iMAC G5 ALS PowerSupply API4PC46-290 REV.A R37 near 3845B

[Toasty]

Re: iMAC G5 ALS PowerSupply API4PC46-290 REV.A R37 near 3845B

Q1 fails are not -that- common and are distinguishable by a clouded bar on the package surface in line with the collector. (or blown to smithereens ) That invariably means the voltage went to high because of the bad caps, causing the 3845 internal (protection) zener to trigger causing Q1 to overcurrent. *poof*

What has been done to this new supply?

[torgeson5kqs]

Re: iMAC G5 ALS PowerSupply API4PC46-290 REV.A R37 near 3845B

I've checked all the caps on the secondary side of the unit (ones that normally bulge) they all check out according to the ESR meter. This is the first supply I've run across that did not have bulging caps that was bad. The only thing done was the replacement of the Q1. All the caps are Ltec, but again they are within tolerance on the ESR meter.

[Toasty]

Re: iMAC G5 ALS PowerSupply API4PC46-290 REV.A R37 near 3845B

...and the little ones as before? Replaced?

Does it even try to start? Meter the outputs and see.

>>they all check out according to the ESR meter.<<

According to Ltec's sheet?

[torgeson5kqs]

Re: iMAC G5 ALS PowerSupply API4PC46-290 REV.A R37 near 3845B

The caps below with voltage readings have all been replaced with Panisonic FC caps. I still just get standby voltages on pins 9 and 22 both in standby and when "ON". All the other Ltec caps in the secondary part all meet Ltec Max impedence specs on this url measuring with Blue ESR meter http://www.ltec.com.tw/lyz.PDF


1st is standby, 2nd value is "On"

C1 - 0, 0
C11 - 0, 0
C37 - 21.3, 21.4
C69 - 5.7, 5.7

[Toasty]

Re: iMAC G5 ALS PowerSupply API4PC46-290 REV.A R37 near 3845B

Locate and measure voltage from output negative to Pin 2 of opto M3 while in standby as follows -

Orient the board with the solder side facing you, the outputs to the left and the hot to the right.
There is a vertical slot in the board for isolation of "HOT" and "COLD" power.
The opto's sit across this.
Looking from top to bottom they are:
M3
M4
M11
M7

At M3, the square pad is Pin 1 and the one above it is Pin 2. (Just to the left is Diode D1)

Report voltage please.
.

[torgeson5kqs]

Re: iMAC G5 ALS PowerSupply API4PC46-290 REV.A R37 near 3845B

I get a reading of 4.6 volts at pin 2 of M3 in standby mode.

[Toasty]

Re: iMAC G5 ALS PowerSupply API4PC46-290 REV.A R37 near 3845B

Good.

Obtain a 1/2w or better resistor with a value between 33 and 68Ω.

Connect a jumper wire (alligator clips) from output negative to one leg of the resistor.

Monitor the 3.3v, 5v, & 12v output voltages with your meter(s).
Briefly touch the other leg (ungrounded) of the resistor to M3, Pin 2.
The supply should come on.
Note each voltage and report.

.

[torgeson5kqs]

Re: iMAC G5 ALS PowerSupply API4PC46-290 REV.A R37 near 3845B

Only pin 15 came on at 5.2 volts along with the regular standby voltages, all the others remained at 0 volts. I only had a 1/4 watt resistor, couldn't find a 1/2 watt.

Pins

8 - 5.2
15 - 5.2
22 - 20.4

[Toasty]

Re: iMAC G5 ALS PowerSupply API4PC46-290 REV.A R37 near 3845B

Umm, please indicate what those pins mean...?

What value resistor did you use?

.

[torgeson5kqs]

Re: iMAC G5 ALS PowerSupply API4PC46-290 REV.A R37 near 3845B

Pins on the output connector where you asked me to monitor the voltages.

