Announcement

Collapse
No announcement yet.

General PSU Discussion and FAQ contributions

Collapse
This is a sticky topic.
X
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    General PSU Discussion and FAQ contributions

    Using ATX and similar power supplies (P/Ss) as a pattern, P/Ss can be described as having seven sections: AC Filter; Rectifier & Filter; Inverter; Pulse Width Modulator; Output Rectifier & Filter; Voltage Sense Amplifier and Protection; Standby Regulator.

    AC Filter
    The AC filter is a network of inductors and capacitors on the “primary” side of the P/S intended to reduce the amount of noise conducted back into the AC power lines (can't be interfering with those TV and radio commercials). Capacitors connected from Line to Neutral (X capacitors) and single-winding inductors reduce noise that is on Line with respect to Neutral. Capacitors connected from Line and Neutral to chassis ground (Y capacitors) and two-winding inductors reduce noise that is common to Line and Neutral with respect to chassis ground. X and Y capacitors are special types that have approval marks from various countries' safety agencies. These should not be replaced with general purpose types. Y capacitors are always in pairs, and if replaced, the same value should be used.

    Rectifier & Filter
    The rectifier and filter convert the AC input voltage to DC. The input voltage selector switch is part of this circuit. In the “115V” position, the rectifier circuit is configured as a voltage doubler. In the “230V” position, the rectifier circuit is configured as a full wave bridge. In either case, the unregulated high voltage bus is about 300VDC. If the switch is in the “115V” position in Europe, the metal oxide varistors (MOVs) in parallel with the two input electrolytic capacitors will fire, blowing the fuse. If the MOVs fail (usually blowing up) before the fuse blows, the two input electrolytic capacitors, usually rated 250V, will vent. If the P/S is operated in the US with the switch in the “230V” position, the P/S may seem to operate OK in a system that isn't fully loaded. In a fully loaded system the P/S may not operate, or it may perform erratically. Either way, the P/S will run hotter and its life may be reduced.

    Inverter
    This is the heart of a switching power supply. The inverter converts the 300VDC into a square wave of varying duty cycle . The positive part of the square wave across the transformer primary (and the secondary windings, except for the -12V and -5V secondary winding) corresponds to the time when the switch device (now usually a MOSFET) is on. The transformer steps the 300Vp-p square wave down to the appropriate voltage levels.

    Pulse Width Modulator
    The output voltage of a switching power supply, after the output rectifier and filter, approximately equals the duty cycle times the peak voltage of the square wave. Since the 300VDC bus is unregulated, that peak voltage varies with the AC input voltage. Regulation of the output voltage is achieved by controlling the duty cycle of the square wave. This is done by the pulse width modulator, which compares the output voltage (indirectly) to a sawtooth voltage to determine the width of each positive output pulse. In some designs, the sawtooth voltage is a sample of the transformer primary current, which is converted to a voltage by a low value resistor between the switch MOSFET Source and primary return.

    Output Rectifier & Filter
    Though the output rectifiers are usually two rectifiers in one package, in the most common topology, the two devices have different functions. One of the devices conducts during the “on” part of the square wave from the transformer, charging the inductor. The other rectifier conducts during the “off” time, providing a current path for the discharge of the inductor. The output inductor and output capacitors form a filter that convert the square wave to a DC voltage. The current through the inductor has two components - the current to the load, and the charging-discharging current super-imposed. The current through the filter capacitor(s) is the inductor charging-discharging current plus some transient currents due to large changes in the load current. Since the ripple current through the capacitor(s) can be several amps rms, low impedance capacitors are absolutely essential. This side of the P/S is commonly referred to as the “secondary” side.

