Cooler Master 1200W Silent Pro Gold - Bad power switch

[tony359]

Hi all,

The PSU in question had a bad power switch. There was not physical feedback when switching on and - if memory serves - it was always ON even when in OFF position.

I researched a bit and got some replacements.

The replacement "got stuck" at first power up - as in "it's now always on".

It looks like this PSU has a pretty large inrush current as my lights flicker when I turn it on with no load.

Probably the switches I got are bad quality but I was wondering if someone could recommend a good replacement?

I THINK the original switch says on the body. I am not sure this is the actual original one. I remember I spent quite some time looking for a suitable replacement when I looked for one.
12A 125/250VAC
8A/100A250V T85
12(4)A 250V-T85

The replacement is
6(4)/250VAC
10(4)/250VAC
5/100A 250VAC
12A125-250VAC
1/3 HP 125VAC
1/2 HP 250VAC

https://www.tweaktown.com/reviews/47...iew/index.html

I'm also wondering whether there is something wrong with the PSU - soft start malfunctioning?

Any help is appreciated!
Thanks

[stj]

Re: Cooler Master 1200W Silent Pro Gold - Bad power switch

they switched 1200w with that tiny switch? i'm not surprised it welded the contacts!
i dont know what to say, if it was mine i'd dremel out the hole and fit a bigger switch with 20A contacts.

[tony359]

Re: Cooler Master 1200W Silent Pro Gold - Bad power switch

Well, I guess the idea is that it will never switch 1200W as the PSU will have some sort of inrush protection and soft-start?

At 240V though that's "only" 5A.

I looked into the circuit a bit. There is a Varistor, some safety capacitors, inductors then it goes into two bridges in parallel. From there to the main capacitors and there is an NTC thermistor there.

It seems to be working (even though the no-load resistance is only 8 ohm - as per specification though).

I found a review claiming that this PSU has "only" 56A of inrush current "thanks to the large thermistor". I don't see a large thermistor to be honest

[stj]

Re: Cooler Master 1200W Silent Pro Gold - Bad power switch

does it have a capacitor across the switch to reduce arcing?

[tony359]

Re: Cooler Master 1200W Silent Pro Gold - Bad power switch

There are two but across L and N, not across the switch.
There also are teo small ones across L and E and N and E.

I thought those were filter capacitors?

I’ll post the actual values when I’m back in front of it.

There is a Varistor on mains input. The thermistor is on D.C., close to the main capacitors?

[tony359]

Re: Cooler Master 1200W Silent Pro Gold - Bad power switch

Adding some pictures with some description to help.
On those pictures I've removed a small thermistor which would normally sit in between the capacitors.

I'm trying to find out what soft-start topology this PSU has. I was looking for a thyristor but I couldn't find one. That thermistor could be the one but it's only 8 ohm when cold and max 4A. This is a 1200W PSU so there must be something else bypassing it when the caps are charged.

Any pointers, please shout! And thanks!

[Per Hansson]

Re: Cooler Master 1200W Silent Pro Gold - Bad power switch

I'm trying to find out what soft-start topology this PSU has. I was looking for a thyristor but I couldn't find one. That thermistor could be the one but it's only 8 ohm when cold and max 4A. This is a 1200W PSU so there must be something else bypassing it when the caps are charged.

You have drawn the rough path up, to confirm you did not miss anything just do a resistance check from the mains to the bridge rectifiers input.
And then from the bridge rectifiers output to the capacitors.
If there is nothing there then the answer is that there is nothing there.

[stj]

Re: Cooler Master 1200W Silent Pro Gold - Bad power switch

when you turn on the switch, even with the pc off the problem will be the inrush charging the 2 390uf caps!

[tony359]

Re: Cooler Master 1200W Silent Pro Gold - Bad power switch

@per - Shall do!
However: what if I'm reading zero resistance BECAUSE the soft start component is shorted - hence the massive inrush?

