PDA

View Full Version : Ticking power supply from Mac LC475



carlsson
August 20th, 2006, 10:44 PM
This is the continuation of the thread I started here:
http://www.vintage-computer.com/vcforum/showthread.php?t=4101

Initially I only meant to post about the PC cluster, but turned out the little power supply was more interesting. Therefore I decided to start a new thread as the forum software doesn't have a function to let a user reply to a thread but put his answer in another group - that is exclusively a moderator privilege.


When you check the DC output with your voltmeter, do the voltages vary rapidly, or is it kind of random?
What is the difference of rapid and random? The digital multimeter reads out values that don't follow a certain pattern. I compared with the original 1541-II PSU, which outputs perfectly stable voltages.

I have a complete Mac LC475. I could try to measure its power supply to see if it fluctuates as much or if this loose one is fishy.

I tried to record it as loud as possible, and then filter away all the other noise. The end result sounds like a bubbling aquarium, but anyone interested can clearly hear the rate of ticking.

http://www.anders.sfks.se/mp3/ticking.mp3

Oh, and .. after powering it off, it takes a long time before the ticking slows down and stops completely. At one point in the slow down, it starts ticking at double speed (!) and then slows down again from there.

modem7
August 21st, 2006, 12:43 AM
Yes, that's definately a 'tick'. Apart from arcing, the only other generator I can think of is relay contacts.
One other very remote possibility - does the PSU have 'made in Iraq' on it?

carlsson
August 21st, 2006, 12:46 AM
No, its made by TDK (!) and assembled in Malaysia. Model number: 699-0153, which is the same as its Apple Part Number.

http://www.cbm.sfks.se/pics/lc475-psu.jpg

modem7
August 22nd, 2006, 04:01 AM
One other very remote possibility - does the PSU have 'made in Iraq' on it?
Tick, tick, tick, tick ..... Boom

nige the hippy
August 22nd, 2006, 04:32 AM
Do I gather that you're measuring the voltages from this PSU off load?

... in which case it's very likely that everything will be wrong. most switch-mode psus require a load, usually on the 5V rail as that is the most critical voltage, and the one that is measured to provide feedback for the switching input stage.

switch mode power supplies usually tick as a result of them repeatedly going over-voltage (due to open circuit) or over-current (usually due to short circuit), resetting, and then trying to start again.

try it on a load of about an amp (e.g. an old hard drive, or 4.7 ohm 5W resistor between 5V and ground ) and measure again.

if it still ticks, and it's not a particularly valuable supply, lose it, as it can be hours of (potentially eyebrow frying) work to fix one.

Nig.

carlsson
August 22nd, 2006, 09:38 AM
Yes, I measured it unloaded and considered there might be a different readout if the circuit is loaded.

After replacing the connector, I plugged in an old, unknown 300 MB SCSI disk made by IBM 1988. At first it didn't power on at all. Then I got a cloud of fine smoke... bye, bye SCSI disk.

Then I tried a known good hard disk, which powered up all right without smoke. The 5V line has a voltage range of 5.13 to 5.16V, which is much better than before, but not 100% stable. The 12V however is rock solid at 11.63V.

nige the hippy
August 22nd, 2006, 10:54 AM
probably it's ok then, maybe it regulates on the 12V, maybe the load on the 5V changes more than the load on the 12V, a variance of .03 volts is not exactly wobbly. Does it still tick??

I suspect that the scsi disk was at fault when you tried that, short, or nearly short on the 5V, the wait was while the PSU pushed masses of current through it.

At this point it's down to a guess, " do you feel like it works properly? "

if so, use it,

if not, either give it to someone who really knows what they're doing with switchers (they really are horrible to fix, accidentally open the feedback loop, and there's hot plastic, and fluffy capacitor filling everywhere), lose it, or put it in a box and forget about it till you're prepared to go round the loop again.

best wishes, Nig

carlsson
August 22nd, 2006, 01:02 PM
As I wrote, I have a complete LC475. Thus, I opened it, disconnected the power from the motherboard and powered on the PSU to compare. It has a different part number and is rated for higher currents!

Loose PSU: +5V 3.25A, +12V 0.75A, -5V 0.075A
In computer: +5V 4.75A, +12V 1.0A, -5V 0.075A

Without load, it ticks a little, but much slower and much more silent. The voltage readout is total rubbish though. While the Mac is running, the power supply delivers 5.15V and 12.06V.

nige the hippy
August 23rd, 2006, 02:59 AM
sounds like your loose one works ok.

don't get obsessive about the ticking! it's normal, just big start-up current pulses through the transformer.

When the thing is running, it still ticks, it's just that the ticks are joined up together into a (usually ultrasonic) whistle.

carlsson
August 23rd, 2006, 07:26 AM
Right now I'm more concerned if the 0.4V difference on the 12V line will be too low to successfully run the disk drive motor. Maybe it will run too slow? The original PSU outputs 5.02V and 12.19V according to my multimeter. Maybe a few decivolt doesn't matter much to these old electronics; it is not like a modern CPU that is rated with a voltage within two decimals.

nige the hippy
August 23rd, 2006, 08:18 AM
yes, 11.63 on the 12V looks a little bit low, and maybe there is something wrong there, or it could just be a cheap and nasty psu , but as there is a tacho on disk drive motors, i doubt it would make a difference. you could replace the rectifier and the capacitors on the 12V output, but I doubt it's worth it, I'd wait for the bang!:shock:

carlsson
August 25th, 2006, 10:18 AM
I bought some cable patch (don't know what it is called in English, when you put the ends of two cables into one hole each connected with a metal plate, and tighten the screws to keep the cables into place) and hooked up a few disk drives.

Upon power on, the drive LEDs flicker twice, but is then stable. The drive functions fine. Maybe the flicker is due to the drive tries to pull the required voltages and currents, and the power supply needs a tenth of a second to build up that voltage? When I connected the drive via my PC, it didn't flicker in the same way. Perhaps it is best used rarily, or until I come across a better solution.. preferrably I want a PSU strong enough to power two drives, but also not so large and cheap, close to for free. :)

nige the hippy
August 31st, 2006, 10:08 AM
.....Lost the thread......


You could always use a standard PC power supply- even if it is an ATX you just short the control input to ground (?) and it turns on.
Loads of watts cheap!

Web link...

http://www.wikihow.com/Convert-a-Computer-ATX-Power-Supply-to-a-Lab-Power-Supply

bbcmicro
August 31st, 2006, 10:41 AM
I bought some cable patch (don't know what it is called in English, when you put the ends of two cables into one hole each connected with a metal plate, and tighten the screws to keep the cables into place) and hooked up a few disk drives.

I think I know what you mean, and the name eludes me!
Is it made of plastic and comes in strips?
This is going to annoy me for the rest of the day now.

nige the hippy
August 31st, 2006, 11:45 AM
chocolate block.

carlsson
August 31st, 2006, 05:07 PM
I think we call it piece of sugar in slang Swedish. We also call Ethernet extenders (where you put two ends of Ethernet cable into it) for a piece of sugar.

Anyway, what I'm talking about is this:
http://www.biltema.se/products/productimage.asp?iItemId=84141

And Nige, yes I am also using power out from my desktop PC to power another drive, but wanted to see if I could make a fairly small footprint standalone PSU from this one.