KDMPower MIPC MI-X8775CD [PCB WF-C rev:E]

[sam_sam_sam]

Re: KDMPower MIPC MI-X8775CD: load-tested and still running… for now (part 2)

After the above tests, it was clearly time to step it up. Unfortunately, my current load test setup with all the jumper leads was not longer up to the task. ******************************
***********************************************************
I knew this PSU would be another “fun” repair project, but I didn’t know it’s going to kick off so many ideas for so many other projects / tests.

I was impressed with the way you went about torture testing your ATX switching power supply

But I have a couple of question for you, you point out the fact that at low input voltage that it should have locked out and not restart but is this with a load or with out ( first of all )

Second what is the correct way to test the low voltage lock out

What is the correct way to test this function

What if it does not work correctly what should you do to make it safer to use if your ATX switching power supply has this issue

Where did you get a heater element that is only 500 to 700 watts

[momaka]

Re: KDMPower MIPC MI-X8775CD: load-tested and still running… for now (part 2)

But I have a couple of question for you, you point out the fact that at low input voltage that it should have locked out and not restart but is this with a load or with out ( first of all )

It doesn't actually matter. The 5VSB may try to start, and that's OK, because it's designed to run all the time. But the main PS (3.3V, 5V, and 12V rails) should not try to keep running when the AC line voltage drops too far (typically 95-100V.) Thus, once the line drops too far, the main PS should shut down and stay off - and by this, I mean the PSU should at least stay off until the PS-ON signal is cycled again, if not until AC power cycled.

I also tested an Enermax ELT400AWT on that same setup (haven't posted the results in its thread yet, though) and that one would not try to run below 90-ish V AC, if I remember correctly. Once the AC input dropped below ~90V, the main PS shut off (5VSB kept running, though). I could get the PSU to try to start again by cycling the PS-ON signal, but the PSU would do the same thing (try to power on, see that it can't really run on that low of an AC line, and shut off and stay off), which is fine. The KDMPower PSU, on the other hand, just tries to keep running as long as the PS-ON signal was held low, and that's NOT OK.

Second what is the correct way to test the low voltage lock out

Well, if you have a Variac, that would probably be easier than my setup above: you just power ON the PSU with a load and then slowly lower the input AC voltage with the Variac until the PSU shuts off (or in the case with this cheapo PSU, probably blow up, if current on the input is not limited - which in my case it was, due to the series heating element.)

Since I don't have a Variac, the series heating element I used was just an "improvised" substitute... and it also helped limit current on the AC input, which is always helpful when testing something you've fixed and don't know how it will do. So essentially, the series heating element help me get "2 birds with 1 stone". The downside is you have to do a few calculations prior to rigging the setup to get a rough idea of what it can and can't handle.

What if it does not work correctly what should you do to make it safer to use if your ATX switching power supply has this issue

That, I haven't figured out yet.

But generally, if the PSU detects a fault (be it a short-circuit on the output, an abnormally-low AC input, or over-voltage on one of the outputs, and etc.), it should shut off and stay off until the PS-ON signal is released (allowed to go high again) at the very least. On most PSUs that use STF designs with current-mode PWM controllers (like UC384x or CM6800/CM6802), the PSU will usually even latch down in the OFF state until the AC power is cycled.

Where did you get a heater element that is only 500 to 700 watts

The 500W one, if I'm not mistaken, was the heating element you see on the bottom of dishwashers. I'm not 100% sure if that's where it came from, though, because I got it from a dumpster many years ago.

The 700/1400 Watt heating element is actually a mini toaster oven that I have in my kitchen. It's just an older type with mechanical switches (i.e. no "smart" electronics that I have to bypass to get it to turn on in the above experiments.) It has 2x 350 Watt heating elements on the top in parallel (i.e. 700W total) for the "broil" function. The bottom also has 2x 350 Watt heating elements in parallel (another 700 Watts), but these come ON with the top heating elements only when the "Bake" function is used (thus, getting all four 350 Watt heating elements in parallel for a total of 1400 Watts.) So depending on whether I have the knob on "bake" or "broil", I can switch between 1400 and 700 Watts on-the-fly.

