listening to the 303. I did own one  in 1971- sold it a year later.I still have the official Quad circuit diagram.of the 303 and its pre-amp the 33.

And yes you would get a slightly lower [uplifted] frequency response with a larger value. And make sure the electrolytics are good quality .

 

[scootsit]

I did a lot tonight, and am very, very frustrated!!!
I'll post details tomorrow.

 

[scootsit]

I replaced the output caps with 1000uF nichicon px caps, bypassed with cheap film caps. I know none of that is top of the line, but they should work. If I get everything working, I may go with better stuff. I replaced c23/24 with 50v rubycon yxg 820uF caps. I got them from this older store, so they aren't brand new, but certainly within the last decade or so. Originally, it didn't work at all, I somehow blew a fuse. Also, there original caps were labeled with the + side marked, not the -, so I initially installed c23/24 backwards. I powered it on for half a second before noticing the mistake. I'm guessing the fuse had already blown, because the safety valve didn't blow on the caps. The noise is quieter, but not to an acceptable level.


To clarify, my main amp is a recapped Sansui 5000x. Most of my amps are from pre 1980. I expect a bit of noise. This is not a tad of noise, it's really loud. Nearly as loud as the music at max volume. I've noticed that the cap voltage ratings are exactly what they need to be. So the 6.3v rated caps have over 5v across them, so the potential for problems seems amplified. Given the rule of thumb that I learned, I'm doubling those voltages for the ratings on the replacements

 

[duncan1]

If the rated caps at 6.3V have over 5V across them somethings wrong .You have 2   addresses US is 110/120V-I think Australia is 230V -have you checked the tappings on the transformer? I have come across several faults in the past where the faulty amp has had its tappings changed to a lower voltage to in the owners words "to make it sound louder"- That means an over voltage is applied all through the amp and it meant all the caps had to be exchanged even if they werent blown if not raise the tapping ABOVE what it normally is set for in that country and see if the noise is reduced. Thats a standard test.
Is this very loud noise actual "white noise" in other words a rough gravelly noise or is it HUM?? 
Running the caps in reverse voltage could have partially damaged them even if it doesn't show.
The caps marked + instead of - isn't standard industrial marking all the Japanese ones I have show the neg. Al;though some OLD ones show the pos
Another fault that can occur giving the same symptoms is a faulty rectifier- yours has 2 rec diodes if a larger voltage than the working voltage of them is applied then that could cause hum/noise or if current carrying capacity is not high enough. Modern rec. diodes have low forward resistance /more efficient and smaller-worth a check .If they are in an old fashioned block with tabs instead of either single diodes or black plastic encapsulated then they could be faulty.Modern diodes are usually reliable.  If you dont find any fault. Then without it in front of me I am running out of ideas.Best of luck.

 

[scootsit]

The caps rated at 6.3V have ~5V over them. My thinking is that this is a radio shack brand amp. They speced it with the caps they sold in their store. It uses exclusively 6.3, 10, 16, and 30V caps. All are pretty standard. Each is as absolutely close to the actual voltage as possible. The parts of the circuit with 25V across them use 30V caps, etc, etc. I learned you double the voltage for the rating, or close to it anyway. So, as I rebuild it, I am doing so.
 
I doubt the amp is australian, it's from radio shack, which is based soley in the us

The noise is white noise, no hum.
 
I have removed the transformer before and tested all voltages. They were consistent with someone else's SA100B, whose amp was working properly.
 
These are old caps, so yes, they were labelled on the + side. It's weird. Even other caps from the 60s and 70s I've seen have been properly labelled, but again, radio shack has a reputation...
 
Regarding the diodes, they were in these big round cases, unlike anything I've ever seen before. I've already replaced them with contemporary equivalents (1N4004).
The PS cap was speced at 1000uF, I replaced it with 2200uF, so it should be plenty large.
 
I have tried bypassing all of the caps with a poly cap, seeing if that helped the noise. It did not.
I'm thinking I need to replace all of the electrolytics, then go from there.

 

[duncan1]

I really admire your determination have to say I am the same way too. I never give up.The   less said about radio shack the better.There are many companies that supply quality caps/resistors. You cant go wrong replacing all the caps as that at least would eliminate them from causing the fault.
Also check the value of all the resistors they can be very noisy from that period I have come across resistors with either highly reduced values due to heat or highly increased  values as little load is on them.The RF section of a lot of old radios and communication receivers  would go dead because of an increase of resistance.If any heat up change them to a higher wattage.

 

[scootsit]

Quote:

I really admire your determination have to say I am the same way too. I never give up.The   less said about radio shack the better.There are many companies that supply quality caps/resistors. You cant go wrong replacing all the caps as that at least would eliminate them from causing the fault.
Also check the value of all the resistors they can be very noisy from that period I have come across resistors with either highly reduced values due to heat or highly increased  values as little load is on them.The RF section of a lot of old radios and communication receivers  would go dead because of an increase of resistance.If any heat up change them to a higher wattage.

I still think most any of this would cause hum, not white noise, right? What is likely to cause white noise?

 

[duncan1]

White noise can come from old carbon resistors especially when heated/old capacitors and active devices-BJT-transistors/Mos-fets etc. So check if the 2 BJT  in the first and 2ND stage are  2SB440 someone could have changed them. The  problem would be that if you thought of changing them the new modern ones could cause instabilty due to higher gain. The ones fitted are not far from the first BJTs  Japan made.But if you feel up to it I could recomend replacements with the same gain-its just a case of three leads being soldered.  Hum is nearly always a product of the power supply.  I used to work for Brit-telecom in the UK the old dial phones had carbon granule mikes they became very noisy and had constantly to be replaced.