@stj
That's why soft start exists! I'm ok with inrush. I am not ok with the main switch getting stuck because of that! Clearly there is something wrong. Could also be that the switch I used was not good but the fact that the existing switch was also stuck, raise a red flag

Also, there is a thermistor in between the capacitors, I suspect it's got something to do with that. But the thermistor needs some device which jumps the rails to the capacitors once they're charged.

[Per Hansson]

Re: Cooler Master 1200W Silent Pro Gold - Bad power switch

I was kind of implying the thing has no soft-start, but of course just follow the traces to confirm that.
Most can be seen directly from the underside, just follow the traces and see what components are on the other side.
To give this thing a proper soft start you could replace the thermistor with say something 20 to 100ohm.
And then in parallel with that put a 5VDC relay that shorts it out.
Connect the relays coil to the 5VSB and you will have a pretty elegant soft-start.
But there is very little room so seems hard, btw are you sure the diode you drawn before the thermistor is really a diode? I don't understand what purpose it would have...

[tony359]

Re: Cooler Master 1200W Silent Pro Gold - Bad power switch

Thank you.

The main issue is the switch sticking to the ON position. I don't have issues with no soft-start but I do with the switch getting stuck every time

Yes, as you say the space is limited but thanks for the input. I might consider that.

I'll see if I can trace things better but the PCB is dual-layer and really there is no way I can follow traces on top because of the crammed components.

The diode looks like a diode to me (attached) but I'd be ready to be proven wrong!


I've measured continuity between the positive of the bridge and the positive of the capacitor and no, there is no continuity. I can measure the junction of the diode of course or megaohms when in resistance mode.

So it's not directly connected as suspected. It's also very difficult to work on that PCB as the output PCB cannot be unscrewed as they soldered the bolts in place...

[Per Hansson]

Re: Cooler Master 1200W Silent Pro Gold - Bad power switch

Ok, I was guessing it was just a wire with an insulator around it from the previous pictures but it is obvious from this picture that it actually is a diode.
That diode is for the Active PFC (APFC) large inductor on the edge of the PCB, it prevents the boosted voltage (390VDC or so) to backfeed into the lower voltage at the bridge rectifiers when the main power supply starts.
Anyway that does not matter, to fix the switch you could just bridge it or leave it on and just unplug the power cord instead.
Or build a proper soft start

[tony359]

Re: Cooler Master 1200W Silent Pro Gold - Bad power switch



I feel the time for tweaking the PSU has yet to come. At the moment I'd like to find out what's wrong with it!

That thermistor is connecting to the capacitors so it seems to be the "soft start" - but there must be something else bridging the rail to the capacitor when they are charged.

Is there a way I can measure the inrush current of that PSU - without a current probe for my scope? I don't have one

[Per Hansson]

Re: Cooler Master 1200W Silent Pro Gold - Bad power switch

I feel the time for tweaking the PSU has yet to come. At the moment I'd like to find out what's wrong with it!

There is nothing wrong with it.

That thermistor is connecting to the capacitors so it seems to be the "soft start" - but there must be something else bridging the rail to the capacitor when they are charged.

Only more expensive power supplies use a relay to switch out the thermistor (to increase efficiency).
Without that when the thermistor heats up to 60°C or so it will have a quite low resistance, probably below 1 ohm.

Is there a way I can measure the inrush current of that PSU - without a current probe for my scope? I don't have one

No but you can calculate it if you measure the voltage rise time of the capacitors.
VERY BIG DISCLAIMER: DO NOT CONNECT THE NEGATIVE GROUND WIRE OF YOUR SCOPE TO THE NEGATIVE OF THE BULK FILTERING CAPACITOR!
READ THIS THREAD: https://www.badcaps.net/forum/showthread.php?t=96508

[tony359]

Re: Cooler Master 1200W Silent Pro Gold - Bad power switch

@Per

Let me disagree with you on that. How can the PSU be ok if its main switch gets stuck every time I use it? It cannot be right.

