Faulty Enermax Liberty ELT620AWT

[Latoof]

My Enermax Liberty ELT620AWT has died...

It will not turn on, when power is applied the fan spins for about 0.5sec and then nothin! There are no apparent crispy bits nor did any magic smoke escape.

When it first occured I noted that the PS_ON line (Pin 14, green wire) was at 1.29VDC and did not change when the power ON sw was depressed.

I tried manually pulling pin 14 to GND and this allowed the Enermax to fire up but this no longer works...??

Installing my spare CoolerMaster PSU, pin 14 starts at 5.19VDC and goes to less than 0.1VDC and stays there and PSU fires up.

I assume my MoBo has some sort of latch that pulls pin 14 low and holds it there but with no cct diagram I have not yet confirmed this.

Given the spare PSU works I also assume that my MoBo is not the cause of the problem....

If my assumptions are correct, it would appear that the PS223 monitoring IC is fauty.

I.C. specs are HERE

I have tried to locate a new PS223 chip withouit sucess.

Any advice, ideas etc appreciated.

[Brian C]

Re: Faulty Enermax Liberty ELT620AWT

Pls check the STANDBY 5V line (purple color wire). It should read 5V.

[Latoof]

Re: Faulty Enermax Liberty ELT620AWT

Pls check the STANDBY 5V line (purple color wire). It should read 5V.

Hi Brian and Tkx for the reply. With the PSU on the bench not connected to anything there is 5.19VDC on the purple wire (pin 9 +5VSB)

[Latoof]

Re: Faulty Enermax Liberty ELT620AWT

A simple ohm measurement across the various o/p donot reveal any hard shorts. These are the measurements I am getting WRT GND.

+3.3VDC (orange) - 16 ohms
+5VDC (red) - 30 ohms
+12VDC (yellow) - 75 ohms
-12VDC (Blue) - 1.5K
PWR_OK (Grey) 10K
PWR_ON (Green) 20K
+5VSB (Purple) 450 ohms

BTW my OP refered to PS_ON being on pin 14, my bad - thats on 20 pin AXT. Should have said pin 16.

[Latoof]

Re: Faulty Enermax Liberty ELT620AWT

I have tried to see what parts of the PSU are working..



So far I have confirmed the rectifier is fine as is the HV DC at around 310VDC. From there I assume it is reconverted into hi frequency AC it appears there is about 240VAC on the primary of the "large" transformer but as I have cheap n nasty DVM I cannot be too sure that it is reading correctly at hi frequency.



I obviously have no DC outputs beyond the +5VSB



I am still convinced that the PSU is actually fine but the PS223 is "telling" it to shut down. Tried tracing the cct with little sucess due to pcb layout etc. Wish I had a cct diagram!!



Measuring the pins of the PS223 I get the following



PIN - SIG - Voltage

1 PGI - 0

2 GND - 0

3 FPO/ - 1.15VDC

4 PSON/ - 0

5 IS12A - 0

6 RI - 0.72VDC

7 IS12B - 0

8 VS12B - 0

9 OTP - 0.2VDC

10 IS5 - 0

11 IS33 - 0

12 VS12A - 0

13 VS33 - 0

14 VS5 - 0

15 VCC - 5.19VDC

16 PGO - 0



I cant get my head around the chicken and egg opetation but I think that the PS223 initially enables the PSU at power up and if no PGI is generated from the MoBo of if any "error" states exist it shuts down.

[Krankshaft]

Re: Faulty Enermax Liberty ELT620AWT

There definately is a problem with the PSU that IC allows the PSU to turn on for a moment in that moment it monitors all of the output voltages connected to it and compares them if all of the voltages going into the IC don't match up to their respective ranges.

It doesn't generate the PS OK signal and shuts the PSU down.

That IC is known as the power good generator.

Did you check the switching MOSFET for shorts?

Also check all secondary rectifiers.

Check for any bulging caps on the secondary side as well.

If all of the above checks out you're looking at either a PWM problem (not driving the switching FET) or a feedback problem (not sending correction pulses to the PWM to keep the output voltage stable).

[Latoof]

Re: Faulty Enermax Liberty ELT620AWT

Did you check the switching MOSFET for shorts?

Also check all secondary rectifiers.

Check for any bulging caps on the secondary side as well.

Thanks for the post.

Forgive my ignorance but how do I Identify "the" switching MOSFET? I have no cct diagram and switchmode power supplies theory was about 27 years ago....

There are no bulging capacitors.

On the secondary side of the main transformer there are 6 devices on the HS and they are probably the recifiers...but not 100%. The one I can read has B2365 and S40S04S on it. Google did not turn up any info on either number that I could find... Gonna remove the whole secondary power stage HS and all from the board - easier than trying to take each device off the HS and out.

Then I can get a good look at them and test them.

[Latoof]

Re: Faulty Enermax Liberty ELT620AWT

Gonna remove the whole secondary power stage HS and all from the board - easier than trying to take each device off the HS and out.

Not today I'm not. My soldering iron is never going to get hot enough to allow me to do that! Bigger Iron or a desolder station required... Who can I call??

[Krankshaft]

Re: Faulty Enermax Liberty ELT620AWT

The switching FET will be on the heatsink closest to the line rectifier and the line filter caps there will be a minimum of 2 semis on the heatsink the large one is for the main supply and the smaller one is for the standby supply.

The rectifiers will be on the second heatsink near the secondary caps and near where the output ATX connector enters the PCB.

[Latoof]

Re: Faulty Enermax Liberty ELT620AWT

Thanks Krankshaft for taking time to post back. Appreciate it.

