Announcement

Collapse
No announcement yet.

CompUSA PowMax LP-6100 – this is as good as they get!

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    CompUSA PowMax LP-6100 – this is as good as they get!

    Ah, good ol' PowMax! I mean, who doesn't love them, right?
    Even Topcat himself used to "frequently recommend and use Powmax supplies"
    Don't believe me? Have a look here.
    (P.S. Sorry TC, I just had to make a funny here. Hope you don't mind. )

    Anyways, folks.
    What I will be showing you today is a CompUSA 500 Watt power supply based on the Leadman/PowMax LP-6100 platform. For those of you who are still scratching your head and wondering about Leadman/PowMax – it's basically a company that used to (and still does?) make some of the worst power supplies ever. Therefore, their PowMax is often called POW to the MAX here… and for a good reason . I think you might have gotten a hint of that even in the above link.

    That said, the LP-6100 platform/design is probably the closest they ever came to almost making a half-decent PSU, in my opinion. (Note my careful wording in that last sentence. ) I say that, because the guts of the LP-6100 visually don't look all that terrible. Yet, there are plenty of stories of PowMax PSUs blowing up, including this model.

    Now let's see some pictures of my model. As usual, I always start with the case/shell and label.
    https://www.badcaps.net/forum/attach...1&d=1451693563
    https://www.badcaps.net/forum/attach...1&d=1451693563
    As you can see, this CompUSA "500 Watt" PSU is a dual-fan design: one pushing air in and one pulling air out of the PSU, both wired in parallel to the same temperature controller. Both fans are made by Yate Loon (FM802512H and FM802012H, rated for 0.18 and 0.19 Amps, respectively.) Now this may sound surprising, but the case metal is actually of proper thickness and doesn't feel flimsy at all. In fact, it feels very sturdy. Even with the guts removed, the case still feels heavy. You can also see from the pictures that the output wires are quite long and not so thin like on other cheap PSUs. In fact, they are all rated for 300V, 18 AWG and have legit UL numbers. In terms of connectors, this one has a 20-pin ATX, 4-pin CPU, 6-pin AUX (AT style), six standard drive connectors, and two Berg floppy drive connectors. So, on the outside, this does actually look like a decent PSU. And it has some weight to it, too.

    Moving onto the label:
    https://www.badcaps.net/forum/attach...1&d=1451693563
    We see 40 Amps on the 5 V rail and 30 Amps on the 3.3 V rail?! That's 1 Watt short of 300 for the 5 V and 3.3 V rails combined. A bit overrated? I think so! After all, how else are the numbers on the label going to add up to the promised rating of 500 Watts?
    Another interesting bit is Leadman/PowMax's UL number (E107407). Nothing comes up in the UL database if you look it up. I guess it could be expired. Or it could be forged. Either way, don't let that comfort you that this is a perfectly safe PSU, because it probably isn't.

    And now, time for the guts:
    https://www.badcaps.net/forum/attach...1&d=1451693563
    Clearly, you can see why this PSU has weight to it. There are good-sized heatsinks and full input filtering. Also note the proper-spec primary wires (18 AWG, 600V). The small board on the secondary heatsink is the fan controller. Again, everything looks decent so far.

    Well, let's see the solder side:
    https://www.badcaps.net/forum/attach...1&d=1451693563
    I'm not impressed here. :\ The soldering is not terrible, but still sloppy in a few places. I didn't measure the creep distance between primary and secondary, but I think it is 5 mm for the most part. The only thing I am not sure about is the solder: it seems too dull to be Pb (Leaded) solder but too shiny for Pb-free.

    Finally, we take a look at the secondary side:
    https://www.badcaps.net/forum/attach...1&d=1451693563
    https://www.badcaps.net/forum/attach...1&d=1451693563
    Now you see why I stopped using this PSU! – Bulged Jun Fu caps at their finest. They lasted maybe 3 years of light-moderate use with a low-power 5 V-based PC (AMD Duron 1400 Applebred CPU and Radeon 9200 SE) – possibly even less than 3000 hours, based on the counter in my Maxtor 32049H2 hard drive. Unacceptable!
    Those bad caps aside, the secondary side doesn't look that bad either. Each major rail has two caps and space for a PI coil. Only the 5 VSB and 3.3 V rails are missing their PI coils. Also note the huge output toroid inductors – the size of the 3.3 V rail toroid is actually what you would find as the *main* output toroid inductor in cheap PSUs. And the main output toroid inductor on this PSU is huge. It has very thick wires, too: three for the 5V rail, two for the 12V rail, and one for the -12V rail. Good job, PowMax! At least you didn't cut corners there.

