Need help identifying a PSU.

Discussion about old PC hardware.

Need help identifying a PSU.

Postby athlon-power » 2019-1-08 @ 23:00

I have a PSU that had the main information sticker ripped off at some point, and so far I have found zero ways to identify it. I took it apart, cleaned it to the best of my ability, and put it back together. All of the capacitors are fine, so if it does happen to be a good-quality PSU, it seems to be in good shape. I must mention that this one had a lot more going on inside than a guaranteed garbage one I took apart the other day. It also weighs a fair bit more, and it's not because of the shell being heavy, or any artificial weighting inside.

It appears to be an OEM PSU of some kind, given that it gives a "Dell-ish" part number on the side, with a serial number. I tried to search up both, to no avail. There's also some other stickers that I found on it that might provide insight into its brand, quality, etc.

IMG_20190108_175224.jpg


IMG_20190108_175257.jpg


IMG_20190108_175312.jpg
Custom Builds:
Intel SE440BX-2, Intel Pentium III SL35E, 128MB PC100, ASUS V3800M 32MB, AOpen AW744L II, Samsung SC-148, WD Caviar WD400, ca. 1999

Upcoming builds:
Athlon XP 3200+ (Socket 754), 512MB DDR-400, rest unknown, ca. 2003
User avatar
athlon-power
Member
 
Posts: 150
Joined: 2018-9-29 @ 00:15

Re: Need help identifying a PSU.

Postby SirNickity » 2019-1-09 @ 00:55

Definitely looks like a Dell OEM to me. Any reason, aside from curiosity, that you care about the branding? Most likely it's OEM'd by one of those operations that cranks out PSUs by the gazillions. Might be something on the PCBs that would implicate a manufacturer, if its lineage is more relevant to you than the badge.

I don't know what vintage this is (seems like 2000s -- is it ATX?), but if it's old enough to have any of those older style X/Y caps, you might want to replace that one(?) at least. It's the across-the-mains type that usually goes up in smoke at this age. If it's one of the boxed film types (usually a large gray, black, blue, or white rectangle near the mains input on the PCB) then you're probably OK. The old ones look more like a big epoxy glob -- yellowish, typically, and with a giant burnt hole in the side.
SirNickity
Member
 
Posts: 327
Joined: 2018-11-08 @ 04:51

Re: Need help identifying a PSU.

Postby gdjacobs » 2019-1-09 @ 01:51

If you open it up, try to look for a UL number.
User avatar
gdjacobs
l33t++
 
Posts: 6041
Joined: 2015-11-03 @ 05:51
Location: The Great White North

Re: Need help identifying a PSU.

Postby athlon-power » 2019-1-09 @ 10:56

SirNickity wrote:Definitely looks like a Dell OEM to me. Any reason, aside from curiosity, that you care about the branding? Most likely it's OEM'd by one of those operations that cranks out PSUs by the gazillions. Might be something on the PCBs that would implicate a manufacturer, if its lineage is more relevant to you than the badge.


I really don't care about the specific brand- as long as it's an okay brand- something that won't blow up and take all of my components with it. It being an OEM PSU is fine, as long as it's a decent OEM PSU. Dell PSUs are usually alright, so I'm hoping that this one shares the same qualities.

SirNickity wrote:I don't know what vintage this is (seems like 2000s -- is it ATX?), but if it's old enough to have any of those older style X/Y caps, you might want to replace that one(?) at least. It's the across-the-mains type that usually goes up in smoke at this age. If it's one of the boxed film types (usually a large gray, black, blue, or white rectangle near the mains input on the PCB) then you're probably OK. The old ones look more like a big epoxy glob -- yellowish, typically, and with a giant burnt hole in the side.


I honestly don't know enough about the innards of PSUs to figure out what looks good and what looks bad in these things. This is the first time that I've disassembled one to really take a look at the internal components. Also, this is an ATX PSU. I'm not lucky enough to own an AT one. :lol:

I've taken it apart again, so that you can see the internals of it, as it's all just a confusing mess for me at this point. I can't tell you what does what, and all I see are some beefy capacitors and wire coils. I turned it on last night, the fan was putting out healthy amounts of air where I had cleaned it, and I hooked up two IDE DVD drives to it (at once), and they worked just fine. I don't have a multimeter, so I can't tell how clean any voltage coming from this thing is. The best thing I can do is run sketchy tests, like with using the DVD drives, for example.