That being said, you can use just about any appliance that has a heating element in it - toasters, sandwich presses, rice cookers, coffee makers... and many others. Just make sure the appliance uses an actual Nichrome -based heating element and not some kind of an induction coil, as I'm not sure how that will work out. Also, better to make sure the appliance is the "dumb" type - i.e. no "smart" electronics, LCD displays, or "soft" buttons, as those probably won't let you turn ON the device.

Better yet, just grab a bunch of oldschool "curled" stovetop/burner heating elements. They can be found for about $10-15 usually (though some years back, Amazon was selling a few at $4 a pop with free S&H.). Most are typically rated 1-3 kW @ 208/230/240V.

Just remember that when running an resistive Nichrome heating element at half of its rated voltage, its output power will drop to 1/4 of its rating. So for example, a 1.2 kW (1200 Watt) heating element rated for 240V AC will output only 300 Watts at 120V AC (i.e. 1200 / 4 = 300W ).

have you ever cut and stripped one of the centech probe leads??
i doubt it - or you wouldnt have put 111v through one!

Actually, I have - not on my own meters, though (I have 3 of these.) Just seen a few in the past with the wire breaking off from the probe handle (very common problem with cheap multimeters.) The wire inside is indeed really laughably thin. The insulation, however, isn't that bad - probably comparable to that of standard-rated 300V wire... though that's not so good, considering many of these cheap meters have dials indicating they can measure up to 1000V DC or AC, which is downright DANGEROUS given their construction. That said, my 3rd Cen-Tech meter is a newer, "revised" version - dial sticker was changed to indicate that the meter can measure only up to 250V (AC or DC) and the unfused current jack downgraded to 5 Amps. But most importantly, the wires on the test probes that came with this 3rd meter are considerably thicker than on the older two versions I have. I've tested the old wire and the new one. The old one will get very warm with just 2-3 Amps of continuous current. And at 5 Amps it's borderline not melting (soft / "flaccid" would be a good way to describe it - which is funny, considering the wire one these meters is normally quite stiff when cold. ) IIRC, I think it was someone on here who measured how much current these cheap wires can handle, and the result was (again, IIRC) that at 6-7 Amps, those old wires will melt. The wire on my 3rd (newer) meter, also gets warm at 5 Amps, but not nearly as much as the old one. HF also added a bunch of safety-related notes on the label and in the manual regarding max. voltage and current, among other things. My guess is they probably got bit by complains / small claims / lawsuits against them before, so the newer revision of the meter covers their asses just a tad better and comes with slightly safer leads... though they will still fall apart at the handle unless glued - which I did to all 3 of my meters as soon as I got them, and their probes are still 100% intact and in good condition after many years of use.

In any case, I'm still well aware of the shortcomings of these cheap meters. I actually posted a thread here on the first one I got a while back. I don't use them anywhere that involves unlimited/unchecked current and high voltage. In the test setups I've showed above, both voltage and current were within fairly safe margins of what these meters can handle, so it's not a big issue. And most importantly, I set everything up for "hands-free" operation, so I'm not touching the probes or changing the settings on anything that's live. That's actually the reason why I got so many of these cheap meters - so I can put one on each spot I want to measure and not have to diddle with moving probes around.

[sam_sam_sam]

Re: KDMPower MIPC MI-X8775CD: load-tested and still running… for now (part 2)

Well, if you have a Variac, that would probably be easier than my setup above: you just power ON the PSU with a load and then slowly lower the input AC voltage with the Variac until the PSU shuts off (or in the case with this cheapo PSU, probably blow up, if current on the input is not limited - which in my case it was, due to the series heating element.)

Do you need something to limit the current with this setup ( I do have a Variac unit I have a 5 amp and a 15 amp ) both of have been modified with a volt meter and current meter

Since I don't have a Variac, the series heating element I used was just an "improvised" substitute... and it also helped limit current on the AC input, which is always helpful when testing something you've fixed and don't know how it will do. So essentially, the series heating element help me get "2 birds with 1 stone". The downside is you have to do a few calculations prior to rigging the setup to get a rough idea of what it can and can't

This might answer the first question but how do you “calculations” are you talking about the limited function of the power supply because of current limiting device that is in place

The 500W one, if I'm not mistaken, was the heating element you see on the bottom of dishwashers. I'm not 100% sure if that's where it came from, though, because I got it from a dumpster many years ago.

