Announcement

Collapse
No announcement yet.

Modding Two PSUs to Output 12V

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

    Modding Two PSUs to Output 12V

    Hello again, this might be a bit of a far-fetched question, but is it possible to modify a 36v and a 48v power supplies to output 12v? Both power supplies appear to use the same topology, they both also have a dedicated 12v winding wired to the control chips, meaning that as soon as i change it so the main output is 12v, then the dedicated windings will not be able to output 12v, meaning it will not start up. However, if i were to then take the main output and tie it to where the dedicated 12v output would go, then it should be able to run fine?

    Anyway here is a quick overview of both PSUs, the first PSU is a Genssi model S-350-36 rated 36V, 9.7A.

    The main switching transistors are two J13009
    The primary capacitors are two real-looking NCC CE 680uf, 230v
    The control chip is a TL494CN with a LM358 op-amp
    The secondary rectifier is a ESAD92-02S
    The (main) secondary capacitors are two Samyong NHA 2200uf, 50v
    The 12v secondary capacitors is one Tocon 47uf, 50v, and one NCC KY 4.7uf, 50v


    And the second PSU is a Generic, model T-500-48, rated 48v, 10.5A, but it might be slightly overrated.

    The switching transistors are two D13009K
    The primary capacitors are two fake-looking NCC KMH 820uf 250v (they have a butterfly in the datecode, do i need to say more)
    The control chip is a KA7500BD
    The secondary rectifiers are 2 20JL2C
    The (main) secondary capacitors are three Su'scon SK 1000uf, 50v
    The 12v secondary capacitor is a yw 100uf, 35v


    Any help will be appreciated. And if anyone is wondering why i need two high current 12 PSUs, the reason involves a lot of 12v LED strips.

    And if 12v is too far-fetched, then would 24v work? I can still find most of the parts i need in that voltage.
    Attached Files
    I'm not a expert, I'm just doing my best.

    #2
    Re: Modding Two PSUs to Output 12V

    Can I ask you why you want to do this

    Because there are 12 volt switching power supply out there that would probably work because you can find up 50 to 60 amp output
    9 PC LCD Monitor
    6 LCD Flat Screen TV
    30 Desk Top Switching Power Supply
    10 Battery Charger Switching Power Supply for Power Tool
    6 18v Lithium Battery Power Boards for Tool Battery Packs
    1 XBox 360 Switching Power Supply and M Board
    25 Servo Drives 220/460 3 Phase
    6 De-soldering Station Switching Power Supply 1 Power Supply
    1 Dell Mother Board
    15 Computer Power Supply
    1 HP Printer Supply & Control Board * lighting finished it *


    These two repairs where found with a ESR meter...> Temp at 50*F then at 90*F the ESR reading more than 10%

    1 Over Head Crane Current Sensing Board ( VFD Failure Five Years Later )
    2 Hem Saw Computer Stack Board

    All of these had CAPs POOF
    All of the mosfet that are taken out by bad caps

    Comment


      #3
      Re: Modding Two PSUs to Output 12V

      Originally posted by sam_sam_sam View Post
      Can I ask you why you want to do this

      Because there are 12 volt switching power supply out there that would probably work because you can find up 50 to 60 amp output
      I don't have any use for a 36v or 48v power supply, so they're kind of useless to me even though they work, and maybe i might be able to learn something along the way. I am fully aware that i can buy a 12v PSU online that outputs what i need out of the box, but i thought this might be an interesting challenge for me to modify one of these PSUs to output 12v. But i did realize this could be pretty far-fetched, and if that is the case, then i can just drop it here.

      One thing i should have added was that these two PSUs were found in a pile of trash, so i basically got them for free, which is part of the reason why i wanted to modify them (if i had bought them, i probably would not be asking this question).
      I'm not a expert, I'm just doing my best.

      Comment


        #4
        Re: Modding Two PSUs to Output 12V

        I going to post a few modification to switching power supply

        http://www.imajeenyus.com/electronic...ge/index.shtml

        http://www.kerrywong.com/2017/10/07/...utput-voltage/

        Just Google this “tl494cn modify”

        There is a lot of information out there but be careful not everything might not work or very complicated to do and sometimes hard to follow

        You also read something that I have been working

        https://www.badcaps.net/forum/showthread.php?t=92612
        Last edited by sam_sam_sam; 02-22-2021, 05:53 PM.
        9 PC LCD Monitor
        6 LCD Flat Screen TV
        30 Desk Top Switching Power Supply
        10 Battery Charger Switching Power Supply for Power Tool
        6 18v Lithium Battery Power Boards for Tool Battery Packs
        1 XBox 360 Switching Power Supply and M Board
        25 Servo Drives 220/460 3 Phase
        6 De-soldering Station Switching Power Supply 1 Power Supply
        1 Dell Mother Board
        15 Computer Power Supply
        1 HP Printer Supply & Control Board * lighting finished it *