IMG_20190109_054328.jpg


IMG_20190109_054349.jpg


The power from the mains runs directly into the PCB that's separated from the rest of the internals at the top, where those blue and white box-looking things are. Is this a boxed-film PSU, or am I just being dumb, and looking at the wrong thing?

IMG_20190109_054605.jpg


As far as I can tell, it does use those capacitors with the little X's and Y's on the top, though none are bulging or damaged. I am not going to fully disassemble this thing again to snap shots of those, because it is very tedious to take this thing apart and put it back together, especially when removing the plug and voltage selector. Unfortunately, this means that I probably won't get a UL number, but I don't want to do all that over again.
Custom Builds:
Intel SE440BX-2, Intel Pentium III SL35E, 128MB PC100, ASUS V3800M 32MB, AOpen AW744L II, Samsung SC-148, WD Caviar WD400, ca. 1999

Upcoming builds:
Athlon XP 3200+ (Socket 754), 512MB DDR-400, rest unknown, ca. 2003
User avatar
athlon-power
Member
 
Posts: 150
Joined: 2018-9-29 @ 00:15

Re: Need help identifying a PSU.

Postby gdjacobs » 2019-1-09 @ 18:38

That's an impressive input filter.
User avatar
gdjacobs
l33t++
 
Posts: 6041
Joined: 2015-11-03 @ 05:51
Location: The Great White North

Re: Need help identifying a PSU.

Postby athlon-power » 2019-1-09 @ 19:39

gdjacobs wrote:That's an impressive input filter.


I took apart an In-Win (aka Power Man) PSU and it had one of those film boxes soldered right to the power input, and from what I can tell those are decent quality PSUs. Does this mean that it's safe to use the unnamed one? I'm just worried, because if I use it, it'll end up in a very valuable PC- one that took me a little under a year to get together. It's actually the first machine that shows up in my signature. As I've said, this is the first time I've delved into PSU quality, and knowing what separates the good ones from the bad ones. I guess I was right when I mentioned that the unnamed one had a lot more going on inside than a bunch of the others that I've got.

If the PSU itself fails, I'm okay with it. I just don't want it to take any components with it.
Custom Builds:
Intel SE440BX-2, Intel Pentium III SL35E, 128MB PC100, ASUS V3800M 32MB, AOpen AW744L II, Samsung SC-148, WD Caviar WD400, ca. 1999

Upcoming builds:
Athlon XP 3200+ (Socket 754), 512MB DDR-400, rest unknown, ca. 2003
User avatar
athlon-power
Member
 
Posts: 150
Joined: 2018-9-29 @ 00:15

Re: Need help identifying a PSU.

Postby gdjacobs » 2019-1-09 @ 20:33

It's hard to tell for sure without testing, but solid construction is always a good sign.

Self contained filter modules are a high quality, slick solution. Any OEM using one is doing that part of their job right.
User avatar
gdjacobs
l33t++
 
Posts: 6041
Joined: 2015-11-03 @ 05:51
Location: The Great White North

Re: Need help identifying a PSU.

Postby athlon-power » 2019-1-09 @ 20:50

gdjacobs wrote:It's hard to tell for sure without testing, but solid construction is always a good sign.

Self contained filter modules are a high quality, slick solution. Any OEM using one is doing that part of their job right.


The problem is, I don't have a multimeter. I can't test if the voltages coming from the PSU are what they're supposed to be, or not. I can't think of any other way to test it off of the top of my head. Everything looks okay in it, and I did attach two of the beefier IDE DVD drives that I have to it at the same time, and they worked just fine. I know that's a very dumb way to test it, but there seems to be very few options for me to actually try and test it.