The 700/1400 Watt heating element is actually a mini toaster oven that I have in my kitchen. It's just an older type with mechanical switches (i.e. no "smart" electronics that I have to bypass to get it to turn on in the above experiments.) It has 2x 350 Watt heating elements on the top in parallel (i.e. 700W total) for the "broil" function. The bottom also has 2x 350 Watt heating elements in parallel (another 700 Watts), but these come ON with the top heating elements only when the "Bake" function is used (thus, getting all four 350 Watt heating elements in parallel for a total of 1400 Watts.) So depending on whether I have the knob on "bake" or "broil", I can switch between 1400 and 700 Watts on-the-fly.

Would you recommend making a “ test jig ” that basically has a outlet that the heat element is in series with the main power coming and outlet to the device that is being tested

That being said, you can use just about any appliance that has a heating element in it - toasters, sandwich presses, rice cookers, coffee makers... and many others. Just make sure the appliance uses an actual Nichrome -based heating element and not some kind of an induction coil, as I'm not sure how that will work out. Also, better to make sure the appliance is the "dumb" type - i.e. no "smart" electronics, LCD displays, or "soft" buttons, as those probably won't let you turn ON the device.

Better yet, just grab a bunch of oldschool "curled" stovetop/burner heating elements. They can be found for about $10-15 usually (though some years back, Amazon was selling a few at $4 a pop with free S&H.). Most are typically rated 1-3 kW @ 208/230/240V.

Just remember that when running an resistive Nichrome heating element at half of its rated voltage, its output power will drop to 1/4 of its rating. So for example, a 1.2 kW (1200 Watt) heating element rated for 240V AC will output only 300 Watts at 120V AC (i.e. 1200 / 4 = 300W ).

Would you need a cooling fan or water to cooling the heating element

Basically I want to make something that I could use with ease and not have to set something up each time I want to troubleshoot some switching power supply issue or issues

Of all of the heating elements that you were talking about earlier which would you use and why

[momaka]

Re: KDMPower MIPC MI-X8775CD: load-tested and still running… for now (part 2)

Do you need something to limit the current with this setup ( I do have a Variac unit I have a 5 amp and a 15 amp ) both of have been modified with a volt meter and current meter

If the PSU decides to blow up, you might want to implement something to limit the current, indeed. Otherwise if you don't care, then no.

This might answer the first question but how do you “calculations” are you talking about the limited function of the power supply because of current limiting device that is in place

I'll probably need to make an Excel spreadsheet for that, as it might be easier to understand.

In general, however, again you just have to know that for an "ideal" maximum power draw, the line AC voltage should be approximately split in half between the series heating element and the PSU... thus each consuming up to 1/4 of the rated power of the heating element. Of course, as I noted above, most PSU's simply won't work at half the AC line voltage, and typically the minimum is 90-100 V for 120V-rated devices and 190-200V for 220/230/240V -rated devices. Therefore, the load you put on the PSU should be much smaller so that the input voltage won't drop as much. IME, taking 1/2 to 1/3 of the 1/4-power rating of the heating element (i.e. 1/8 to 1/9 total) is probably *around* the maximum load you can put on the PSU. So with a 500 Watt heating element for example, the 1/4 rating is 500 / 4 = 125 Watts. Taking 1/2 of that is ~60-ish Watts... or ~42 Watts with 1/3. So with a 40-60 Watt load on the PSU output, most likely the voltage on the PSU input would be just high enough for the PSU to stay working. And then, to induce a low input AC line, just continue increasing the load (power draw) on the PSU output (i.e. connect more fans, light bulbs, HDDs, or whatever else.) The more you load the PSU output, the more power (and thus current) the PSU will draw on the input / AC side, which means more voltage dropped across the series heating element. As you continue increasing the load on the PSU, the line input should eventually reach an level where the PSU shuts off and stop working (or not, like the PSU in this thread.)

