Truepower 2.0 recap and repair

[PCBONEZ]

Re: Truepower 2.0 recap and repair

That sucks. - I just bought a Truepower 2.0.

Interesting that there's an Antec ad up top as I read this... LOL

[999999999]

Re: Truepower 2.0 recap and repair

...Now i just need to find a replacement fuse

far as i can tell it's an M205 12A 250V ceramic (i broke it getting it out, so no pics)

Can anyone confirm for me if it would be a slow blow/fast blow, if i will have any problems with a glass replacement and whether i could get away with 10A in a 550W supply? 12A fuses are like hens teeth here, particularly in the ceramic cartrige variety. The only place i've been able to find an exact replacement is at a specialist fuse dealer but minimum buy is 30 at a cost of about $35 with tails

A 10A fuse is ok. Ceramic are a little safer but glass are ok, it should be slow-blow as spec'd in ATX12V PS Design Guide, use of fast blow would require significantly higher current rating making it less effective protection in this use.

[Brian C]

Re: Truepower 2.0 recap and repair

10x110v = 1100w

Need more clarification? :P

It is not that way how we choose fuse rating. You need to consider the efficiency. Whether slow blow or fast blow, you may need to check the in-rush current.
e.g. for a 110V, 550W SMPS. with 80% efficiency.
---------------------------------------------------------------------------------
(1) input power = 550/0.8 = 687.5W
(2) effective input current = 687.5/110 = 6.25A
(3) assuming an effective input resistance of 0.4Ohm, resistance factor = 0.4 x 687.5 = 275 Ohm.W
(4) for resistance factor of 275 Ohm.W. The RMS input current ~ 6.25 x 1.45 = 9.06A.
---------------------------------------------------------------------------------
The continuous current rating of fuse is 9.06A. RMS current rating varies with efficiency and minimum operating voltage. PSU with min operating voltage of 90Vac requires higher rms input current...

[Kaine]

Re: Truepower 2.0 recap and repair

OK keep in mind i'm on ~240v down here (depending on who you ask it's between 230v and 300v, i honestly can't remember what it actually is right now) I'll go with 230v because it's the "worst case scenario" as far as current draw goes.

---------------------------------------------------------------------------------
(1) input power = 550/0.8 = 687.5W
(2) effective input current = 687.5/230 = ~2.99A
(3) assuming an effective input resistance of 0.4Ohm, resistance factor = 0.4 x 687.5 = 275 Ohm.W
(4) for resistance factor of 275 Ohm.W. The RMS input current ~ 2.99 x 1.45 = 4.33A.
---------------------------------------------------------------------------------

So a 10A fuse should be more than enough considering we run the higher voltage in household supply.

Thanks a lot guys you've really cleared it up for me

[Sparky]

Re: Truepower 2.0 recap and repair

That sucks. - I just bought a Truepower 2.0.

Interesting that there's an Antec ad up top as I read this... LOL

You can always return it. Or when it fails (which could be in as little as 6 months or as long as several years) if it is still under warranty RMA it and they give you a newer PSU which doesn't have any crap caps.

Hence another reason for their ad - they finally learned and don't have crap caps anymore (they don't use CWT for their OEM anymore either).

[PCBONEZ]

Re: Truepower 2.0 recap and repair

They don't make the TP2's anymore.
The problem with a replacement is that the newer models don't have the power I need on the rails where I need it. They would have to upgrade the total watts to get me what I need and I don't think they'd go that far. - I have to pay particular attention to both the 5v and 12v rails on this one because I'll be pushing a lot of drives in this system.

.

[Kaine]

Re: Truepower 2.0 recap and repair

Would this fuse be appropriate? From what i can gather from the datasheet it's slow blow, even if they don't advertise it there.

[999999999]

Re: Truepower 2.0 recap and repair

Would this fuse be appropriate? From what i can gather from the datasheet it's slow blow, even if they don't advertise it there.

RS includes session ID info in the browser URL which broke your link. The easiest way to link a RS product is probably to use the "Email to colleague" link at the bottom.

> New link to that fuse <

Assuming your present fuse holder is for 5 x 20mm size (which is the most common), yes that would work. You might have some old PSU lying around that you could pull the fuse out of instead, if it's rated high enough, or if there's a friendly computer shop nearby you might get them to hold a dead PSU for you to take the fuse out of (if it isn't blown). I just suggest these alternatives because it seems expensive to pay shipping and the 10 unit minimum for a 48 cent part unless you needed other parts from RS too.

[Brian C]

Re: Truepower 2.0 recap and repair

The I^2t rating of the 10A fuse is 490.000. In case of over-current for 0.1 second, you need 70 Ampere of current to melt the fuse element...

[Kaine]

Re: Truepower 2.0 recap and repair

You might have some old PSU lying around that you could pull the fuse out of instead, if it's rated high enough, or if there's a friendly computer shop nearby you might get them to hold a dead PSU for you to take the fuse out of (if it isn't blown). I just suggest these alternatives because it seems expensive to pay shipping and the 10 unit minimum for a 48 cent part unless you needed other parts from RS too.

Good call, I've contacted a mate at a local computer shop so i'm gonna go scab some of their dead/dying psus. Unfortunately all the ones i have around here only have 6A fuses in them

[Brian C]

Re: Truepower 2.0 recap and repair

Good call, I've contacted a mate at a local computer shop so i'm gonna go scab some of their dead/dying psus. Unfortunately all the ones i have around here only have 6A fuses in them

Try to get fuses with safety mark eg. UL, CSA, BS. They are more reliable.