        These two repairs where found with a ESR meter...> Temp at 50*F then at 90*F the ESR reading more than 10%

        1 Over Head Crane Current Sensing Board ( VFD Failure Five Years Later )
        2 Hem Saw Computer Stack Board

        All of these had CAPs POOF
        All of the mosfet that are taken out by bad caps

        Comment


          #5
          Re: Modding Two PSUs to Output 12V

          you could add a 12v regulator after the 36 or 48v outputs .
          the design of the circuit would depend on how much current you actually need
          Last edited by petehall347; 02-22-2021, 07:15 PM.

          Comment


            #6
            Re: Modding Two PSUs to Output 12V

            So i've mostly only read the first article and one thing i'm starting to notice is the Genssi 36v PSU is a slightly beefed up version of the one in the article, whereas the generic 48v PSU is a stripped down version.

            Now in the first article, there's a schematic of the 12v PSU he's modding:
            https://cdn.badcaps-static.com/pdfs/...0826aa92ed.pdf
            Now one thing i noticed is that the part of the circuit marked voltage sense is nearly identical between my two PSUs and the 12v one in the article, the only difference being the resistors and capacitors.

            Now i'm guessing if i were to swap the voltage sense resistors and capacitors on my two PSUs, then that should set the voltage to 12v on both units, right?

            I plan to try this on the 48v PSU since it lacks the protections the 36v PSU has.

            Though one thing i think might happen when i reduce the voltage is the auxiliary winding will no longer output enough volts to get the chip started, and if that does happen, i had two possible fixes in mind, the first one involves hooking up the secondary right where the auxiliary would connect, however, i have a sneaking suspicion this will most likely not work for some reason. Which brings me to my second one, which is to wrap a few extra windings on top of the coil of the transformer, and hooking those extra windings in series to the auxiliary windings, in that case, it might take a bit more trial and error, but i don't any reason why it wouldn't work (until someone tells me why it won't).

            In the meantime, i plan to go through the other article and thread that was linked, and maybe reverse engineer a schematic for the 36v PSU, since its design is slightly more complicated, and i think doing this will help me understand it a bit more.
            I'm not a expert, I'm just doing my best.

            Comment


              #7
              Re: Modding Two PSUs to Output 12V

              I hate to be a negative nancy, but I don't think these mods will work too well (if at all), because you're trying to change the voltages a bit too much. The reason I say this is because the main transformers of these PSUs have a certain number of primary and secondary turns. So this sort of dictates what the output voltages are going to look like. Sure that with very low PWM values (narrow pulse widths), the output may look like 12V. But because you're still trying to pull the same current as whatever the PSU was originally rated for, this may become infeasable at those narrow PWM pulses without modifying the output filter (caps and inductor), which in turn will also warrant modifying the compensation circuit of the PSU. So in the best case, you may get 12V on the output, but the PSU will likely oscillate badly and regulate very poorly. I seriously doubt you would be able to pull anywhere near the rated current at this low voltage. And that's of course assuming you can even get the PSU to turn on (will probably need modifications to the PWM IC power supply, as you noted.)

              Since one of the PSUs is rated for ~9 Amps and the other for ~10 Amps, why not get a single PSU that can do 20 Amps at 12V. A lot of used ATX PSUs can easily do that. I understand you stated you already have these and they were free. But I think the time and parts spend trying to modify these will not be worthwhile in the end... and moreover, you probably still won't get good working PSUs.

              Another question I have is... what kind of LED circuits are you powering that need this kind of current? Is it like some really big LED panel board? 20 Amps at 12V is 240 Watts - that's A LOT of power for LEDs (and a lot of power to dissipate too.) Are you driving the LEDs properly to limit power/current to what they are rated for?
              Last edited by momaka; 02-24-2021, 11:42 PM.

              Comment


                #8
                Re: Modding Two PSUs to Output 12V

                Yeah i'm starting to think about that, partially because i accidentally blew up the 48v PSU when i accidentally shorted the aux output while trying to check its voltage (before modding it), the odd thing is that the PSU took a few seconds to blow its transistors after shorting the aux output. I tried swapping the transistors and checking every component that was in the line of fire, but so far, no luck (i'm starting to think i somehow blew the PWM IC). And so because of that, i kind of lost a lot of motivation to work on this.