Yes, the thermistor will be very low resistance but this is a 1200W power supply. The thermistor is rated 4 Amps. Indeed, the thermistor is on rectified AC which at 240V is 338V. 338V x 4A is 1353W which is what the PSU is rated for - even though it feels a bit tight!

So are you suggesting that the whole current drawn by the PSU flows through that little component? And the thermistor is constantly at 60+ degrees whenever the PSU is being used?

Thanks about measuring the voltage rise time and no worries, I have a differential probe for power supplies.

[Per Hansson]

Re: Cooler Master 1200W Silent Pro Gold - Bad power switch

@Per

Let me disagree with you on that. How can the PSU be ok if its main switch gets stuck every time I use it? It cannot be right.

The power supply is 13 years old according to text on the PCB, the original switch failed once, and then your replacement failed, that hardly counts as "every time" unless you did not use the power supply for 13 years.

Yes, the thermistor will be very low resistance but this is a 1200W power supply. The thermistor is rated 4 Amps. Indeed, the thermistor is on rectified AC which at 240V is 338V. 338V x 4A is 1353W which is what the PSU is rated for - even though it feels a bit tight!

The main PSU is not on when you switch on the mains switch, only the standby 5VSB is on.

So are you suggesting that the whole current drawn by the PSU flows through that little component? And the thermistor is constantly at 60+ degrees whenever the PSU is being used?

I think it is better if we take a step back and look at the whole picture:
240VAC (or 115VAC in the US) comes in through the black and white cables.
They go through a 16A fuse F1 and through two common mode rejection chokes into two bridge rectifiers.
The bridge rectifiers change the voltage to DC at 240VAC x √2 = 340VDC (Or 115VAC x √2 = 163VDC in the US)

The positive goes up to the diode we discussed before and through the thermistor that you wrote on Reddit is a SCK-084 and then directly into the bulk filtering capacitors.
The negative presumably goes directly to the negative of the bulk filtering capacitors, you can check this with a multimeter please.

This completes the standby portion of the power supply, once the voltage reaches something like 150VDC the 5VSB control chip (that I can't see) will start working and after startup it will produce 5VSB via the small transformer.
When the main power supply is started by taking the green ATX cable PS_ON low the APFC stage in the power supply is activated.
This boosts the voltage from 163VDC (in the US) or 340VDC (Europe) up to around 390VDC via the large APFC inductor.
This is done with the two MOSFETS closest to that inductor (I was unable to see their part number) and the large diode on the same heatsink.
This is where the previously mentioned diode becomes useful: it now prevents the higher DC voltage from feeding back towards the lower DC voltage at the bridge rectifiers.
At this stage or rather the same time the other two mosfets 35N60C3 will start pulsing the high voltage DC through the large transformer on the left (check it's soldering in attached photo!)
This is what produces the 12VDC in the power supply, finally on the secondary there are two DC-DC boards that step the 12v down to 5v and 3.3v respectively.


Now since your problem is with the switch you can ignore the third paragraph: the inrush only happens when the standby stage starts.
As you can see the voltage on the positive side can only go through the thermistor, so there is nothing "wrong" with the soft start. (Unless there is something on the negative side I'm not seeing).
The boosted voltage does not go through the thermistor, but instead via the APFC inductor, the two large MOSFETS and then I guess via the heatsink to the big diode on the heatsink.
But in any case the DC coming in from mains is still enough to heat the thermistor to somewhere around 60°C to 100°C lowering its resistance to 1 ohm or so. (depending on load on the power supply of course).

[tony359]

Re: Cooler Master 1200W Silent Pro Gold - Bad power switch

Thank you for taking the time to guide me through the process, and for marking up the PCB picture, I appreciate it.

I confirm the negative of the bridges is going straight to the negative of the capacitors. I will check the soldering you indicated.

Indeed the inrush current happens when I flip the mains switch at the back (PSU not connected to anything): this is where the caps get charged with 340V.