    But this is a PowMax PSU after all, so there has to be something wrong with it.
    https://www.badcaps.net/forum/attach...1&d=1451693563
    It may not seem like anything, until you look carefully. At the lower left side of this picture, there are two jumpers. Between them is connected the leg of ceramic cap C17. The reason I pick on this is because the two jumpers are directly connected to the 5 V rail, whereas the leg of C17 that goes between them is connected to ground underneath. If the ceramic insulation breaks away (and it easily can from a physical disturbance), there will be a hard short-circuit from the 5 V rail to ground. And if that happens, let's hope this PSU has proper short-circuit protection. Otherwise, it will likely live up to it's name

    The primary-side RCD snubber network (seen to the right of the transformer) is another point of concern, in my opinion. The big diode (FR307) you see is actually not part of this RCD snubber network. Instead, it is a snubber for an additional winding on the primary side of the transformer, which is something I have never seen in STF topologies from good manufacturers, like HiPro, Delta, and Lite-ON. Why PowMax has designed it this way rather than a more standard method is beyond my understanding. And given that I have seen pictures of blown primary sides of this PSU before, I wouldn't be surprised if marginal snubber design has something to do with it. As for the actual RCD snubber network, there is a second (small) diode underneath the big diode. I believe it is an FR107 (1 A, 1000 V). It is connected in series with a 22 KOhm 2 (or 3) Watt resistor parallel with a 100 nF 400V metalized polypropylene cap. To me, that 22-KOhm resistor seems a bit too high to keep the inductive kickback of the primary in check (though, I will admit I haven't measured it). The primary MOSFET is rated for 600V max only.

    Finally, we have a look at the 5 VSB section:
    https://www.badcaps.net/forum/attach...1&d=1451693563
    Again, we see more disappointments here. First off, this *is* a 2-transistor design *with* a "critical" cap (25 V, 47 uF, spot C32). The 5 VSB transformer is very tiny. According to everell, that transformer runs hot even with just 5 Watts load (1 Amp at 5 Volts) after a few minutes. Another thing to note is that the 5 VSB primary side drive transistor (a 2SK3067 MOSFET) is not heatsinked at all! The PCB area underneath its pins is slightly darkened in my unit, so I definitely think it needs one. The snubber RCD network is nearly the same as for the main PS, except the snubber cap is only a 10 nF ceramic rather than 100 nF polypropylene. So I am not sure how good that is. But I suppose it's good enough for the 5 VSB rail.

    Here is a schematic of the 5 VSB circuit that I drew:
    https://www.badcaps.net/forum/attach...1&d=1451693563
    The last thing to mention about the 5 VSB circuit is that the primary side auxiliary winding is also used for providing power to the PWM controller. The secondary side of the transformer has only one winding – the 5 Volt output. It is used both for the 5 VSB and powering the supervisory IC. The PCB has spots for components so that the supervisory IC can have its own filtered output. However, PowMax went cheap here and connected the supervisory IC Vcc directly to the 5 VSB output through a diode rather than installing all of those components.

    Anyways, that is all for the 5 VSB section. I haven't ran any load tests on it like I did with some of my other PSUs (since the primary caps are currently removed).
    Attached Files

    #2
    CompUSA PowMax LP-6100 – this is as good as they get!

    Summary

    Primary Side:

    * 6.3 Amp glass fuse + SDK 053 NTC thermistor
    * full input filtering: five Y2 safety-rated caps, two X2 rated caps (0.22 uF and 0.047 uF – both Teapo XG-VS), one common-mode choke and two single-mode chokes
    * two 6 Amp diodes and two 1N5406 (3 Amp) diodes for the bridge rectifier
    * two JEE IM series, 200 V, 1000 uF, 25 x 46 mm (dia. x h) bulk caps
    * one W8NA60 MOSFET (TO-247 package) for the main PS, STF topology
    * one 2SK3067 MOSFET (TO-220FP) for 5 VSB, unheatsinked, 2-transistor design
    * critical 5 VSB cap: JEE IM series, 25V, 47 uF, 5 x 11 mm, electrolytic
    * main transformer with "35" size EE/EI core
    * ATC AP3842 PWM controller for main PS

    Secondary Side:
    ST1412A Supervisory IC (I can't find any info on it, but I am guessing it is similar to TPS3510.)

