Power Computing Google Group

Re: Power supplies for PW machines

2008-05-03T05:31:04Z

Removing a board changes nothing else? To be careful about not
changing the fault, we change only one or a few things; then test.
Then if a fault actually located elsewhere disappears, then reason for
that change is easier to locate.

Meanwhile, that green wire jumper is unnecessary. Remove some

Re: Power supplies for PW machines

2008-05-02T03:40:44Z

I don't know about your memory for details, but the safest way, is to
re-install all the boards for storage ---- that way, if you get
another MLB, it's a simple set them side by side, and swap everything
over to the new MLB. Rather than a BIG PUZZLE of where everything
goes. [Of course, a good set of photos would work well as 'memory

Re: Power supplies for PW machines

2008-05-02T00:21:03Z

I'm assuming that it is more likely that the fault is in the MLB than in one of the peripheral boards -- is that a good assumption? If so, then I think it would be faster and easier if I removed all the peripheral boards & cards first and then made my first voltage measurements. If the voltages are still incorrect I could then put everything back, (or simply throw them in a box, awaiting another MLB someday far into the future!)

Re: Power supplies for PW machines

2008-04-25T04:27:45Z

Anything that plugs into the MLB - those are peripheral cards.
Although the computer may not operate, still, the board should have
voltages when PSU powers without all those other attached
peripherals. Remove everything - one by one.

Re: Power supplies for PW machines

2008-04-25T04:07:30Z

So what exactly do you mean by "peripheral boards"? Does this include the cache card, the video card, and all the DIMMs? Does it also include the CPU card? Anything else?
Dave L.
-----Original Message-----

Re: Power supplies for PW machines

2008-04-24T19:36:28Z

Defect is now seen. As peripheral boards are removed, either the
leakage should disappear (purple wire rise back to 5.0 volts), or the
MLB remains as the only defective part.

Next move on to inspection. For example, has a metallic fragment
lodged under an IC? Change only one thing and test again. Eventually

Re: Power supplies for PW machines

2008-04-24T12:33:25Z

Thanks w_tom,
Yes, I too wondered about the violet wire, but since it perked up after I started the machine would that be evidence that the root problem is not inside the Power supply box but is a short elsewhere?
And I also was wondering if a voltage leakage could cause the unexpected and slightly fluctuating voltages in several wires during the OFF state.

Re: Power supplies for PW machines

2008-04-22T01:10:01Z

Purple wire voltage must always measure above 4.87 volts. At, 3.57
volts, circuit that controls power become unstable, if it works at
all. Either power supply is defective or something connected to +5VSB
(purple wire) has failed (consuming more current than +5VSB can supply
while also shorting red wire +5V to that +5VSB).

Re: Power supplies for PW machines

2008-04-21T22:51:36Z

Here are the voltages I have measured from the ATX power supply in a PowerWave machine. The symptom is failure to start when pressing the startup button on the keyboard and the buttons on the bezel don't start it either. (I hope this table shows up correctly -- It appears to be a monospaced font here, probably Courier.)

Re: Power supplies for PW machines

2008-04-21T22:22:38Z

Thanks for your ideas Oliver.
First, to answer you questions:
* No, pressing any buttons on the front bezel does not produce any response.
* Yes, the PRAM battery was replaced about 3 weeks ago.
* I'm not sure about the non-volatile memory question (I surmize that this may be more of a PC term). The only memory I'm familiar with other than the regular DIMMs is called PRAM, and it depends on the integrity of the small lithium battery I've already replaced. I did press what is called the CUDA button and that supposedly cleared parts of the PRAM memory that would not otherwise be flushed by leaving the battery out for a while (which I had already done), by using various key combinations on startup, or by using some utility like TechTool.

Re: Power supplies for PW machines

2008-04-19T00:46:52Z

Yes, I have measured all the voltages for all colors of the wires entering the 20 pin connector while it is plugged into the MLB and I will post these soon (don't have them with me at the library right now.) However, my conclusion was they were all normal after I started the machine by shorting green to ground. Nothing happens when I press the start button on the keyboard (i.e., nothing starts), but I will follow your advice and measure each voltage while holding down the start button. I'll post it all in a day or two.

Re: Power supplies for PW machines

2008-04-17T15:20:43Z

Hi,
I can provide some general ideas about what may be involved with such a
problem, but please keep in mind that I don't have specific knowledge
about ATX power supplies or Macintosh hardware. I do have a background
in electronics though.
As you have no doubt discovered, ATX power supplies (that is what you

Re: Power supplies for PW machines

2008-04-17T04:51:16Z

It's called a power supply 'system'. "System' has multiple
components. Best way to find which component is defective is to
remove or disconnect nothing. Then measure one of each color wire as
defined in "When your computer dies without warning....." starting 6
Feb 2007 in the newsgroup alt.windows-xp at:

RE: Power supplies for PW machines

2008-04-17T00:28:28Z

Well, I've moved on a bit in my understanding of the start up problem with this machine. I thought it was the PS, but after reading some more info on various web pages I've now realized that my power supply may not be the component at fault. I read that ATX power supplies may not fully start (or may shut down) if there is not some sort of load on at least one of the outputs. I had measured low voltage on two wires (yellow, and I think the other one was blue )and so concluded on that basis that the PS was defective. But then when I plugged a hard drive into one of the 4-pin outputs from the PS, all the output voltages became normal. So I put the PS back in the PowerWave, connected the 20-pin power plug to the MLB and another output cable to two hard drives. Again it would not start, so this time I used the trick I had used while bench testing the PS, i.e., shorting the green to ground -- and presto it started up. Then I measured voltages for all colors of wires going into the MLB and all were normal. Perfect. So now I'm doubting that the problem is the PS at all. I'm wondering what component should be in charge of shorting the green wire from the PS to ground when one presses the start button on the keyboard. Does anyone know if this switching system would be on the MLB or could it actually be somewhere inside of the PS? I know that the keyboard button and the keyboard connecting cable (ADB) are both functioning because the same keyboard & cable will start up another Mac I have. And I know that the violet wire from the PS supplies around 5 v. of standby power even when the computer is plugged in but not running. Does anyone happen to know if I should be looking for the problem somewhere outside of the PS, and if so where?

RE: Power supplies for PW machines

2008-04-16T11:04:41Z

No worries. Hope it helps.
Oliver
-----Original Message-----
Behalf Of dave_lis...@earthlink.net
Sent: Sunday, April 13, 2008 6:52 AM
To: powerlist@googlegroups.com
Thank you. Those are all good tips that I will stay aware of.
Dave L.
-----Original Message-----
variations are
components and
over +4V. Some