At least we shouldn't have to worry about the input part of the PSU anymore.
Custom Builds:
Intel SE440BX-2, Intel Pentium III SL35E, 128MB PC100, ASUS V3800M 32MB, AOpen AW744L II, Samsung SC-148, WD Caviar WD400, ca. 1999

Upcoming builds:
Athlon XP 3200+ (Socket 754), 512MB DDR-400, rest unknown, ca. 2003
User avatar
athlon-power
Member
 
Posts: 150
Joined: 2018-9-29 @ 00:15

Re: Need help identifying a PSU.

Postby cyclone3d » 2019-1-09 @ 21:03

Found it.. The part number is definitely Dell so I started there and had no luck with the full part number... and then I remembered that the full part number is not actually the part number that Dell uses.

So I just used CN-03t938 to search.

It is a Mitac X-200/12V 200W power supply.

Here are some pics I shamelessly stole from this ebay listing:
https://www.ebay.com/itm/Mitac-X-200-12V-200W-Power-Supply-Dell-Dimensions-2300-CN-03T938-Tested-/292711872244

part_num_sticker.jpg

psu1.jpg

back.jpg

label.jpg
User avatar
cyclone3d
l33t
 
Posts: 2605
Joined: 2015-4-08 @ 06:06
Location: Huntsville, AL USA

Re: Need help identifying a PSU.

Postby athlon-power » 2019-1-09 @ 21:41

At least I now know the specs of this thing. This is probably a really dumb question, but is Mitac any good? I think I've heard that name before, and usually with the no-name brands, you'll see the label and have no idea about the company beforehand, period. A good example is from a PSU I got out of a "new" case I bought recently- It was a 'L&C' PSU. It weighed practically nothing and is basically garbage.

If I can figure out whether or not Mitac is a reputable brand, I may actually use this PSU. My 350W Power Man PSU is a little overkill for this build, and I could use the PCI-E power connector elsewhere, instead of letting it languish in a 1999 build.
Custom Builds:
Intel SE440BX-2, Intel Pentium III SL35E, 128MB PC100, ASUS V3800M 32MB, AOpen AW744L II, Samsung SC-148, WD Caviar WD400, ca. 1999

Upcoming builds:
Athlon XP 3200+ (Socket 754), 512MB DDR-400, rest unknown, ca. 2003
User avatar
athlon-power
Member
 
Posts: 150
Joined: 2018-9-29 @ 00:15

Re: Need help identifying a PSU.

Postby SW-SSG » 2019-1-09 @ 22:02

Mitac are BIG and actually owned Tyan for a while (although they were producing well-built motherboards even prior to the merger). Not sure about Mitac PSUs, though, but that one certainly looks high-quality...
User avatar
SW-SSG
Oldbie
 
Posts: 670
Joined: 2016-9-29 @ 03:44
Location: Canada

Re: Need help identifying a PSU.

Postby athlon-power » 2019-1-09 @ 22:12

SW-SSG wrote:Mitac are BIG and actually owned Tyan for a while (although they were producing well-built motherboards even prior to the merger). Not sure about Mitac PSUs, though, but that one certainly looks high-quality...


That does bode well for the PSU. I'm just nervous about what might happen if it messes up. I've had a PSU die in one of my old main computers, actually- I was playing a game and this really nasty ozone smell smacked me in the face. I turned the PC off, and sure enough, it wouldn't turn back on. I got a new PSU and it worked just fine again. The moral of the story is that I don't mind a PSU dying, but I do mind if it risks taking everything else with it, especially the motherboard. I cleaned it out, so any heat-related failures should be mitigated.

It also has a bit of coil whine when it's off, but still plugged in- I don't know if that signifies anything. I don't think it's a big deal, as it's not very loud in the first place, and even then a huge amount of old forum posts consistently mentioned that it was harmless. I figured it would be a good idea to mention here, though, just to be sure.
Custom Builds:
Intel SE440BX-2, Intel Pentium III SL35E, 128MB PC100, ASUS V3800M 32MB, AOpen AW744L II, Samsung SC-148, WD Caviar WD400, ca. 1999

Upcoming builds:
Athlon XP 3200+ (Socket 754), 512MB DDR-400, rest unknown, ca. 2003
User avatar
athlon-power
Member
 
Posts: 150
Joined: 2018-9-29 @ 00:15

Re: Need help identifying a PSU.