    3.3 V rail (mag-amp regulated):
    * SBL2040CT schottky rectifier (20 A, 40 V, TO-220AB)
    * 2x Jun Fu WG series, 10 V, 2200 uF, 10x25 mm caps with missing PI coil between them
    * 47-Ohm, ½ Watt load resistor

    5 V rail:
    * two SBL3040PT schottky rectifiers in parallel (30 A, 40 V, TO-247)
    * 2x Jun Fu WG series, 10 V, 2200 uF, 10x25 mm caps with a PI coil (6 turns, 5 mm core) between them
    * 100-Ohm, ½ Watt load resistor

    12 V rail:
    * FEP16BT fast recovery rectifier (16 A, 100 V, TO-220)
    * 2x Fuhjyyu TNR series, 16 V, 1000 uF, 10x20 mm caps with a PI coil (19 turns, 6 mm core) between them.
    * 270-Ohm, 1 Watt load resistor

    -12 V rail:
    * HER302 fast recovery diode (3 A, 100 V)
    * 1x Fuhjyyu TNR series, 16 V, 470 uF, 8x14 cap before 4.7 uH PI coil
    * 1x JEE IM series, 16 V, 100 uF, 5x11 cap after 4.7 uH PI coil
    * 270-Ohm, 1 Watt load resistor.

    -5 V rail:
    * derived from -12 V rail through 7905 linear regulator (TO-220AB, no heatsink)
    * 1x Fuhjyyu TNR series, 16V, 47 uF, 5x11 mm cap after 7905 regulator

    5 VSB rail:
    * 1N5822 schottky diode (3 A, 40 V)
    * 1x Jun Fu WG series, 10 V, 1000 uF, 10x12 cap before missing PI coil
    * 1x JEE IM series, 10 V, 100 uF, 5x11 cap after missing PI coil
    * 51-Ohm, 1 Watt load resistor

    Overall Thoughts
    It's not too bad of a PSU – at least it doesn't look like it on the outside and judging by all of the part. Of course, it's a PowMax, so one never knows.

    I have been saving it with the intention to fix it and use it. Of course, now that I found some potential flaws (and a pun here ), I am probably going to wait to build a load tester first. I simply can't trust this thing to run in a computer quite yet. Regardless, I will save it - maybe even just for the nostalgia. After all, this was the PSU that powered my first "gaming" PC that I built. And it didn't go *POW* on me either.

    Comment


      #3
      Re: CompUSA PowMax LP-6100 – this is as good as they get!

      Linkworld likes its 5vsb mosfets/transistors without heatsinks too, fortunately not too often.

      Comment


        #4
        Re: CompUSA PowMax LP-6100 – this is as good as they get!

        oh no. this is a badcap double whammy here. fuh-joo-yoos with jun fuck. this deserves to go in the badcap hall of shame. i'm looking forward for more bulging badcap pr0n. u know, when its viewed from the side, the bulge looks like a tiny mammary gland...

        Comment


          #5
          Re: CompUSA PowMax LP-6100 – this is as good as they get!

          Hi momaka. You have inspired me to drag out my 250 watt CompUSA power supply. The 5vsb transformer ran hot, so I replaced it with a larger transformer from a junker. Also removed old 5vsb components and added a ViPER 22 circuit outboarded. Had to replace the diode and capacitor in the snubber to reduce singing. Removed the R6 resistor which sets the 5vsb voltage regulation point and installed a 5K trim pot. Set it for 5.10 volts. Also added an indicator LED circuit. I took out the FET regulator in the 3.3 volt rail and installed a mag amp circuit. It is regulating at 3.90 volts - close enough. I also removed resistor R59 which sets the regulation voltage for the 5 volt and 12 volt rails, and installed a 5k trim pot for it also. Set the voltage on a working mother board at 12.14 volts and 5.12 volts. While working on it, I found two more Rulycon capacitors with high esr. They got replaced. Also replaced the rectifier in the -12 volt circuit. It also was having stability issues. Fixing this thing has been somewhat like fixing an IMicro power supply. End result is......it does work!
          Attached Files
          Old proverb say.........If you shoot at nothing, you will hit nothing (George Henry 10-14-11)

          Comment


            #6
            Re: CompUSA PowMax LP-6100 – this is as good as they get!

            ^ Nice mods, everell . Is that the one that blew the main MOSFET and 3842 PWM controller? Looks like your board has a lot more "burn-in" than mine.