                As for the LED project, it involves roughly 12 5 meter rolls of RGBW LED tape, i'm not exactly sure how much power it's going to need yet, as i may under-drive the LEDs to make them not as intense. But i don't really need to get it done till November.

                While an ATX PSU may work, i sort of like these PSUs due to the fact that it's pretty easy to tweak the voltage up and down a bit, which can make a big difference with these LED strips.

                One thing i should add, i already have quite a few ATX PSUs on hand, way more than i'll ever realistically need, so i might have something in that department that could work with the right parts.
                I'm not a expert, I'm just doing my best.

                Comment


                  #9
                  Re: Modding Two PSUs to Output 12V

                  Originally posted by RukyCon View Post
                  Yeah i'm starting to think about that, partially because i accidentally blew up the 48v PSU when i accidentally shorted the aux output while trying to check its voltage (before modding it), the odd thing is that the PSU took a few seconds to blow its transistors after shorting the aux output.
                  Yeah, when working on PSUs - be it modding or just probing voltages - always have it plugged in to a series current/power limiting device, such as an incandescent light bulb or two (or a more powerful 200-500 Watt heating element if you want to still be able to use the PSU to power something.)

                  Otherwise you always run the risk of slipping or dropping something on the board and shorting something out.

                  Originally posted by RukyCon View Post
                  I tried swapping the transistors and checking every component that was in the line of fire, but so far, no luck (i'm starting to think i somehow blew the PWM IC). And so because of that, i kind of lost a lot of motivation to work on this.
                  It's a half-bridge design based around a KA7500 PWM IC, so it should be fairly straight-forward to troubleshoot. Just have to really get in there and remove all of the resistors and diodes connected to the primary BJTs, then check them out of circuit. The other parts that like to blow are the two driver transistor on the secondary side that are connected to the PWM IC, along with the protection diodes connected between E-C junctions on those driver transistors. See this thread:
                  https://www.badcaps.net/forum/showthread.php?t=90392
                  ^It's for an ATX PSU based on the same topology, but the troubleshooting technique should be pretty much identical.

                  If you don't find any of the abovementioned parts bad, then it could indeed be the PWM IC that got blown.

                  Originally posted by RukyCon View Post
                  As for the LED project, it involves roughly 12 5 meter rolls of RGBW LED tape, i'm not exactly sure how much power it's going to need yet, as i may under-drive the LEDs to make them not as intense.
                  Depends on how these LED strips are constructed. Generally, most of these have 3 same-color LEDs in series along with a resistor in series with those 3 LEDs. Depending on the resistor value, some of these strips may be way too over-driven at 12V, while others are OK. One of the first LED strips I bought about 8 years ago came with 120-Ohm resistors for the series resistors in each string. On 12V, those cheap LEDs did not last more than 2-3 months of 4-5 hours use per day and drew a lot of current per 5M roll (over 1 Amp, IIRC.) Running the few remaining identical strips at 9V yielded lower brightness but no more burned LEDs.

                  In contrast, I bought another LED strip recently (about 1 year ago), and that one came with 3x 180-Ohm resistors in series with each LED string for a total of 540 Ohms total series resistance per 3 series LEDs. Now these draw a lot less current @ 12V and IMO are a lot less likely to die in such short order (though we will see about that, given then are still cheap Chinese LEDs.) If I recall correctly, this new 5 Meter strip ate around 500 mA or less at 12V. So you may be able to run yours just fine from the 12V rail on a single ATX PC PSU. Just take one 5-M strip and measure how much current it draws at 12V, then multiply that by 12 obviously (since you have 12 strips.) Oh course, check your LED strips first and see how it is constructed, as well if there are series resistors and what their value is. From what I've saw once, some of the RGB strips have an MCU inside each LED, so brightness is controlled differently on those.

                  Originally posted by RukyCon View Post
                  While an ATX PSU may work, i sort of like these PSUs due to the fact that it's pretty easy to tweak the voltage up and down a bit, which can make a big difference with these LED strips.
                  If the voltage affects the brightness of the LED strips, then they likely have series resistors with the LEDs on each string.

                  However, you DON'T have to do voltage adjustment to tweak brightness/current consumption. You can do that with a simple PWM circuit while still feeding the LED strip with 12V. Just look up "555 circuit PWM motor control". Something like this should do, though is not the only way to build such circuit:
                  https://howtomechatronics.com/how-it...-555-timer-ic/

                  In fact, you can even find such pre-made modules on eBay nowadays for as much as it would cost you to get all the parts to build one yourself (and not to mention probably cheaper than buying parts to try to modify the two power supplies you showed in this thread.)

