Antec SL-400 5vsb Mod with NCP1200

[LDSisHere]

Re: Antec SL-400 5vsb Mod with NCP1200

I do not think the soldering was the problem but at this point who knows. That supply has not been in a case since I got it working many months ago and it was upside down when I powered it on so that I could probe around to see if I could find any problem areas. I think the FET may have been damaged and I did not catch it or the gate was "floating" on, but I have no evidence one way or the other at this point. All I can do now, is put Humpty back together and try not to push him back off the wall again as I did in this situation. I want to save that for the load tests.

[ben7]

Re: Antec SL-400 5vsb Mod with NCP1200

I don't know how else it could have exploded besides bad soldering, bottom of the board touching the psu case, or, you had floating wires inside the thing since you removed the control board.

[LDSisHere]

Re: Antec SL-400 5vsb Mod with NCP1200

The control board was laying off to the side on the table when I applied power. Hopefully I will not make that mistake again.

[ben7]

Re: Antec SL-400 5vsb Mod with NCP1200

^Oooo ahhh!

Did you have the whole control board removed, or did you forget to attach the feedback to the opto-isolator? Or, did you make a silly boo-boo, er, shorted pcb trace?

[LDSisHere]

Re: Antec SL-400 5vsb Mod with NCP1200

Carnage pics?

I was finally able to take this supply apart to see what had been damaged. There were more components damaged than what I have posted pictures of but the rest did not have any visible damage so I am not bothering to post them. I am now in the process of scrounging up replacement parts, so I am not sure how long it is going to take for me to get this supply back together so I can continue testing.

[ben7]

Re: Antec SL-400 5vsb Mod with NCP1200

Carnage pics?

[budm]

Re: Antec SL-400 5vsb Mod with NCP1200

Darn, that is too bad, I was waiting to find out about the snubber to see what value they turn out to be, will wait for you next report when you go back to it.

[LDSisHere]

Re: Antec SL-400 5vsb Mod with NCP1200

Thursday night I decided to take a little time to measure the wave form without the snubber installed so I could then do some testing with various values of capacitors to determine what value would double the waveform. Unfortunately, things did not go as planned.

I connected everything up with no load on the supply. When I applied power I heard a loud pop and I had an interesting show. As is the case most of the time when I make a stupid mistake, I was tired and probably should have called it night before working on this supply. It turns out that I forgot to solder the winding and diode lead back after removing the snubber. What I found was a shorted 100V 1A diode on the primary side and a popped Nichicon PW 47uF 50V.

I replaced the 100V 1A diode with a 200V 1A diode and the capacitor with a Nichicon PW 47uF 63V. I apply power and get no output. I cannot easily determine the problem at this point but I think the output diode on the secondary may have been damaged so I replace it with a 60V 10A diode I received earlier in the week.

This is the point where I should have really quit while I was ahead but nooooooo, I had to keep at it. This time when I apply power I get a real fireworks show.

As I sit there in disbelief looking at this catastrophe, I immediately see where I really screwed up this time. -> In my haste to test the supply I had for gotten to reattach the NCP1200 board. I think the 5vsb ran away without the board in place to throttle it back. Needless to say, I have a lot of work to do before I will be able to do anymore testing with this supply. Since I have a lot of other projects that are more pressing, it will probably be a while before I am able to get back to this one. Ugh!

[ben7]

Re: Antec SL-400 5vsb Mod with NCP1200

If you look in the bottom left corner of the screen you will see a "1" in a red circle, next to that is the V/Div (10V AC in this case). I do not have any issue with the scopes interface yet, since this is the first scope I have touched almost 20 years I do not have anything to compare it with and that may be to my benefit.

Saying it is well away from from 40V may have been a stretch but it at least looks like it should not be overvolting the diode. I have not had a chance to do any farther experiments with this supply.

I think it might be fine like that (at least at that load), but you should try to get the reverse voltage a bit lower, just for a safety margin.

[LDSisHere]

Re: Antec SL-400 5vsb Mod with NCP1200

If you look in the bottom left corner of the screen you will see a "1" in a red circle, next to that is the V/Div (10V AC in this case). I do not have any issue with the scopes interface yet, since this is the first scope I have touched almost 20 years I do not have anything to compare it with and that may be to my benefit.

Saying it is well away from from 40V may have been a stretch but it at least looks like it should not be overvolting the diode. I have not had a chance to do any farther experiments with this supply.

[ben7]

Re: Antec SL-400 5vsb Mod with NCP1200

I discovered yesterday that it does indicate when the bandwidth is limited, it is just not that noticeable. If you look at the bottom left of the screen (post 43, picture 1) where it has the channel 1 information, just to the right of 10V it shows BW. Not exactly obvious but at least it is there. I learn more about this scope and just using one, each time I use it. Trying to get it to show you what you want to see is sometimes tricky when you do not have a clue about what you are doing.



BudM thanks for this information, it is the first snubber document I have read that I can actually understand and it applies directly to my problem.