[Brian C]

Re: Truepower 2.0 recap and repair

The I^2t rating of the 10A fuse is 490.000. In case of over-current for 0.1 second, you need 70 Ampere of current to melt the fuse element...

Slow blow fuses need huge current to melt the element in short time (from the I^t rating). It doesn't mean the fuse is not appropriate.

[Kaine]

Re: Truepower 2.0 recap and repair

ok first test power up was a no go

While i try and work out how to troubleshoot a circuit i know very little about(help!), here are some After pics. Warning: big.

[999999999]

Re: Truepower 2.0 recap and repair

I was waiting for someone else who had one to chime in, since further replies by me would only be speculation... so I only give a few random thoughts.

Check the Topswitch application notes, they give an example circuit which is probably not so much different than your PSU.

Check the function of the two diodes you wrote were in parallel to the one you replaced.

Check that baked solder joint with the crusty blackness in the picture.

Check the (3A?) diode adjacent to the smaller transformer as the high current had to be passing through it.

Check the topswitch too if you feel comfortable testing while the PSU is live, the output from the bridge rectifier +DC should be continuous to one leg of the topswitch. If the Topswitch is SOIC on the bottom which it looked like in the picture, it looked a little crusty/thermally-stressed so I'd also wonder about it.

It all starts with tracing what you can while there is no power applied, but ultimately if that doesn't find a fault you are left applying power and then checking functionality (reasonably speaking, it is not reasonable IMO to replicate the same environment off-board to test each IC).

[Brian C]

Re: Truepower 2.0 recap and repair

ok first test power up was a no go

While i try and work out how to troubleshoot a circuit i know very little about(help!), here are some After pics. Warning: big.

From ur picture (PCB layout), it should have no TOPSWITCH. The design uses MOSFET transistor as switcher. Please refer to the attached picture.... (click on the attached thumnails for full size)



First of all, is standby 5V (purple color wire) OK ? If standby 5V is normal, u need to check the main supply section. Well, for the time being I assume the standby 5V is not working.

Here are some suggestions from me.
----------------------------------
(1) Check the replacement diode connection (polarity). It should be as shown in the picture (F).
(2) (A) could be a jumper or resistor. Check the resistance if it is a resistor. Usually small resistance value (4.7 Ohm).
(3) Check the power resistor (D) for open circuit. The power resistor is connected to the Gate of MOSFET transistor (C). It is 2M Ohms from ur picture.
(4) Check the MOSFET transistor (C). It should be 4 to 6 Ampere, 600V or higher MOSFET.
(5) (B) should be 78xx series regulator, probably 18V or 10V. It is the supply to the UC3844B (startup Vth=17.5Vmax) or UC3845B (startup Vth=9.0Vmax) controller IC. Check this IC if ur standby 5V is OK but cannot power ON.

[Kaine]

Re: Truepower 2.0 recap and repair

Wow brian, that is precisely what i was hoping for. I'll hopefully get the motivation up to go through that later today. It's competing for time with the cannibalization of a gigabyte motherboard full of rubycon MBZz, nichicons, chemi-con KZGs and... teapos? I'm debating whether to do any live testing... i'll have to run some long wires from the power input to the board itself so i can get to the underside of it while it's on... good times. Any particular areas/points on the board i should check for certain voltages?

[Kaine]

Re: Truepower 2.0 recap and repair

well i'm just about ready to give up. I couldn't find any malfunctioning components (haven't checked ICs, don't have the resources) so i tried to test it live on my bench today and couldn't even get a voltage reading on the board where the leads come from the AC input circuitry. Then i went to try to get a reading on the circuit that sits on the back of the 3-pin plug and shorted it, blowing a house fuse and making a nice big spark.

can't tell if i cooked any components, but i did blow a nice chunk out of the solder and gave myself a fright. My multi seems to have survived the ordeal too. should i just pull out the caps i put in it and move on? Maybe i should try recapping a working supply 8-)

[Kaine]

Re: Truepower 2.0 recap and repair

Forgot my testing before i shorted it :P

Resistor D checks out at 2Mohms
A appears to be a jumper/link
Diodes are in correct polarity and seem to be functioning (closed circuit in one direction only, little to no resistance)
MOSFET C i can only get momentary continuity between the base and outside pins, but none across the pins. is this correct right or should i pull it out and test it again? (major pain in the ass, it's screwed to the HS)

stupid question:
where do i check for standby 5v? i can't find any voltage anywhere during operation.

[999999999]

Re: Truepower 2.0 recap and repair

I find it handy to keep a second set of probes nearby, upon which I have put heat-shink tubing so they're insulated right up to the tip. I'm sure you can buy probes already insulated further up, but I have a few sets of probes from various meters so it was quicker and cheaper to use what I had.

So far as further diagnosis, trace the circuit after each key component. Take a voltage reading at a point, then after a connector or component there's the next point, and so on.

When you wrote that you couldn't get a voltage reading on the board where the leads come from the AC input, that seems unusual, were you measuring AC or DC or (maybe meter was set wrong?) ?

[Kaine]

Re: Truepower 2.0 recap and repair

thanks for the hasty reply 9s.

I'll probably take your advice on insulating my probes that little bit more. They've taken only superficial damage this time but i think i'll grab another set all the same.

I tried taking readings on both AC and DC settings and got nothing. I was fully expecting to blow the fuse in my meter at the very least. I had the power-on pin shorted and all. Beginning to think the input circuit may need a look at. It's pretty simple but i have to de-solder it from the power socket to get a good look at it. The short seemed to only have blown a bit of solder off, it wasn't in immediate vicinity to any sensitive components anyway.