            I don't know if I will do that many mods to mine. Might do the 5VSB with a chip, if it seems to run hot after I heatsink the MOSFET and maybe also tweak the primary-side snubber network of the main supply. Probably need to fix up some jumper messiness on the secondary, though, and check if I can pull at least 200 Watts from the PSU without it blowing up. But I definitely won't be touching the output voltages and adding pots. If I remember correctly, which was many many years ago since I ran it in a computer, that PSU had a pretty solid 5V rail, so I might use it in a 5V-based PC again.

            Comment


              #7
              Re: CompUSA PowMax LP-6100 – this is as good as they get!

              This CompUSA was the second power supply I worked on (the first being a Bestec 12E mobo killer...which still works fine!). This CompUSA had ALL Rulycon capacitors most of which were bulging. I am sure that every capacitor in this thing has been replaced at least once. The switching FET and 3842 shorted at least twice, and the output diodes shorted several times. Lots of burnt spots on the bottom. Truely a learning experience. Now you can understand why the IMicros and CoolMax were a breeze to work on. Bestec even easier. I still don't understand how that CompUSA main transformer primary works! Converting the 3.3 volt FET circuit to a mag amp circuit was another wild experience.
              Old proverb say.........If you shoot at nothing, you will hit nothing (George Henry 10-14-11)

              Comment


                #8
                Re: CompUSA PowMax LP-6100 – this is as good as they get!

                Originally posted by everell View Post
                I still don't understand how that CompUSA main transformer primary works!
                Frankly, I don't think Leadman/PowMax does either. Looks like they bodged something together that happened to work by accident .

                Comment


                  #9
                  Re: CompUSA PowMax LP-6100 – this is as good as they get!

                  Hah, I think that's one the few times I've seen different diodes used in one bridge rectifier. (On the AC input)
                  Funny, I really wonder why they did that XD

                  -Ben
                  Muh-soggy-knee

                  Comment


                    #10
                    Re: CompUSA PowMax LP-6100 – this is as good as they get!

                    The dark spots on the PCB remind me of Deer!
                    ASRock B550 PG Velocita

                    Ryzen 9 "Vermeer" 5900X

                    16 GB AData XPG Spectrix D41

                    Sapphire Nitro+ Radeon RX 6750 XT

                    eVGA Supernova G3 750W

                    Western Digital Black SN850 1TB NVMe SSD

                    Alienware AW3423DWF OLED




                    "¡Me encanta "Me Encanta o Enlistarlo con Hilary Farr!" -Mí mismo

                    "There's nothing more unattractive than a chick smoking a cigarette" -Topcat

                    "Today's lesson in pissivity comes in the form of a ziplock baggie full of GPU extension brackets & hardware that for the last ~3 years have been on my bench, always in my way, getting moved around constantly....and yesterday I found myself in need of them....and the bastards are now nowhere to be found! Motherfracker!!" -Topcat

                    "did I see a chair fly? I think I did! Time for popcorn!" -ratdude747

                    Comment


                      #11
                      Re: CompUSA PowMax LP-6100 – this is as good as they get!

                      Originally posted by ben7 View Post
                      Hah, I think that's one the few times I've seen different diodes used in one bridge rectifier. (On the AC input)
                      Funny, I really wonder why they did that XD
                      Sun Pro did it in a lot of their PSUs as well. And given the similar designs of some units, at one point I even thought Sun Pro = PowMax. But whether that is really true or not, I don't know.

                      As for why they do the different diode sizes:
                      When the voltage selector switch is set for 230 V AC, all four diodes in the bridge rectifier are used, and you get about 340 V DC total across the two bulk caps after rectification.

                      However, with the voltage selector switch set for 115 V AC, the two caps on primary are wired as a voltage doubler circuit. When this is the case, only two of the diodes in the bridge rectifier are used (usually on the Live rail). The other two are bypassed by the voltage selector switch.

                      Now, it takes twice the current at 115 V AC to get the same amount of power when on 230 V AC. But when on 115 V AC, only two diodes are used. So what did PowMax/Leadman and some Sun Pro units do? - Only put the bigger diodes on the rail that gets used both on 115 V AC and 230 V AC. In other words, the diodes that get bypassed by the voltage selector switch on 115 V AC can be kept the same small size as when on 230 V AC.

                      I can't imagine how that saves on production costs, though. Sure two small diodes are cheaper than two bigger diodes... but if you buy all of the same diodes in bulk, I think it might be cheaper to go with one size. Moreover, you have to spend time and train the "workers" (i.e. little children) where to place the large diodes and where to place the small ones - both of which takes more time IMO. So overall, I think it is a silly thing to do this with the diodes.