I did get around to trying the .01uF 100ohm combo yesterday and you should be able to see the results in the images. It is far from perfect but it does seem to keep the voltage well away from the 40V range. When I get a chance I will do as the document BudM provided suggests and see if I can find a value that doubles the time period and calculate the resistor from that information. Now I just need to figure out if and/or how to get my scope to display that information.

How many V/div? I don't like the layout of the interface there xD

[budm]

Re: Antec SL-400 5vsb Mod with NCP1200

Page 11 has good info.

[LDSisHere]

Re: Antec SL-400 5vsb Mod with NCP1200

Your scope display doesn't indicate, in the first image, that the BW is limited to 20MHz. I'm not quibbling, I'm just used to Tek scopes, which put "BWL" next to any channel which is set to 20MHz BW. But I certainly understand that Tek scopes are not inexpensive ($5K-$6K for a 100MHz, 4-channel TDS-3000C, IIRC).

I discovered yesterday that it does indicate when the bandwidth is limited, it is just not that noticeable. If you look at the bottom left of the screen (post 43, picture 1) where it has the channel 1 information, just to the right of 10V it shows BW. Not exactly obvious but at least it is there. I learn more about this scope and just using one, each time I use it. Trying to get it to show you what you want to see is sometimes tricky when you do not have a clue about what you are doing.

Snubber design.

BudM thanks for this information, it is the first snubber document I have read that I can actually understand and it applies directly to my problem.



I did get around to trying the .01uF 100ohm combo yesterday and you should be able to see the results in the images. It is far from perfect but it does seem to keep the voltage well away from the 40V range. When I get a chance I will do as the document BudM provided suggests and see if I can find a value that doubles the time period and calculate the resistor from that information. Now I just need to figure out if and/or how to get my scope to display that information.

[budm]

Re: Antec SL-400 5vsb Mod with NCP1200

Snubber design.

[ben7]

Re: Antec SL-400 5vsb Mod with NCP1200

That's it. Just be sure your resistor isn't installed backwards.

LOL???!!!

Yes indeed the snubber circuit you drew is good. The resistor might need to be rated at a watt or two, so be careful not to burn yourself on an underrated resistor, or start a fire!

[PeteS in CA]

Re: Antec SL-400 5vsb Mod with NCP1200

That's it. Just be sure your resistor isn't installed backwards.

[LDSisHere]

Re: Antec SL-400 5vsb Mod with NCP1200

Thanks again for all this great information. I will probably start with the .01uF capacitor (50V) with the 100ohm resistor (1/2W) since I have those on hand. I am just not sure if I understand how to connect them correctly for the desired result. I am attaching a schematic of how I think it is supposed to be done, but I want to make sure before proceeding.

[ben7]

Re: Antec SL-400 5vsb Mod with NCP1200

The norm for mil-spec electronics used to be 100% or greater margin. E.G. a 1/4W resistor would be required if the part dissipated 1/10W. That's not practical in this case, of course. Looks like a 60V Schottky would suffice: using a 100V ultrafast would be double the power dissipation; even a 100V Schottky would dissipate more power than a 60V part.

Since LDSiH is experimenting, trying several permutations of an RC snubber might be interesting. I don't know the math, but trying several caps between .01uF and .1uF in series with several resistors between 10 ohms and 100 ohms, across each Schottky might do interesting things to that ringing when the rectifier turns off. The resistors would have to be .5W or 1W, the .1uF--10 ohm RC would dissipate the most power. The idea is to limit the magnitude of the spike and dissipate that energy in that resistor without slowing down the rectifier turn off or dissipating too much power in the resistor. So start with the lowest value C and the highest value R and gradually optimize the snubber.

And make sure it doesn't cook the capacitors! xD

[PeteS in CA]

Re: Antec SL-400 5vsb Mod with NCP1200

The norm for mil-spec electronics used to be 100% or greater margin. E.G. a 1/4W resistor would be required if the part dissipated 1/10W. That's not practical in this case, of course. Looks like a 60V Schottky would suffice: using a 100V ultrafast would be double the power dissipation; even a 100V Schottky would dissipate more power than a 60V part.

Since LDSiH is experimenting, trying several permutations of an RC snubber might be interesting. I don't know the math, but trying several caps between .01uF and .1uF in series with several resistors between 10 ohms and 100 ohms, across each Schottky might do interesting things to that ringing when the rectifier turns off. The resistors would have to be .5W or 1W, the .1uF--10 ohm RC would dissipate the most power. The idea is to limit the magnitude of the spike and dissipate that energy in that resistor without slowing down the rectifier turn off or dissipating too much power in the resistor. So start with the lowest value C and the highest value R and gradually optimize the snubber.

[ben7]

Re: Antec SL-400 5vsb Mod with NCP1200

The real world of electronic circuits has all kinds of interesting limitations and parasitic effects. More basic electronics classes tend to ignore them, for simplicity. But in the real world they can bite you. But just think of it as increasing your understanding of how things work. And as opportunities in the future to help others.

Couldn't agree more!

Hence why I put in parts that are better than "needed". 16v capacitor on a 15v rail?- a bit too close for comfort, etc...