Agent24
I see dead caps
Last Activity: Yesterday, 11:25 PM
Joined: 10-04-2007
Location: Hiding inside a plated-through hole
-
Every time I see a news article about Windows or Microsoft in general these days, I'm glad I switched to Linux. -
Oh for sure it's 470uF for the one you posted a photo of. 47uF is too small to be of use filtering power rails on a GPU.
I haven't seen any electrolytics at all use codes with a multiplier digit, except maybe some tiny SMD ones.Leave a comment:
-
Good to know! Might order some myself just in case.Leave a comment:
-
A vrm is a vrm. It can't really be "designed badly", only underrated parts used.
Rubycon ZL would be good I think. I used those to replace Sacon FZ before.Leave a comment:
-
I know, I was hoping their website would tell what it mates with, but Molex website seems to be useless now....Leave a comment:
-
I haven't done any BGA work at all, but I imagine replacing individual pins may well be easier assuming they can be released from the socket assembly somehow, which seems like an interesting concept.
Do you have a link to any videos showing this?
I guess you can watch the videos and try to follow along. Does your heater have temperature or power control of any kind? You will need it so you don't overheat the board and melt/warp the socket.
However you try, I think you should definitely practice on something else first. The risk of ruining the board on a first attempt...Leave a comment:
-
Oh lovely, it's got the black self-destruct glue on it...
Quite possibly you have some corrosion damage under or near some of that glue, which could be causing the problem.
It's also possible some voltage rail is bad, there will be some lower voltage rails on the board for the digital processor and bluetooth etc like 5v or 3.3v etc. Check that these are OK, you may have bad capacitors on them if they are low or noisy.Leave a comment:
-
Nice!Leave a comment:
-
Are the ICs ultra cheap in a pack of 10 from China? If so they're probably fake.Leave a comment:
-
Datasheet for that connector says "for use with 8993 series terminal" but I can't find that number on Molex catalogue....
Leave a comment:
-
I'm guessing it's some Molex part or equivalent, is there a manufacturer's mark on it?
I know some ATX PSUs used the same connector before they had 24-Pin. Maybe the ATX spec sheet has some information. I don't know which revision used that connector though.
Edit: ATX Spec document lists the connector as the Molex 90331-0010 (Page 21)
Seems to be replaced with https://www.molex.com/en-us/products...tail/903311003
Not sure what model number the part on the PCB is, though....Leave a comment:
-
It reminds me of the crappy Sacon FZ with that fake polymer style can, but 2001 is years before their time.Leave a comment:
-
How nice of the bad one to identify itself so easily!Leave a comment:
-
PCBs do darken over time in areas where components run hot, this is normal. As long as it's not carbonised it's OK. However it can also be indicative of a fault and something running hot that shouldn't. It looks like the big dark area has a few power resistors there so that's the kind of thing that makes sense to happen in normal operation (probably).Leave a comment:
-
I think this means the power management IC is not turning it on, because some required condition is not met, it detects a fault, or the IC is bad, or the circuitry to provide information to the IC is bad (like open resistor in sense circuit etc)
Unfortunately I can't view your PDF file because this website tells me I have to subscribe now to download attachments. Please post it as a JPG or such.Last edited by Agent24; 03-03-2026, 09:44 PM.Leave a comment:
-
The fact it was working until you had a power outage does suggest bad capacitor(s), although other causes are possible.
One failure mode of electrolytic capacitors is high ESR, and ESR changes as a function of temperature. When the capacitor is warm, the ESR will be lower than when it's cold. If the capacitor is marginal, but the supply is kept powered on, it may stay warm enough for its ESR to stay within acceptable limits. However after a power failure, and subsequent cooling down, the ESR rises and the circuit no longer works normally.
Of course other components can...Last edited by Agent24; 03-02-2026, 03:07 PM.Leave a comment:
-
I guess a fault in the relay drive circuit or its power supply, again, a bad capacitor could be the reasonLeave a comment:
-
Are the optocouplers actually being driven though?Leave a comment:
-
That seems very strange. Normally PS_ON signal is derived from 5VSB rail. So I can't see how you can have PS_ON but not 5VSB. If you don't get 5VSB on the main ATX connector it could be the wiring is bad. I see this is a modular PSU. Maybe the cable or plug/socket is faulty. Or some internal connection in the PSU.Leave a comment:
No activity results to display
Leave a comment: