Beckman Coulter Optima XE-90 Ultracentrifuge Repair - Power Supply Failure (Again)
HTML-код
- Опубликовано: 7 авг 2024
- What's up everyone! In this video I perform a repair of a Beckman Coulter Optima XE-90 power supply that provides power to the unit's SCB and TEM. The supply has part # B05755 and is $7500 to replace (that's just for the part!)
I did a 3 part video series on repairing this same supply about 6-7 months ago. And it has failed again! Same module but different cause. However, both issues seem related to this power supply's poor heat dissipation and possibly the manufacturer (i.e. Beckman) not selecting a supply with a large enough output.
0:00 - Overview/recap of first failure on 11/2023
3:10 - Overview of today’s failure
5:00 - side by side comparison of 60V module transformers
7:00 - note on module heat dissipation
7:50 - output voltage test before installation
8:45 - fuse blow #2/load testing working 60V module
14:30 - loading testing non-working 60V module
18:30 - another secondary fault found! (gate MOSFETs shorted)
22:00 - post-repair testing… no dice…
23:40 - testing the UC2845s
25:45 - primary fault possibly found!
27:07 - primary fault repaired; testing the UC2845
29:00 - load testing the non-working 60V module post-repair
36:20 - load testing the working 60V module
37:40 - load testing the 24V module
38:45 - testing the supply installed in the centrifuge - success!
40:00 - outro 
Наука
Nice job young man👍
Thank you Joel!! Always appreciate your feedback!
Oh ! get the boss to buy you a thermal camera ! Are those transistors mounted on a heat sink ?....Nice vid !.....cheers.
yes! been meaning to get one of those for a while!!!
@@fmashockie :)
Mechanical stress on caps due to differing expansion coefficients of ceramic and copper with pc board substrate is a known cause of smd failures.
Was not surprised to see that it was the cause of the fets smoking and burning up.
That high frequency would have run the transformer hot, luckily it did not get hot enough for permanent damage it seems.
It was an interesting video, fault finding is fascinating stuff.
Best regards,
Duncan
South Africa
Took me a while to make that connection but glad you still enjoyed the video! This is my favorite part of my job but I don't get to do it as much as I'd like. We'd be in bad shape if I was working on PCBs like this everyday!
@@fmashockie I use a cheap usb microscope with 800 times magnification to examine boards, and take screenshots of what I find.
Not a week goes by that I do not use the microscope to look for stuff or take images to put into reports.
Or look at the numbers on components.
@@Uncle-Duncan-Shack I might need to make an investment in a microscope that powerful. I think mine is only 10x
This was an interesting Video. I think it also verifies rule 0: Do a very detailed visual inspection. Followed by rule 1: Check your voltages 😂.
Maybe somebody has a suggestion for rule 2?
For switching power supplies, rule 2 might be check your frequencies (only makes sense if you have a known good unit). Aside from microscope inspection that someone else mentions, a thermal camera would often be helpful.
Glad you found it! That's some serious bad design there... falling off? Holy fuck!
For a bit I thought you had counterfeit FETs there (Even from the big 3 it happens, ask how I know...)
Great video, a tiny bit scary ;)
Thanks Charles! It's crazy right!? A SMPS that costs $7500 shouldn't be having this many failures so early in its lifespan!
Around a decade ago, I watched someone open up an audio amp. I don't recall whether they planned to fix it or wanted parts out of it, but we were all amazed when they saw that surface mount resistors had fallen off or tombstoned. I think they surrounded a microcontroller or something else that shouldn't get hot, so it was quite strange. The board also wasn't discolored IIRC. If the unit was produced when RoHS was new, I wonder if the wrong solder was used, with a melting point that was too low.
Question for the viewers out there who might have more experience: am I wrong to say the manufacturer of this centrifuge should have picked a supply with greater output? The thermoelectric module (TEM) that cools this beast can draw 8A 60V at max setting. The supply is only rated for a max of 8A. I think they should have went with a higher rating but curious to hear your thoughts!
I'd say it depends on how it's rated. If the manufacturer simply says 8 A max, I'd expect it to actually live up to that for its rated lifetime, as long as it stays in the temperature range (and whatever other parameters). If they derate it under certain conditions, it should still be able to do 8 A for lifetime, provided you avoid the derating conditions.
@@markp5726 thank you! I was just curious because I've worked on a lot of lab equipment power supplies, and I've yet to see one that was not overrated a little bit. So I thought this was odd.