SX1980 Voltage Regulator / current source question

Current sources built, tested (both outputting 2.2mA) and installed and working. I notice the + / - 80V's seem to ramp up now over a few seconds rather than immediately, assume this is a characteristic of the circuit.

Transistors used...
KSA1013YTA (PNP)
KSC2383YTA (NPN)

Something still wrong with the low voltage regulator(s) so I'll have a bit more of a poke around there and see what I can find.

I Must be almost there.... fingers crossed.
 
At least you seem to grasp the importance of the correct connection arrangements of emitter, base and collector.
Five chances in six to get it wrong, one chance in six to get it right.

Ok, now you made me look at the 13.5v & 80v regulators. And I immediately found some discrepancies schematic to reality.
Which I double checked physically on one of my boards (yes - I have a sx-1980 power supply board)
There is a resistor from the +80v regulated power line to the base of Q201.
This Q201 base point is the SAME topology as the current sources you just installed. It provides a current to operate the regulator.
IF you have the ORIGINAL resistor you are OK. The original resistor is R201 a 22,000 (red red orange gold) ohm 1/4 watt resistor.
It provides (80v - 14.7v) 65.3v / 22,000 ohms = 2.9 mA of current to run the 13v feedback control system.

BAD schematics show 2.2k (red red red gold) which would make the feedback system heat up and fail - maybe within your time frame.
The feedback system would have to suck down almost 27 mA ( (80v - 14.7v)= 65.3v / 2,200 ohms = 29.6 mA - 2mA) to regulate.
That's through Q202 and D207.
At 30 mA D207 at 6v has about 0.18w to dissipate, Q202 is 8.7v and 0.261 w. 0.44 watt total.

The 22k resistor is fine if there, although 1 CCS from R202's low side to q201's base would probably work if the ccs's will work with just a 10v differential.
 
Last edited:
The plot thickens...

I've today gone over the 13.5v, 8v, 5v regulator with a fine toothed comb, checked all transistors, resistors etc, checked for shorts, nothing apparent.

I checked R201 and its the original 22k.

I removed Q203 and checked the voltage at collector... 35V!
Same at Q202.

Any thoughts? Could I have something wrong with the transformer? Is it worth changing the two diodes D203 / 204? Assume they are rectifier diodes?
 
You can try (1n4004) , but the 35v is an UNLOADED voltage. Try a 250 ohm resistor across the cap (100mA 2.4w load), it has to be 3 watts at LEAST. Then see what voltage is a result. It should drop down to the expected 24 to 26 volts.
 
I have removed all real loads, and placed a 100 ohm resistor between pins 13 (13.5V) & 10 (Gnd).

It tries to regulate but the voltage keeps wandering up and then sharply dropping down in a peculiar manner.

Q201 gets very hot. Q203 does not get at all hot.

I assume that R202 should be 3.3 ohms as per manual

I have measured AC input voltage on pins 19 & 20, getting just over 21V in each case, so I suppose the transformer is not to blame.

I will change the diodes and also that 22k resistor for good measure, and report back..

Thanks for your assistance on this Mark, it really is appreciated.
 
The 100 ohms between pins 13 and 10 is fine. A reasonable load.

q201 is the darlington DRIVE transistor, and isn't supposed to get hot, Q203 should be the hot one.
methinks you still have a transistor mis wired.
The thing is, that Q203 and Q205v are TO-220 transistors and HARD to miswire while still being heat sunk.
Which transistor(s) show damage after the regulator fails?

List EXACTLY which transistor numbers (prefixes, SUFFIXES and such) you are installing at each position.
Q201 13.5v darlington drive for series regulator old-2sc945a, new = ?
Q202 13.5v series regulator feedback element old-2sc945a, new = ?
Q203 13.5v darlington power pass series regulator old-2sd712, new = ?
Q204 8.0v darlington drive for series regulator old-2sc945a, new = ?
Q205 8.0v darlington power pass series regulator old-2sd325r, new = ?

Look for other components heating up during operation, that energy's gotta go somewhere.

Posting pictures of that board location to double check the installation would be a good step.