I didn't buy this PSU, I acquired some time ago second hand. When I got it, I found the original switch had got stuck. So I replaced it with another one. It **immediately** got stuck when I first powered up the PSU.

Powered up the PSU = Applied 240VAC then toggled the mains switch on. Nothing connected to the PSU.

I replaced it a second time and it **immediately** got stuck again. I never used this PSU besides those two tests.

I have a feeling we are disagreeing on some wording here
What I am trying to say is that it seems the soft-start circuitry might be malfunctioning as the main switch fails every single time.

I do not know what happened during the past 13 years. I got the PSU a couple of years ago and only tested those few times to ascertain the switch was getting stuck. So yes, it fails "every time" from my point of view!

I am still not understanding how PFC works but are you saying that once the PFC starts, there is no current flowing into the thermistor as the PFC is doing all the work? So the thermistor should only get to 60-100C when the mains switch is toggled and capacitors are charged?

Thanks again!

[Per Hansson]

Re: Cooler Master 1200W Silent Pro Gold - Bad power switch

I am still not understanding how PFC works but are you saying that once the PFC starts, there is no current flowing into the thermistor as the PFC is doing all the work? So the thermistor should only get to 60-100C when the mains switch is toggled and capacitors are charged?

I have not seen the thermistor put in this position before, it is pretty clever:
Normally it sits directly at the AC input side, and thus has to be rated for the full load of the PSU (i.e. be big and expensive and dissipate allot of heat)
But where they put it that is after the APFC inductor.
So the energy that is in the APFC inductor never has to pass through the thermistor, since it is located after the inductor.
There is still energy passing through it, but just allot less than would be if it was positioned in a more traditional place like directly at the input.

I confirm the negative of the bridges is going straight to the negative of the capacitors.

I have a feeling we are disagreeing on some wording here
What I am trying to say is that it seems the soft-start circuitry might be malfunctioning as the main switch fails every single time.

No we are not disagreeing on wording: I'm saying the only soft start that exists is the thermistor.
In post #11 you checked the positive side and this goes directly to the big capacitors through a diode and the thermistor.
And then in the quote above you confirm that the negative goes directly to the big capacitors.
Inrush protection can only be in series with the capacitors: so you can ignore all other circuitry and thus reach the conclusion that the thermistor is the soft start
This was what I was alluding to in post #7, maybe I should have been more clear.

Thank you for taking the time to guide me through the process, and for marking up the PCB picture, I appreciate it.

You're welcome, can you also please check resistance at the point I marked with red text as "Heatsink Pin???" in the same picture, if that leads to the heatsink or if you can see where it leads in the top of the PCB?

[tony359]

Re: Cooler Master 1200W Silent Pro Gold - Bad power switch

No we are not disagreeing on wording: I'm saying the only soft start that exists is the thermistor.

Then I'm not sure I understand what you mean with

There is nothing wrong with it.

I'm not trying to argue with you of course!

Inrush protection can only be in series with the capacitors: so you can ignore all other circuitry and thus reach the conclusion that the thermistor is the soft start

Great. So, since every time I turn the switch on it gets stuck, would you agree that maybe the thermistor has failed - or partially failed?

You're welcome, can you also please check resistance at the point I marked with red text as "Heatsink Pin???" in the same picture, if that leads to the heatsink or if you can see where it leads in the top of the PCB?

I confirm those are the heatsink pins.

And thanks for explaining how the soft start works on this unit, pretty clever indeed.

[Per Hansson]

Re: Cooler Master 1200W Silent Pro Gold - Bad power switch

What I meant when I said "There is nothing wrong with it" is that there is no fault per-se, it works as intended.
You should instead ask yourself why did the original switch fail?
Most people never use that switch, so how do you know the original did not fail very quickly also?
And that your replacement switches are much too weak to handle the inrush just like the original was?

I would either just bypass the switch or go all in and attach two wires where the thermistor used to be and connect them up to a relay like this one.
Then I'd crimp a thermistor with a much larger ohm value together with those two wires on to the relay and power the relay from the +5VSB.