[torgeson5kqs]

Re: iMAC G5 ALS PowerSupply API4PC46-290 REV.A R37 near 3845B

resistor was 56 Ohm's

[Toasty]

Re: iMAC G5 ALS PowerSupply API4PC46-290 REV.A R37 near 3845B

I have no idea what those pins mean (nor does anyone else) without pulling out the sheet and looking it up.

Please tell me what voltage is supposed to be on each one.

If I understand you correctly, 3.3/5/12 were -ALL- zero (and stayed there) with the supply jumpered from output negative thru the resistor to M3 Pin 2? Correct?

56Ω is good.

[torgeson5kqs]

Re: iMAC G5 ALS PowerSupply API4PC46-290 REV.A R37 near 3845B

Sorry I don't know how else to show you what I meant, the correct values are shown below. Pin 22 should be 24v but it seems to be anywhere from 20 to 24 volts either when in standby or when on.

Pin / Correct Volts / Actual Volts

pin 8 / 5.1v / 5.2v
pin 15 / 5.1v / 5.2v
pin 22 / 24v / 20.4v

[Toasty]

Re: iMAC G5 ALS PowerSupply API4PC46-290 REV.A R37 near 3845B

The 20.4v is correct until the secondary voltages come up. Then it will be ~24v.

5.1v or 5.2v is fine.

Again:
If I understand you correctly, 3.3/5/12 were -ALL- zero (and stayed there) with the supply jumpered from output negative thru the resistor to M3 Pin 2? Correct?

They didn't jump or even attempt to come on? That is -very- odd.

.

[torgeson5kqs]

Re: iMAC G5 ALS PowerSupply API4PC46-290 REV.A R37 near 3845B

Yes, that is correct. there was no jump on the meter at all, I'll try again tonight with another identical power supple that is also in need of repair to see if I get similar values and report back to you.

[torgeson5kqs]

Re: iMAC G5 ALS PowerSupply API4PC46-290 REV.A R37 near 3845B

I went ahead and tried again tonight to get voltages on another power supply. It's almost like you need 3 hands to do this. The only alligator clips I have are all attached to test equipment. So I had fiddle with it. Anyway now I'm not even sure I'm doing this right. I've attached a pictures with the resistor on the underside of the board to better show what I've done. Again, I'm not confident I'm following your instructions correctly. On the opto for M3 (underside) is pin 1 (square) the negative output and you short it with pin 2 using the resistor? If thats the case I'm not getting any voltages at other than the ones I gave you in the previous post. Those previous voltages I gave you are from standby regardless if I do anything with the opto.

[Toasty]

Re: iMAC G5 ALS PowerSupply API4PC46-290 REV.A R37 near 3845B

>>Connect a jumper wire (alligator clips) from output negative to one leg of the resistor.<<

Means: the output of the power supply negative terminal(s).

You could also solder one end of the resistor right to the negative which is 3/4" to the left of M3 / D1. The copper heatsink (on top) to the left is negative and you can see where the 2 feet for it come thru the board and are soldered in. Any trace on that will do.

.

[torgeson5kqs]

Re: iMAC G5 ALS PowerSupply API4PC46-290 REV.A R37 near 3845B

Ok I finally got some time to try this again and soldering the resistor to the feet of copper heatsink really helps. I was able to get voltages on another similar PSU but the layout is slightly different. However, on the PSU in question I still have no voltages coming out. They are all zero except for the ones that are the standby voltages which I mentioned in a previous post.

[Toasty]

Re: iMAC G5 ALS PowerSupply API4PC46-290 REV.A R37 near 3845B

Ok. I'll check some other voltages and get back to you. Might be a bad M3.

[Andy3665]

Re: iMAC G5 ALS PowerSupply API4PC46-290 REV.A R37 near 3845B

Have you checked to see if there is a short in the power factor controller in the primary side. I've found about 10% of these have a blown PFC chip.

Also I found in some cases, that the 3.3V rail is pulled low. It appears the housekeeping circuit uses this as a reference voltage and the PSU increases the output, but by doing so over volts the 5v and 12v rails, which get detected and then the PSU instantly shuts down. This all happens in a split second, but can be viewed using a digital storage scope.