I don't know if this has been explicitly said to you, so here it is:

EACH transistor's lead layout has to be identified by IT'S downloaded data sheet,
there is NO CORRELATION between the old transistor lead layout and the replacement transistor's lead layout.
NONE NADA NNN AAAAA DDDDD AAAAA NNNN OOOO NNNN EEEE !!!!!!!!!!!

It's hard enough getting an operating specification match, it is IMPOSSIBLE to get a lead arrangement match as well.

Then you install as directed by the transistor symbol on the board.

I will eventually ask you for voltage readings to pursue this problem,
but I have no confidence yet that that when I ask for q201's base voltage or Q202 collector voltage I will actually get them.
 
Mark, I am sure no offence intended, and none taken, as of course you don't know me from Adam. As one of my comedy T-shirts will attest, I am not a complete idiot, some parts are missing. :banana:

As requested:

Q201,202,204 - All now KSC945. Which seems to be the same as 2SC945A, except base & collector swapped.

Q203,205 - Both now On Semi D44H11G

Some pics attached. Ignore R201 which has since been replaced.

Below readings are with 100ohm load between pins 13 & 10.... nothing connected to pins 15 & 16.

Q201 base voltage ~17.5V at initial switch on (reduces)

Q203 base voltage ~17V at initial switch on (reduces)

Q203 emitter / pin 13 voltage ~16.5V initial switch on (reduces)

Q201 / 203 collector voltage - 27.5V (stable)

On feed side of R202 - 28.1V (stable) - I note this is a couple of volts over spec.

Pin 13 Voltage does drop down to ~13.5V after a few seconds but Q201 getting excessively hot and I'm shutting down before damage occurs.

I am wondering if my transistor combinations are not correct.

Worth noting - original failure was R202, Q201, Q202 & D207 with R202 visibily scorched.
 

Attachments

  • 001.JPG
    001.JPG
    67.2 KB · Views: 12
  • 005.JPG
    005.JPG
    60.6 KB · Views: 11
  • 011.JPG
    011.JPG
    43.7 KB · Views: 11
Just taken a reading of 16V across the 6.2V zener, so I think I can safely say that has failed. Will substitute and see what happens
 
Found a mistake - I had accidentally put a 24V zener in. I had both ammo reels on the table and must've grabbed the wrong one. The result I now get the correct voltage 13.5V at pin 13, but Q201 gets even hotter and starts to smoke after a few seconds, so I'm none the wiser as to finding the fault.

Q201 & 202 are definitely KSC945 though and I've tried changing Q203 to MJE15032 but that hasn't helped.
 
New voltage measurements:

Input (supply side R202) 28.5V

Q201 Base 14.9V

Q202 Base 6.7V
Q202 Emitter 6V

Q203 Base 14.2V
Q203 Collector 28.1V
Q203 Emitter 13.4V
 
Might have found the problem here. There seems to be a bad connection to Q203 collector. Could this have the affect of making little Q201 do all the work?
 
Probably. I just repaired a JVC R-X60, it was the same sort of problem, intermittent op as the devices get so hot, packed in there so tight, that the heat gets to the solder joints and opens them up over time.
 
I seem to have cured it! Q203 remotely mounted via flying leads and a bad solder joint onto the PCB. I'd never looked at it closely enough it seems. Also those following this saga will note I had similar problem with the -80V and I found a suspect solder joint there, too. So now with that taken care of and the CCS's installed, giving nice stable voltages, hoping I can put her back together.
 
You can have a lot of problems with traces on these power boards at times so maybe next time if you suspect something or a bad connection, try checking the resistance from component to component. You should get very near 0 Ohms maybe 0.5 Ohms depending on your test leads. Anything high is a definite pointer to a bad solder joint or damaged trace.

IIRC the smaller transistors can get very hot as you saw if the path to the larger transistor is damaged/intermittent as then, as you said, it has to do all the work of the larger one and the current required is, in effect, dragged through the little transistor. If it gets hot but the voltage remains as expected, that's another good indicator of bad traces or solder joints.
 
Back
Top Bottom