Postby SirNickity » 2019-1-10 @ 20:56

Coil whine is not a sign of anything really. Switching PSUs use various techniques to regulate a high input voltage to a low output voltage. Basically, they allow the high voltage to pass through in short bursts, and then through passive components that form a sort of low-pass filter, those short bursts of high voltage get averaged out into a steady but much lower voltage. When the PSU is off, there's very little load on it, so it takes less of an "on" burst to keep that steady output going. Some PSUs achieve this by keeping a regular interval of bursts, but scaling back how long the burst is -- the venerable pulse width modulation scheme, or PWM. Others keep the pulse WIDTH the same but pass that burst through less often -- modulating frequency instead of pulse width. In those cases, the frequency of pulses can start to dip down into the audible frequency range -- maybe 10 to 15kHz versus 50kHz, or 100kHz, or even up into the MHz (though not typically for this kind of PSU.) When it's ~10-15kHz, you'll hear the whine. Just like you do in a TV set where the yoke deflection coils are humming along at 15kHz raster line scanning frequency.

That said, sometimes that coil whine is indeed because of lower pulse frequency, but because the PSU has gone into current-limit or other protection mode, and so it's trying to start up, detecting a fault, and turning off again. Then it retries some milliseconds later. Obviously, if your PSU is working properly, this probably isn't the case. It's just a long way of saying "it depends."

The boxed film caps are the blue and white ones near the large coils in the side shot you took. Those are pretty reliable, so you should be OK for a while. The other caps look healthy as well. Impossible to say for sure, but if the PC seems stable, you can make a judgement call on whether to pursue cap replacement now, or only when/if you have trouble with it.

Unfortunately there's just no such thing as insurance against damage to downstream components. A PSU will ideally fail gracefully, and protect the downstream load at all costs. But sometimes you're just not that lucky. All you can do is try to use a quality PSU and hope for the best. There's just no getting around the risk entirely. This one seems like it's built pretty well, so I would not hesitate to use it.

A multimeter is nice, so you can see the output voltages, but you really should be loading the supply before you measure it anyway. An unloaded supply doesn't necessarily look safe to use. I have some quality Delta supplies that show wildly wrong 12V and 5V rails when run wide open. And they will only allow me to do that for a minute or so, then they go into protect mode. Attach a motherboard and they'll settle down and run like champs. A multimeter also won't tell you what the DC really looks like. It'll give you an average (maybe even RMS) voltage, but it won't necessarily tell you there's 4V of ripple on your 12V line. You need an oscilloscope to get the full picture.

That is to say, unless you have a sacrificial test jig of some sort, eventually you just have to hold your breath, plug it in, turn it on, and hope for the best. Maybe put in a cheap video card at first, and save your OG Sound Blaster 2.0 and Voodoo 5 for when you have more confidence in the supply.
SirNickity
Member
 
Posts: 327
Joined: 2018-11-08 @ 04:51

Re: Need help identifying a PSU.

Postby athlon-power » 2019-1-10 @ 22:33

SirNickity wrote:The boxed film caps are the blue and white ones near the large coils in the side shot you took. Those are pretty reliable, so you should be OK for a while. The other caps look healthy as well. Impossible to say for sure, but if the PC seems stable, you can make a judgement call on whether to pursue cap replacement now, or only when/if you have trouble with it.


The PC has been running just fine, and I've ran a few tests running the Pentium III Katmai 500MHz at full load. It only has a 28W TDP, so it's not the most vigorous of tests, but it ran fine like that.

SirNickity wrote:Unfortunately there's just no such thing as insurance against damage to downstream components. A PSU will ideally fail gracefully, and protect the downstream load at all costs. But sometimes you're just not that lucky. All you can do is try to use a quality PSU and hope for the best. There's just no getting around the risk entirely. This one seems like it's built pretty well, so I would not hesitate to use it.


So far, I've been using it, and it's been working fine, knock on wood. I'm sure cleaning it out helped quite a bit- the fan was originally very dusty, and so were most of the internals. I cleaned it as well as I could with my hodge-podge setup, and the air output from the back of the PSU is now nice and healthy.