                  I suspect this module, for example, is based on the same or similar 555-timer design mentioned above:
                  https://www.ebay.com/itm/12V-24V-PWM...A/233691771070
                  If you swap the MOSFET and free-wheeling diode on the above module with higher-current parts, you should be able to drive all of the LED strips without issue. And with the PWM circuit, you should be able to get much more precise brightness control on the LEDs - from near 0% to near 100% brightness.
                  Last edited by momaka; 02-25-2021, 11:14 PM.

                  Comment


                    #10
                    Re: Modding Two PSUs to Output 12V

                    So i decided to troubleshoot the 48v PSU a bit more today, and found a dead .47 ohm resistor, so i replaced it with what i had on hand, and now the PSU tries to turn on, but is unable to. The main output manages to hover up to 32v after 15 seconds and the aux output sticks around 2.5v.

                    At this point, i've tested most of the componets between the transistors and the PWM IC, and so i'm starting to think the IC is bad.

                    I'm not sure if i should spit this off into its own thread, but i can do that if necessary.
                    I'm not a expert, I'm just doing my best.

                    Comment


                      #11
                      Re: Modding Two PSUs to Output 12V

                      Originally posted by RukyCon View Post
                      So i decided to troubleshoot the 48v PSU a bit more today, and found a dead .47 ohm resistor, so i replaced it with what i had on hand, and now the PSU tries to turn on, but is unable to.
                      If you'd like more input, please let us know which resistor you replaced (PCB designator) and where found / on which picture (if designator is not clearly visible in the pictures.)

                      Originally posted by RukyCon View Post
                      At this point, i've tested most of the componets between the transistors and the PWM IC, and so i'm starting to think the IC is bad.
                      IC is usually the last to go in half-bridge PSUs. 7500 and 494 chips tend to be tough nuts to crack. Also, of the secondary-side driver transistor (attached to the PWM chip) are not blown nor their protection diodes, chances are the PWM IC is still OK.

                      Originally posted by RukyCon View Post
                      I'm not sure if i should spit this off into its own thread, but i can do that if necessary.
                      Up to you.
                      I guess pictures of the PSUs are already here, so we might as well continue. Though on that note, the pictures of the 48V PSU are a little blurry and the solder-side one a little too dark to be able to trace paths. Might need better pictures, if possible.

                      Comment


                        #12
                        Re: Modding Two PSUs to Output 12V

                        Sorry it's been a while, i did not have a lot of time last week for working on this, and when i finally did, my 2 1/2 month old phone, which i use for taking photos, bricked itself.

                        Anyway, the resistor i replaced was R6, and i replaced it with two 1 ohm resistors in parallel.

                        The secondary transistors and diodes all appear to be in good shape, so i'm not really sure what to check next.
                        Attached Files
                        I'm not a expert, I'm just doing my best.

                        Comment


                          #13
                          Re: Modding Two PSUs to Output 12V

                          Originally posted by RukyCon View Post
                          The secondary transistors and diodes all appear to be in good shape, so i'm not really sure what to check next.
                          "Appear"? Or did you really pull them out and check?

                          If you didn't, remove diodes D10, D11, ZD1, and ZD2 and test them out of circuit on diode test mode. They should conduct one way but not the other. For the Zener diodes, it's even recommended to put a 10k resistor in series with the out of circuit and apply a high-enough voltage (I imagine up to 30V will do) to see what their reverse voltage is. Most likely it should match - especially if they have matching part numbers.
                          Also remove the two TO-92L transistors there - that is Q3(?) and Q4. These are probably NPN BJTs, so again check them for proper diode drops on the BE and BC junctions. They could be open-circuit or partially shorted. All those small resistors in the viscinity of the above diodes and transistors should be checked too. If anything tests off, remove from circuit and test again.

                          Since R6 was open, you need to remove both primary-side BJTs and check them for open-circuit between the BE and BC junctions. Can't do it in circuit - the protection diodes and primary caps will mess with the readings, along with the drive circuit. After you do this, proceed to test all of the resistors close to R6 too. It looks like the PWM controller is working if the PSU is able to pump out any voltage, so most likely that's not the problem. However, if all of the above parts do check out fine, then I guess just swap that PWM controller with another 7500 or 494 part (doesn't matter which - they are pin-compatible) and see if that fixes it.

                          On a different note, I don't see any other small BJTs or protection ICs on the secondary side, so I'm not sure if this PSU even has any protection circuits - certainly no OV and UV, I bet. SC might just be implemented in the form of limiting from the driver transformer... which is far from optimal. Ugh, this is why I dislike these cheapo Chinese designs - can never tell how many corners they cut too much.
                          Last edited by momaka; 03-19-2021, 10:30 PM.

                          Comment

                          Working...
                          X