Given there was no obvious carnage or obvious shorts on the o/p's.. I bit the bullet last night and decided to disconnect pin 3 of the PS223 from the rest of the PSU.

My theory was that there are only 2 outputs from the PS223 and one of them "had" to be the enable/disable line. These are pin 3 and 16. Pin 16 went direct to the MoBo via ATX pin 8 and is the "power OK signal". Pin 3 on the other hand went "somewhere" and logically (presumably) enabled the PSU, but without a cct diagram who knew exactly??

Anyway, by doing so the PSU came to life and all rails are at expected voltages. :-)

This confirmed my original diagnosis that the PS223 was (1.) faulty, and (2.) disabling the PSU.

Anyway further measurements have proved that pin 4 (PS_ON) on the PS223 is internally shorthed to GND when it should be at 5VDC under "normal" conditions.

The "ON" signal from the MoBo is a "LOW" telling the PS223 to enable the PSU and then monitor the O/P for (1.) correct voltage and (2.) signs of overload/short cct.

If any "fault" condition exists then the PSU is disabled via the output on pin 3. This all happens in milliseconds so my DVM was never going to see it. **wishes he owned a CRO**

As determined earlier, there is no apparent short so I assume Pin 3 is always low disabling the PSU because one of the "monitor" inputs is being interpreted as a fault.

Further poking about with the meter showed that the monitor voltage on pin 11 of the PS223 (3.3V current monitor) was incongruent with the readings I was getting on the other current monitor inputs. I isolated both pin 11 and pin 13 from their respecive sources and now conclude that the IS33 comparator is faulty and posts a false "fault" to the PS223 logic. This generates the "low" on pin 3 disabling the PSU.

In summary the PS223 has 2 faults (1.) power on line shorted and (2.) 3.3V rail current monitor line faulty

Ideally I'd like to load test the PSU and confirm the power ccts are in fact OK but I dont want to risk my MoBo being damaged by the hard power up sequence and no fault protection. I am awaiting confirmation on a new PS223 from the US so hopefully I'll have the required spare soon.

[Krankshaft]

Re: Faulty Enermax Liberty ELT620AWT

Thanks Krankshaft for taking time to post back. Appreciate it.

If any "fault" condition exists then the PSU is disabled via the output on pin 3. This all happens in milliseconds so my DVM was never going to see it. **wishes he owned a CRO**

One note about that I have a Tektronix 475 and I never hook it up to line powered equipment (switching PSUs, TVs, LCD inverters, etc) because I don't have an isolation transformer.

Unless the equipment under test is running off a linear transformer then its not isolated and is dangerous to use on any scope.

If a fault occurs on the device it could destroy te scope or you could receive a shock simply from touching the metal body (ground) on the scope.

A dead comparator on the power good IC well as long as the monitored voltages are in spec a new IC should fix you up.

I can't recommend a full load test (other then some large 10 watt wirewound resistors on REAL LARGE heatsinks to keep them from overheating) but if you rig automotive light bulbs to the outputs they will present a load for testing.

[i4004]

Re: Faulty Enermax Liberty ELT620AWT

if tek475 is too precious for you, buy something cheap and hook it to mains powered stuff.

it's kinda moot if you ask me, because in tv-set(for example) you have (isolated for example stuff on h.output) voltages that are much higher than mains: are you thinking these pose less risk?

scope doesn't have an idea about what is hooked to it.
it will burn if you bring too high voltage on low range. so prior to scoping measure the voltage and you'll be fine.

something else to think about: if you're scoping the circuit and suddenly short-circuit occurs there, then essentially you have voltage dropping to zero, and that can't destroy the scope.

so in the end mains are as dangerous to scope as to any other electric-meter; wrong range can make a mess, but other than that....

[Latoof]

Re: Faulty Enermax Liberty ELT620AWT

The real question I need answered is where can I get a replacement PS223 IC from.

I have had no luck so far finding one. I found one supplier so far but the MSQ is 2000 units....

I'm now trying to source a similar function IC and will build a functional replacement if required.

Is anyone interested in poping the covers off their PSU and working out what power monitoring IC it uses?

[Krankshaft]

Re: Faulty Enermax Liberty ELT620AWT

Try checking Mouser or Digikey.

[Latoof]

Re: Faulty Enermax Liberty ELT620AWT

Try checking Mouser or Digikey.

Thanks but neither site has the PS223 listed...

[andlcs]

Re: Faulty Enermax Liberty ELT620AWT

If its a Liberty can be the switcher or 12V rectifier that blew... Or the secondary capacitors (from CEC) leaking.

[Latoof]

Re: Faulty Enermax Liberty ELT620AWT

If its a Liberty can be the switcher or 12V rectifier that blew... Or the secondary capacitors (from CEC) leaking.

thanks for your ideas. The PSU works fine IF it can turn on but it cannot becaause the monitor/controller is faulty. I have proved this. Just need to find a new one somewhere... PS223 I.C.

[gonzo0815]

Re: Faulty Enermax Liberty ELT620AWT

I think this is getting realy dificult to get your hands on, as it is too specialized to be carried by the large electronic retailers.

[Latoof]

Re: Faulty Enermax Liberty ELT620AWT

Still trying to repair this sucker... anyone got a faulty Enermax, Silverstone, or Antec that I can salvage this IC from?

[Latoof]

Re: Faulty Enermax Liberty ELT620AWT

I have finally got some good news on the PSU...



The Fairchild SG6520 is a direct replacement for the PS223 so have a couple on order from Mouser elec in the US and should have them in a couple weeks