Pioneer SA-9800 current limiter resistor

AFJ

New Member
Hi all, I recently acquired a SA-9800. It is an "S" version from Europe. It looks in very good nick and the circuitry does not appear to have been touched. Now I am planning to rebuild the power supply and recap it.

My electronics knowledge is basic, but I am good with soldering iron and have successfully recapped a SA-8800

Things I am planning to do are as follows:

1) Use the rebuild list created by solution room and overseen by markthefixer - http://www.audiokarma.org/forums/index.php?threads/sa-9800-re-work-begins.409557/page-3
I will also replace the parts in the rectifier board as well

2) I may upgrade some of the amp board transistors as found in Lee's link, this list was approved by Echo - http://audiokarma.org/forums/index.php?threads/pioneer-sa-9800-restoration-and-upgrade.420407/page-2

3) I want to upgrade the AWR-194 power supply board to a AWR-196 (as the GH version) - http://audiokarma.org/forums/index.php?threads/sa-9800-in-need-of-some-tlc.609273/page-2

Questions I want to ask for the moment are as follows:

1) I will replace the filter caps with 661-EKMH800T183MB80T - these are 180000uf / 80V - I have read that a limiter resistor upgrade will be a good idea with this - I want to know what are the limiter resistors in this power supply and what it should be upgraded to?

2) What is the function of the 2 chassis mount wire wound resistors in front of the transformers? - they are called R1 and R2 in the schematics and they are R1- 150 Ohms (5W), R2 - 220 Ohms (5W)
 
2. R1 and R2 are part of the +/- 32v power supply regulator assemblies, and they dissipate heat that would strain Q11 and Q12. These are actually discussed at the bottom of the page you linked to in (2).

1. Limiter resistors do not exist in this unit, please provide a link to what you saw. They are used in units that have Toroidal transformers to tame the inrush current that would possibly pop breakers. The square frame E - I transformers do not need them, they saturate at a maximum power transfer level. Toroidal transformers, for our purposes, really don't saturate or limit. Your change from 15,000 to 18,000 is negligible, as these old caps had a -20% +80% tolerance.

If the original amp board power (driver) transistors (Q33, Q34, Q35, Q36) are working, leave them be. They are special and crazy fast, and are very hard to find replacements.
Some of the SMALL transistors on the amp (Q7, Q8, Q22, Q12) can be beneficially upgraded to to-126 packaged transistors for heat purposes.
These small transistors to be swapped out MAY have a THICKER center collector lead than the emitter or base leads on either side of that collector lead.
I have worked with these NSA amps quite a bit, few others have that understand their ramifications in the design.
Screwups have been accompanied by SHRAPNEL flying as they blew up, and sometimes they just quietly miss their distortion numbers by a factor of TEN. ("HELLA SLOW")
 
Hi Mark, thank you for the response,

1) I have uploaded an image of the R1 and R2 I'm talking about.

2) Here is a link to a page where there is talk about upgrading the limiter resistors and upgrading few more transistors - http://marantzhallo-fi.blogspot.com/2016/10/stefans-pioneer-sa-9800-integrated.html. This link shows some impressive numbers with the THD though it does not go in to too much details, but I can see on the images that the drivers they have swapped are - 512-KSC2073TU and 512-KSA940TU_Q. Thease are not the much faster Toshiba 2SA1837 and 2SC4793 recommendations I found on this site. It will be interesting to know your thoughts about the link.

3) About "screwups", I have read a lot about it in this site. So all my work will happen in stages and starting with the power supply. I have ordered an Atlas DCA55 to help with the transistors
 

Attachments

  • Untitled drawing (2).pdf
    24.7 KB · Views: 36
I myself could not see the attachment, so I uploaded an image
Untitled%20drawing%20(2).pdf
 
Sigh, those are EXACTLY the R1 and R2 I was talking about. They ARE part of the +/- 32v regulator, OFF the board, connected to the power supply board by pins 12,13 and 9,15.


As for the slow driver transistors on the link they say:
During the rebuild several driver transistors were tried, resolving on the fact that the NSA circuit requires a higher bandwidth operating device of at least 95Mz to obtain its rated distortion factors
So the picture showing the slow drivers had to be taken during the earlier part of the rebuild, where they found out the hard way (nasty surprise) that the drivers were too slow as evidenced by their published results of accurate testing of it's specified distortion numbers.

It's a very good article, with a very good rebuild, and it shows many indications that the AudioKarma threads on the SA-9800 had an influence upon various choices.

There even may have been a period of time in AK where somehow the association of the drivers and the a940 and c2073 parts numbers got loose into the world without being tested first. Since I maintain (private) records of my errors and screwups , I checked and nothing showed up in my sa-9800 work. I haven't backtracked the threads yet, where I would edit out the errors (WITH attribution) or urge others to correct errors in their posts to avoid repeating mistakes. That task gets even larger when ALL NSA amp threads have to be checked for slow drivers. Added to that is the fact that those fast drivers are becoming unavailable and you have a hella storm coming.
 
"many indications that the AudioKarma threads on the SA-9800 had an influence upon various choices" - that is quite obvious

Last night I eye balled the power-amp and power supply boards before ordering parts. I can hardly see any indication of heat damage. So apart from the cap replacement and few resistors, the other work done on it (as per solution rooms list and the above recommendations) will be prophylactic/preventive.

I have the following questions from the observations:

1) R22 - appears to be drifted, but I have checked this in-circuit - If it needs replacement, does it have to be cement cased or can it be replaced with high precision wire wound resistors (71-RS005560R0FE12 or 588-45F560E) - this is because I need to order all the potential parts beforehand even though they may not be replaced, because this is cheaper than paying for postage at mouser again.

2) Is output transistor replacement prophylactic?
 
If you change from a 50mA (8vac) incandescent pilot lamp power indicator to something else, R22 WILL change. Drift in the lighting circuit isn't a problem.
Wire wound can be used, it is a very non-critical circuit. Cement cased take the heat better.

are the original outputs behaving? If so just regrease them.

IF the replacements were higher power rated units, they are still working into the same heat sinks, with the same power supplies - thus no real advantage.
 
If you change from a 50mA (8vac) incandescent pilot lamp power indicator to something else, R22 WILL change. Drift in the lighting circuit isn't a problem.
Wire wound can be used, it is a very non-critical circuit. Cement cased take the heat better.

are the original outputs behaving? If so just regrease them.

IF the replacements were higher power rated units, they are still working into the same heat sinks, with the same power supplies - thus no real advantage.

The original outputs are behaving so I will leave it alone.

With regards to the R22 - If I consider LED pilot lamp conversion.

1) If I do a reverse calculation from the voltage/current and at the pilot lamp - it appears that the voltage output from the T3 transformer is 76V, this is doped to 59.5V by the 330 Ohm R23. This is further doped to 8V/50mA by the combined resistance of 1030 Ohm of R 21 and R 22 which has to be at least 3W.

2) If I want to drop the voltage to 3V/20mA for an LED, I need a 56.5V drop across R 21 and R 22.

3) For this voltage drop I need a resistance of 2825 Ohm across R 21 and R 22 which should be above 1.5W.

4) After the new resistors a diode can be used before the LED.

Is this calculation correct?
 
Last edited:
All above calculations are based on assumptions. It will be helpful if an expert can confirm the T3 transformer output.
 
Replacing the drifted R22 and adding a 250-275 Ohm (1/2W) resistor after that will drop the voltage/ current to 3-2.5V/20mA for a LED. This approach is cheaper than upgrading the R22 and R21.

I will order the parts for the rebuild today
 
Back
Top Bottom