SX-1980 power supply problems
- Thread starter txturbo
- Start date
You should be able to calculate this as a ball park number knowing all the loading calculations that could/should have been done previously as part of a re-design effort.
And you are going to power up the new PS unit with the 1980 ckt's as your loads? = eek. And if (hopefuuly not for your sake) there is one or more faults, such as a over voltage condition, then what, smoked ckts. I do not recommend , in anyway possible, to do it this way. Suggest to use some load resistors( based on nominal loads) instead and I guess you will need to remove the supplies to the PA o/p bjt devices to isolate them as well.
I 2nd that, sometimes it is just easier to chop them out and remove one pin at a time rather than pulling out PTH barrels. If possible heat up all the pins at once, using a copper wire or other techniques.
Good luck.
Last edited:
You need plated through holes because of the ground plane, it's technically a 4 layer board.
It can be a pain to pull parts without a proper desoldering system.
The circuit was not redesigned so loading calculations were not done. It's simply a combination of previous mods in a new board with some updated features.
And I was just looking for the idle current, it's in many service manuals. It's useful for setting a trip current when using a variac with a trip setting. honestly, Idid not even look in the manual, it may be there.
I do not know what would happen if I try and power up with the new board because the receiver I am using for a test was dead when I got it and it looks like more than one section needs work. I will try and craft up something for the supply to chew on. We have an old power supply test stand at work that has a bunch of stuff in it that may be useful for setting up a test jig.
Last edited:
pin 17 +80v regulated 30mA 2.66k ohms 2.4 watts
pin 9 +34v regulated 64mA 530 ohms 2.2 watts
pin 14 +13.5v regulated 250mA 54 ohms 3.375 watts
pin 16 +8v regulated 100mA 80 ohms 0.8 watts
pin 15 +5v regulated 80mA 62.5 ohms 0.4 watts
pin 18 -80v regulated 30mA 2.66k 2.4 watts
pin 11 -34v regulated 64mA 530 ohms 2.2 watts
pin 8 -30v regulated 3mA 10k ohms 0.25w
pin 9 +34v regulated 64mA 530 ohms 2.2 watts
pin 14 +13.5v regulated 250mA 54 ohms 3.375 watts
pin 16 +8v regulated 100mA 80 ohms 0.8 watts
pin 15 +5v regulated 80mA 62.5 ohms 0.4 watts
pin 18 -80v regulated 30mA 2.66k 2.4 watts
pin 11 -34v regulated 64mA 530 ohms 2.2 watts
pin 8 -30v regulated 3mA 10k ohms 0.25w
Thanks gentlemen..
We have two of thees at work collecting dust. They are part of an old test fixture that was used to test custom power supplies. There are a bunch of other load resistors and such to go with it. It may be a bit overkill for this. Anyway we should be able to rig something up.
Good call on the lab power supply(s) to test the receiver first. I am sure we have one that will do up to 34V but I am not sure about the 80V.
Very, very well done. I have been offline for a few days and the progress is great to see. I am jealous in wishing I had the same drive you are displaying. The neatest part is the incorporation of the members ideas and advice. So far a well executed plan.
I had control of a BUNCH of something similar, but they dumped the heat to water, when I worked for Gould Lab's Battery Test and Development Center.
PDP-11 Computer controlled cyclers that charged and discharged batteries under test.
14 cyclers with hundreds of data collection channels , ever hear of CAMAC?
The power supplies ran on 440 3 phase. That was some POWER!!!
Well sort of good, but I did think that +/-80V would not be readily available but +/-34V ( you need 2 supplies for this) should be more available in a lab. Do not power up +34V without a -34V, as well, that is for sure. So with +/-34V, you should be able to test out most of the pre-amp section
I would suspect that +/-80V supplies might be the ones that are more problematic to locate. So what to do, jump in the deep end? Well at least make sure your PS puts out these correct voltages first before applying it to any 1980 loads.
Last edited:
GrampaDave
Active Member
That board was so beautiful it made me weak in the knees.
yes, except when the lead is shrouded by the body of the component - think radial electrolytic cap.
Time for an update...
I have been off this project for the last few days and finally got a chance to start some testing.
We found a couple problems and luckily were able to correct them quite easily. One problem was in the current source, the schematic was not translated properly so a trace was in the wrong place, we managed to move it and all was good. The other was a silkscreen error on a transistor, flipped the part and it was good.
Once that was all sorted out the voltages all seemed to be within a reasonable tolerance. This is off the top of my head...
pin 17 +80v (+76)
pin 9 +34v (+34)
pin 14 +13.5v (+13.4)
pin 16 +8v (+8)
pin 15 +5v (+5)
pin 18 -80v (-77)
pin 11 -34v (-34)
pin 8 -30v (-30)
I am not sure if the 80V's being a few volts low means anything.