Would you recommend making a “ test jig ” that basically has a outlet that the heat element is in series with the main power coming and outlet to the device that is being tested

You can, if you're going to use it often.
On that note, I actually did make one many years ago... but for some reason just never got to using it before (probably because I never added a proper input plug on it) and completely forgot about it until now. So thanks for reminding me. Maybe I will put it to use. Here's what I have:
https://www.badcaps.net/forum/showpo...&postcount=183

Would you need a cooling fan or water to cooling the heating element

No, if the heating element is made to run at full power without active cooling.

In the case of my toaster oven... well, it's a toaster oven, and those heating elements inside are meant to get red-hot under max load (which you will never see in a series configuration in normal operation - unless your PSU short-circuits and you don't notice and leave the setup running like that for a while.)

In the case with the 500W dishwasher heating element... I'm not sure if it's made to run at max power without being in water, so I can't say. I think it is, as I think that's how dishwashers "dry" the dishes on the dry cycle. But because that heating element would run hot left air-coupled, I would need to put it on something that won't burn. A brick or two would do the job. But I chose the pot of water, as that's cleaner than bringing dirty bricks from outside (which I don't even have ATM.) Also, I originally started using that heater in that pot of water back in late December when we had some really cold temperatures and the air got really dry from the house heating. Thus closing my work area door (and air vents in the room), then running the heating element at half power (via. single diode half-wave rectification) kept the water in the pot right around 50-60C mark - perfect where it's releasing a lot of moisture in the air but not boiling. In short, it was my ghetto air humidifier.

But no, you don't need to cool your heating elements if they are designed to take full power without need for active cooling. Most cooking oven elements are of this type.

Of all of the heating elements that you were talking about earlier which would you use and why

Any toaster ovens and toasters come first, as the heating elements in those can run in "free air" without active cooling. Then curled stovetop/burner heating elements, as those *can* usually take full power while being in "free air"... though they are really designed to have a "load" (i.e. in contact with a pan or pot) on top of them. Other than these considerations and the power rating, there really is no difference which one you choose/use. I used the ones shown here because that's just what I have on hand. (Well, I have a couple more parts boxes in the garage that I haven't looked through what's in there... though I do recall I have a few more different heating elements around. )

[sam_sam_sam]

Re: KDMPower MIPC MI-X8775CD [PCB WF-C rev:E]

Thanks for the information

[goodpsusearch]

Re: KDMPower MIPC MI-X8775CD: load-tested and still running… for now (part 2)

(I guess goodpsusearch was right in suggesting to leave the extra rectifier on the 12V rail… though I still think the PSU would do better with one bigger rectifier on the 12V rail vs. two smaller ones.)

I agree! It will also highly depend on the Vf (forward voltage drop).

[Th3_uN1Qu3]

Re: KDMPower MIPC MI-X8775CD [PCB WF-C rev:E]

Heck, does a half-bridge design over 500 Watts even exist as far as ATX PSUs go? If yes, I don’t want to know how much one of those weights or what kind of BJTs it has on the primary.

Yes. There's a 450W Spire with Passive PFC. It does do rated power. It's HEAVY.

It uses TO-3P D209Ls on the primary, the PI filters are complete, input caps either a pair of 680uF 250v or 1000uF 250v, output toroid large. Transformer is a tad bigger than the one in yours iirc, and I made it into a 600W +/-30V and +/-12v supply for my amplifier. Just the PCB was left of it for that purpose though.

[Dan81]

Re: KDMPower MIPC MI-X8775CD [PCB WF-C rev:E]

Who was the OEM on the Spire? I remember I have a Sun Pro AT-2005B based one (which someone recased into a RPC shell, of all things, lol - I think it was a 380W unit, and *should* do its rated power I guess) and two YX-GP labeled ones that claim 420W - they're the shiny golden SP-ATX-420WTN-PFC models.

[Th3_uN1Qu3]

Re: KDMPower MIPC MI-X8775CD [PCB WF-C rev:E]

Who was the OEM on the Spire? I remember I have a Sun Pro AT-2005B based one

Sun Pro and it had the AT2005B also.