SA9900 Restoration Challege

Fhamre

Super Member
Picked up a nice 9900 dead on one channel. Traced problem to a bad cap on the control board.Working great. Then following all the recommendations in this forum I went through the unit recapping and replacing all the transistors ie. 2SC1451's,2SA809's,2Sa725's etc and the odd diode. Did one board at a time. All went well. Saved the amp board for last. Triple checked my work and put it back together. Using dim bulb without the output transistors installed the relay clicks and no flashing bulb. Carefully replaced the outputs with new thermal paste and mica in the correct location as per my diagram. Checked to ensure no broken wires etc. Dim bulb flashes when powered up.
Any suggestions? Don't want to fry these original outputs.
 
What watt bulb are you using? 100 to 150 watt should cover it.
It's takes a second or two for the big filter caps to charge. If a short, the dim bulb will light brightly and stay lit brightly protecting the set. If OK she lights up brightly and then goes dim after a few moments.

Is there a "Tickle the Dragon" Procedure for the Outputs settings? If so Idle trim VR's to zero (CCW) all the way. Then fire it up on the DBT.
 
Last edited:
Yes there is a "Tickle the Dragon" Procedure.
It may require a 8 ohm dummy load to set the idles. That set is beyond my experience so help from the Guru's is needed. The adjustment is on page 19 and 20.
 
Yes I have the service manual,thx. Using a 100 bulb.
Not sure whether I should be tickling the dragon yet as I may have a short or transistor incorrectly installed. Although I painstakingly reviewed my work as assembly and dissembling the amp board is a lot of work.Don't want to risk breaking wires etc if there is something else I am missing.
 
I wish you guys would just ASK about what you intend to do BEFORE doing it...

In a word, Your procedures are all wet.

There is enough variation in these amp topologies to make a set procedure nonsense.

Without the outputs in place in THIS type of amp there is NO DC FEEDBACK. The closest similar analogy is the sx-1010 amplifier. The on the pre-driver and driver stages the positive and negative halves are INDEPENDENT, NOT CONNECTED - thus UNABLE to tap DC feedback - and half of the stage can fail To get DC feedback and balance the amplifier, Two guys have now done the "resistor trick" to measure and correct the base drive voltages for the output transistors BEFORE connecting them.

You make no mention of the power supply, it's voltages, NOR OF SETTING THEM!!! THERE ARE FOUR POWER SUPPLY SET POTS DAMMIT!!!!!

And for general information - a DBT is ONLY useful to prevent damage to the output transistors - you are perfectly able to send EVERYTHING ELSE up in smoke!!!
 
Last edited:
So how would I go about taking those voltage measurements and adjusting the the power supply voltages given that now the dim bulb is flashing. Prior to the work I did to the amp board the unit was functioning . I had planned to check and adjust once the amp board was done.
 
did you replace the trim pots on the amp board and if you did, what did you do about the initial resistance settings of those trim pots? A bad / too high setting can be causing this.
 
9900 uses a simple totem-pole bias scheme, so minimum setting on the bias pot is the safe zone. Whether that's clockwise or counter-clockwise is up to you to figure out.
 
You need to check that they are ok, we have started to recognize the idle current pots as ticking time bombs. UNCLEANABLE ticking time bombs.

Their construction does not lend to effective cleaning, in fact I expect cleaning to induce damage.

Check ohms across them, as they are rotated, if they go higher than 470 ohms - even momentarily - that's BAD.

The initial setting of the pot for minimum idle current (done after ANY work) is zero ohms, and while this is probably fully CCW, use the ohmmeter to BE SURE. Because the wrong direction will be full resistance and excessive idle current. Probably damaging levels of excessive idle current.

That is why we ask people to publish their list and intentions, to point out exceptions and additions specific to what list they are using, and the pitfalls to avoid.
 
I just checked those pots they are working as intended. I will set to them to zero and go from there?
I have the dummy loads and the rca 5.1 K ohm resistors.
 
So I did the junction point and idle current as per the service manual. Seemed to go well. At this point all settings done with dim bulb.
 
Found this previous post of Mark's. Will now try without dim bulb.

on DBT:

before power, VR3, VR4 fully CCW
turn on power, warm up time
pots on the power supply adjust to +58 and -58v
read p.s. pin 4 voltage to ground, set p.s. vr1 to +58v
read p.s. pin 5 voltage to ground, set p.s. vr2 to -58v

All the 8.xxx stuff is in the manual. Page 20 of the paper manual that was scanned, pages 18 and 19 of the 61 page pdf file.

8a obvious - has to be open...
8b VR3, VR4 fully CCW
8c warm up time
8.1 ignore, leave it at normal
8.2 ignore, volume control to minimum, that 's the infinity symbol when fully CCW
8.3 ignore
8.4 ignore
8.5 DO
8.6 DO

WHEN we remove the DBT, the above will change and has to be redone.

Tickling the dragon's tail:
This is a TRIAL set of the idle current for each channel, to tease out any faults while STILL on the DBT. You might NOT make it to 0.050VDC on each channel - that's fine as long as there is SOME idle current and it is adjustable.
The idle current will cause the bulb to glow a bit brighter, if you have strange things happening, you probably need some more wattage on the DBT.
FULL ON BRIGHT would be a fault, and you would stop.

8.7 DO
turn idle current VR3 adjust BACK to fully ccw
8.8 DO
turn idle current VR4 adjust BACK to fully ccw


now, plugged DIRECTLY into the power line:

before power, amplifier VR3, VR4 fully CCW
turn on power, warm up time

pots on the power supply adjust to +58 and -58v
read p.s. pin 4 voltage to ground, set p.s. vr1 to +58v
read p.s. pin 5 voltage to ground, set p.s. vr2 to -58v

then follow the procedure


8.1 ignore, leave it at normal
8.2 ignore, volume control to minimum, that 's the infinity symbol when fully CCW
8.3 ignore
8.4 ignore
8.5 DO, target = 0.000VDC
8.6 DO target = 0.000VDC
8.7 DO target = 0.050VDC
8.8 DO target = 0.050VDC
 
Adjustments done. Seems stable,sounds great. Will continue to play and monitor before reassembling.
Appreciate the input and guidance.
 
Back
Top Bottom