The protection circuit is preventing the speaker relay from pulling in. One of the amp outputs has ~70V on it and the other one has a volt and a half, so it's no surprise that it's not working. What I don't quite understand is when I remove the amp outputs from the board (easy to do now) the protection circuit still does not allow the relay to pull in. Anyway I need to study it a bit more. What I really want is to make sure that the power supply board works completely.
I need to trouble shoot the amp section. Since I am a virgin with the 1980 I planed to study some of the previous threads and see if I can come up to speed a bit. I think the first thing I will look at is separating the pre and power sections. I am not sure if one or both of the channels in the preamp is simply driving the amps to saturation. Before I powered the unit up I checked the outputs and they looked OK but who knows. I am open for any suggestions
I'll try and work on it a bit more this weekend.
I have been off this project for the last few days and finally got a chance to start some testing.
We found a couple problems and luckily were able to correct them quite easily. One problem was in the current source, the schematic was not translated properly so a trace was in the wrong place, we managed to move it and all was good. The other was a silkscreen error on a transistor, flipped the part and it was good.
Once that was all sorted out the voltages all seemed to be within a reasonable tolerance. This is off the top of my head...
pin 17 +80v (+76)
pin 9 +34v (+34)
pin 14 +13.5v (+13.4)
pin 16 +8v (+8)
pin 15 +5v (+5)
pin 18 -80v (-77)
pin 11 -34v (-34)
pin 8 -30v (-30)
I am not sure if the 80V's being a few volts low means anything.
The protection circuit is preventing the speaker relay from pulling in. One of the amp outputs has ~70V on it and the other one has a volt and a half, so it's no surprise that it's not working. What I don't quite understand is when I remove the amp outputs from the board (easy to do now) the protection circuit still does not allow the relay to pull in. Anyway I need to study it a bit more. What I really want is to make sure that the power supply board works completely.
I need to trouble shoot the amp section. Since I am a virgin with the 1980 I planed to study some of the previous threads and see if I can come up to speed a bit. I think the first thing I will look at is separating the pre and power sections. I am not sure if one or both of the channels in the preamp is simply driving the amps to saturation. Before I powered the unit up I checked the outputs and they looked OK but who knows. I am open for any suggestions
I'll try and work on it a bit more this weekend.
Perhaps a new thread for non-power supply related SX-1980 issues?
See if the DC offset adjust can zero either channel.
See if the DC offset adjust can zero either channel.
Well, the +/-80V is used in the input stage and biasing stage. If you're a few volts short of 80V, you may not be able to drive the output of the amp to its full potential.
Odd that I didn't bring this up earlier (like all you want to hear about is another damned mod for the board), but it's so damn common for the output to be short of +/-80V that I've been connecting a resistor sub box to R222 and R223 to find the resistance value that gets as close to 80V as I can get (have to monitor for a while, and I usually settle for about +/-79.5V to make up for the heating that will occur when the unit is all buttoned up and sitting on a shelf). Once I find the proper value (commonly somewhere from 600K to 1M), I solder the part in parallel to R222 or R223 on the foil side of the board.
Ideally, R222 and R223 should be replaced with a 24K resistor, and a 50K trimpot (Bourns 3386F-503LF) put in series with them to make the +/-80V adjustable.
My 2 cents worth.
First off, good job on you PS pcb = cheers
The +/-80V tolerance sounds about right. Not sure what tolerance values that you used for the feedback network resistors? One usually uses 1% or better tolerance for those comps. The zener references also have tolerances (their are two 5% tolerance zeners in series?) that are all part of the overall accuracy.
Are the +/-80V, as they are, sufficient? I say they are fine. Why? well I believe the amp clips based on the supply to the o/p stage, under loads, not the +/-80V to the Input Stage(IPS), voltage amp(VAS).
If you do the math, with the +/-80V supply, theoretically you should be able to get close to (80Vpk * 0.7071) to get the rms V then E squared/8 to get the avg power into 8 ohms which calcs to be 399W. Well we all know it does not put out that much juice!
For the rated 290W/8ohm, you need a sine wave of 48.17Vrms, 68V peak, 136V pk-pk. Yes there are losses to consider in the IPS/VAS and the OPS. So it will never be exactly as I calculated above, but it just shows you what determines the o/p power rating. Under load, it is the o/p V supply sagging from the regulation (losses) in the transformer, rectification losses, the ripple V in the big ecaps, Vce sat losses in the o/p bjts, emitter R losses etc.
I am not saying the EW suggestion of some fine adjustment is not a good one, ( they did do it in the sx-1250 after all) just that I am not certain (well pretty sure) that it will not affect or be the limiting factor in the o/p power, as he suggests.