SirNickity wrote:A multimeter is nice, so you can see the output voltages, but you really should be loading the supply before you measure it anyway. An unloaded supply doesn't necessarily look safe to use. I have some quality Delta supplies that show wildly wrong 12V and 5V rails when run wide open. And they will only allow me to do that for a minute or so, then they go into protect mode. Attach a motherboard and they'll settle down and run like champs. A multimeter also won't tell you what the DC really looks like. It'll give you an average (maybe even RMS) voltage, but it won't necessarily tell you there's 4V of ripple on your 12V line. You need an oscilloscope to get the full picture.


I installed an old version of SpeedFan to try and grab some CPU temperatures, and there is a real-time voltage meter you can look at. I really can't say how reliable it actually is, but looking at it, the voltages seem to be alright.

IMG_20190110_172201.jpg


I don't know how much variation is bad, but at the most the variation is under half a volt, and the 12v rails seem to be nearly spot-on. The only questionable rails here seem to be the 5V rail and the 3V rail. The wandering seems to be the worst on the 5V rail, by 0.32V. That seems a little high to me, but maybe it's within acceptable variation. Hopefully this can give at least some vague insight into the condition of the PSU.

SirNickity wrote:That is to say, unless you have a sacrificial test jig of some sort, eventually you just have to hold your breath, plug it in, turn it on, and hope for the best. Maybe put in a cheap video card at first, and save your OG Sound Blaster 2.0 and Voodoo 5 for when you have more confidence in the supply.


I don't actually have any hardware that's very rare- my budget is just fairly low, and so the more common stuff tends to be harder to come by. It can take long periods of time to get builds like this one together, which is why I don't want to lose anything- tracking down another non-OEM Slot 1 motherboard in my acceptable price range is going to be difficult.

I also found your explanation on coil while interesting- I always wondered why the older CRT displays whined like that. Almost every CRT TV I've ever owned has made that whining sound, but not so much my current vintage monitor. It works great, but there is very little audible sound coming from it at all. If it's of any interest to know, it's a Hyundai ImageQuest P910+.
Custom Builds:
Intel SE440BX-2, Intel Pentium III SL35E, 128MB PC100, ASUS V3800M 32MB, AOpen AW744L II, Samsung SC-148, WD Caviar WD400, ca. 1999

Upcoming builds:
Athlon XP 3200+ (Socket 754), 512MB DDR-400, rest unknown, ca. 2003
User avatar
athlon-power
Member
 
Posts: 150
Joined: 2018-9-29 @ 00:15

Re: Need help identifying a PSU.

Postby athlon-power » 2019-1-11 @ 21:33

I've been using the PSU for a little while now, and the coil whine I mentioned before is now much louder than it was before. I'm not sure what's causing it, and it's still audible only when the PSU is off and plugged in. As soon as I turn the computer on, it goes away and seemingly functions normally.

I'm not exactly sure what the louder coil whine signifies, at least with this particular PSU. It may just be aging, but I'm worried that it may indicate that this thing will soon fail. Apparently, this thing came out of a Dell PC, but it's labeled as a Mitac- it's either the same model or a very similar model to the one that cyclone3d posted pictures of.
Custom Builds:
Intel SE440BX-2, Intel Pentium III SL35E, 128MB PC100, ASUS V3800M 32MB, AOpen AW744L II, Samsung SC-148, WD Caviar WD400, ca. 1999

Upcoming builds:
Athlon XP 3200+ (Socket 754), 512MB DDR-400, rest unknown, ca. 2003
User avatar
athlon-power
Member
 
Posts: 150
Joined: 2018-9-29 @ 00:15

Re: Need help identifying a PSU.

Postby gdjacobs » 2019-1-12 @ 00:12

Could be the coil on the 5V SB output filter. In the spirit of nothing ventured, nothing gained, if you positively identify the coil making noise, you could try to goop the loops in place to each other using RTV silicone or epoxy and prevent them from vibrating.
User avatar
gdjacobs
l33t++
 
Posts: 6041
Joined: 2015-11-03 @ 05:51
Location: The Great White North


Return to General Old Hardware

Who is online

Users browsing this forum: Google [Bot], mixerjdp and 5 guests