                      But this is a PowMax, after all. What more (or less) can you expect?

                      Comment


                        #12
                        Re: CompUSA PowMax LP-6100 – this is as good as they get!

                        Originally posted by momaka View Post
                        However, with the voltage selector switch set for 115 V AC, the two caps on primary are wired as a voltage doubler circuit. When this is the case, only two of the diodes in the bridge rectifier are used (usually on the Live rail). The other two are bypassed by the voltage selector switch.
                        What about in 120V only PSUs, wide range flyback PSUs, or in APFC PSUs (IE power supplies that don't use the voltage doubler or power supplies with no voltage selector switch)? On an 115V input, I mean. What I gather from this is, at least, is that a 6A bridge rectifier in a voltage doubler circuit, when the switch is set to 115V, is really a 3A bridge (two 1.5A diodes = 3A, four = 6A, obviously ).
                        Last edited by Wester547; 02-09-2016, 12:21 PM.

                        Comment


                          #13
                          Re: CompUSA PowMax LP-6100 – this is as good as they get!

                          Originally posted by Wester547 View Post
                          What about in 120V only PSUs, wide range flyback PSUs, or in APFC PSUs? On an 115V input, I mean. What I gather from this is, at least, is that a 6A bridge rectifier in a voltage doubler circuit, when the switch is set to 115V, is really a 3A bridge (two 1.5A diodes = 3A, four = 6A, obviously ).
                          Hmm, interesting point

                          -Ben
                          Muh-soggy-knee

                          Comment


                            #14
                            Re: CompUSA PowMax LP-6100 – this is as good as they get!

                            Originally posted by Wester547 View Post
                            What I gather from this is, at least, is that a 6A bridge rectifier in a voltage doubler circuit, when the switch is set to 115V, is really a 3A bridge (two 1.5A diodes = 3A, four = 6A, obviously ).
                            I'm not quite sure what you mean with this.

                            It is rather simple. Let's say, for simplicity, your PSU is drawing 240 Watts from the wall. At 240 V AC, it would pull 1 Amp of current ( P = V x I ). But at 120 V AC, it would draw 2 Amps. So the bridge rectifier diodes for the "120V" version need to be rated twice as high.

                            Now this only applies for PSUs with a voltage selector switch and two bulk caps on the primary:
                            - When the switch is set for 240V operation, all four diodes in the bridge rectifier are used.
                            - When the switch is set for 120V operation, only two out of the four diodes in the bridge rectifier are used. The other two are bypassed and not doing anything.

                            Because of that, PowMax/Leadman only puts the larger diodes in those spots that are used both by 120V and 240V operation.

                            Originally posted by Wester547 View Post
                            What about in 120V only PSUs, wide range flyback PSUs, or in APFC PSUs (IE power supplies that don't use the voltage doubler or power supplies with no voltage selector switch)? On an 115V input, I mean.
                            Those are different animals.

                            Comment


                              #15
                              Re: CompUSA PowMax LP-6100 – this is as good as they get!

                              Originally posted by momaka View Post
                              I'm not quite sure what you mean with this.

                              It is rather simple. Let's say, for simplicity, your PSU is drawing 240 Watts from the wall. At 240 V AC, it would pull 1 Amp of current ( P = V x I ). But at 120 V AC, it would draw 2 Amps. So the bridge rectifier diodes for the "120V" version need to be rated twice as high.

                              Now this only applies for PSUs with a voltage selector switch and two bulk caps on the primary:
                              - When the switch is set for 240V operation, all four diodes in the bridge rectifier are used.
                              - When the switch is set for 120V operation, only two out of the four diodes in the bridge rectifier are used. The other two are bypassed and not doing anything.

                              Because of that, PowMax/Leadman only puts the larger diodes in those spots that are used both by 120V and 240V operation.


                              Those are different animals.
                              Thanks for the explanation. But I guess what I'm asking is, in power supplies that don't employ a voltage doubler circuit (and thus no voltage selector switch), are all four diodes used in the bridge rectifier on an 115V input or less than that (or does it depend on the current draw)?

                              Comment


                                #16
                                Re: CompUSA PowMax LP-6100 – this is as good as they get!

                                Anything that doesn't have a voltage doubler circuit will use all four diodes in the bridge rectifier. The only exceptions may be perhaps some odd-ball small power adapters that rectify half the mains... but those are extremely rare. Usually even the small power adapters have a BD or four separate diodes.

                                Comment

                                Working...
                                X