Sorry for a long 2 cents worth, hope you got your monies worth
and i did not step on any ones toes
Cheers
First off, good job on you PS pcb = cheers
The +/-80V tolerance sounds about right. Not sure what tolerance values that you used for the feedback network resistors? One usually uses 1% or better tolerance for those comps. The zener references also have tolerances (their are two 5% tolerance zeners in series?) that are all part of the overall accuracy.
Are the +/-80V, as they are, sufficient? I say they are fine. Why? well I believe the amp clips based on the supply to the o/p stage, under loads, not the +/-80V to the Input Stage(IPS), voltage amp(VAS).
If you do the math, with the +/-80V supply, theoretically you should be able to get close to (80Vpk * 0.7071) to get the rms V then E squared/8 to get the avg power into 8 ohms which calcs to be 399W. Well we all know it does not put out that much juice!
For the rated 290W/8ohm, you need a sine wave of 48.17Vrms, 68V peak, 136V pk-pk. Yes there are losses to consider in the IPS/VAS and the OPS. So it will never be exactly as I calculated above, but it just shows you what determines the o/p power rating. Under load, it is the o/p V supply sagging from the regulation (losses) in the transformer, rectification losses, the ripple V in the big ecaps, Vce sat losses in the o/p bjts, emitter R losses etc.
I am not saying the EW suggestion of some fine adjustment is not a good one, ( they did do it in the sx-1250 after all) just that I am not certain (well pretty sure) that it will not affect or be the limiting factor in the o/p power, as he suggests.
Sorry for a long 2 cents worth, hope you got your monies worth
and i did not step on any ones toes Cheers
Last edited:
My 2 cents worth.
First off, good job on you PS pcb = cheers
The +/-80V tolerance sounds about right. Not sure what tolerance values that you used for the feedback network resistors? One usually uses 1% or better tolerance for those comps. The zener references also have tolerances (their are two 5% tolerance zeners in series?) that are all part of the overall accuracy.
Are the +/-80V, as they are, sufficient? I say they are fine. Why? well I believe the amp clips based on the supply to the o/p stage, under loads, not the +/-80V to the Input Stage(IPS).
If you do the math, with the +/-80V supply, theoretically you should be able to get close to (80Vpk * 0.7071) to get the rms V then E squared/8 to get the avg power into 8 ohms which calcs to be 399W. Well we all know it does not put out that much juice!
For the rated 290W/8ohm, you need a sine wave of 48V.17Vrms, 68V peak, 136V pk-pk. Yes there are losses to consider in the IPS and the OPS. So it will never be exactly as I calculated above but it just shows you what determines the o/p power rating. Under load, it is the o/p V supply sagging from the regulation( losses) of the transformer, rectification losses, the ripple V in the big ecaps, Vce sat losses in the o/p bjts, emitter R losses etc.
I am not saying the EW suggestion of some fine adjustment is not a good one, ( they did do it in the sx-1250 after all) just that I am not certain (well pretty sure) that it will not affect or be the limiting factor in the o/p power, as he suggests.
Sorry for a long 2 cents worth, hope you got your monies worth
Cheers
Thanks gentlemen for all of the comments and suggestions.
I'll try and make some adjustments to see if I can get the DC offset in check. If not I'll start another thread regarding restoration of the unit. I really want to make sure that all sections of the power supply board work. I think the only thing left to confirm is the protection section.
As for the 80V adjustment, it may or may not be critical to the operation but the bottom line is that it's no problem at all to make the circuit as EW suggests. In the end all of the other voltages are spot on with the manual, the 80V can be as well.
Time to work on the wife's car...
Last edited:
progress but....
So We found a transistor on the driver board that was not really soldered in all that well. Fixed it and then we were able to set the dc and bias. The protection was still not working. Found that R301 was some how purchased as a 1/4 watt when it should have been a 1/2 watt, it was open. Fixed that and then it was working.
I let it run for 30 min, noticed that the controls are all dirty and the tuner acts weird at the bottom of the dial.
I turned the unit off and back on a min later and I had 70+ volts on the outputs again. I suspect that it's in the driver boards.
But the good news is that the board seems to work!
So We found a transistor on the driver board that was not really soldered in all that well. Fixed it and then we were able to set the dc and bias. The protection was still not working. Found that R301 was some how purchased as a 1/4 watt when it should have been a 1/2 watt, it was open. Fixed that and then it was working.
I let it run for 30 min, noticed that the controls are all dirty and the tuner acts weird at the bottom of the dial.
I turned the unit off and back on a min later and I had 70+ volts on the outputs again. I suspect that it's in the driver boards.
But the good news is that the board seems to work!
Last edited:
Have you polished all the connector pins in the unit especially the driver to output transistor connectors?
The pencil eraser trick?
The pencil eraser trick?