    Voltage Sense Amplifier and Protection
    Here is the control center of the power supply. The main feedback circuit senses the +5V output and often the +12V output as well. The +3.3V output is usually mag-amp (magnetic amplifier) post-regulated. The -12V and -5V outputs are usually post-regulated with three-terminal linear regulators (Mac doesn't use a -5V output). P/Ss also typically have circuits that monitor the +5V, +3.3V, and +12V outputs going out of regulation high. If this occurs, the circuit will latch the P/S off (cycling AC power will clear the latch condition, but if the problem cause is still present, the P/S will latch off again). There may be circuits that sense the output currents for the +5V, +3.3V, and +12V outputs that will limit the current on the output(s) and reduce the output voltages of those outputs. Many P/Ss have a temperature sensor located on the heatsink for the output rectifiers (this is usually expected to be the hottest spot). This sensor connects to a circuit that will turn off or latch off the P/S if the temperature exceeds a preset threshold. In some P/Ss, the output voltage sense amplifier and protection circuits are integrated into a custom IC.

    Standby Regulator
    The Standby Regulator is a separate, low power, P/S. Unlike the main P/S, it is active almost immediately after plugging in the power cord. Depending on the current rating, this circuit may be a self-oscillating inverter with a 3-terminal linear post-regulator or a higher current switching regulator. This circuit powers power management circuitry on the computer motherboard (including the on/off function), as well as secondary-side protection circuits (e.g. latch-off protection functions).
    PeteS in CA

    Power Supplies should be boring: No loud noises, no bright flashes, and no bad smells.
    ****************************
    To kill personal responsibility, initiative or success, punish it by taxing it. To encourage irresponsibility, improvidence, dependence and failure, reward it by subsidizing it.
    ****************************

    #2
    Re: Power Supply Design Description (first post is long)

    Please PM me with suggestions where this isn't clear and with GP questions. I'll update the first post A/R and answer Qs by PM or here.
    PeteS in CA

    Power Supplies should be boring: No loud noises, no bright flashes, and no bad smells.
    ****************************
    To kill personal responsibility, initiative or success, punish it by taxing it. To encourage irresponsibility, improvidence, dependence and failure, reward it by subsidizing it.
    ****************************

    Comment


      #3
      Re: Power Supply Design Description (first post is long)

      looks good to me.
      good work!

      Comment


        #4
        Re: Power Supply Design Description (first post is long)

        Here are links to TI's datasheets for three of the most common pulse width modulator ICs:

        SG3524 (KA7500 is similar): https://cdn.badcaps-static.com/pdfs/...0b64dfc01f.pdf

        TL494: https://cdn.badcaps-static.com/pdfs/...22b539cab9.pdf

        UC3842: https://cdn.badcaps-static.com/pdfs/...7f7a055a4f.pdf

        The datasheets incude chip schematics, descriptions of how the ICs operate, and sample application circuits. The SG3524 was introduced in 1975; the TL494 a few years later. The UC3842 family is newer, introduced in the mid-80s. Good stuff remains usable for a long time.
        PeteS in CA

        Power Supplies should be boring: No loud noises, no bright flashes, and no bad smells.
        ****************************
        To kill personal responsibility, initiative or success, punish it by taxing it. To encourage irresponsibility, improvidence, dependence and failure, reward it by subsidizing it.
        ****************************

        Comment


          #5
          Re: Power Supply Design Description (first post is long)

          KA7500B is pin-compatible with TL494 - they can be interchanged. They are also used on some ECS/PC Chips boards (like K7S5A) for CPU power, some older boards (like M726MRT) use MC3842 (or UC3842 or something3842 - they're same). I've never seen these chips bad on a board, however I've seen many of the special VRM controllers dead (because of PSU or shorted mosfet).
          Last edited by Rainbow; 08-31-2005, 12:43 PM.

          Comment


            #6
            Re: Power Supply Design Description (first post is long)

            The multi voltage high amperage switching power supplies have always boggled me.
            I just replace them thinking that when the go they're gone, chared and burnt.
            For some reason i don't dispose of them and have a closet full.
            I have modified some PSU's but have not been able to repair them.
            Your description is very helpful and the links to chip specs a nice touch.
            My IC Master is way outdated and HW Sams may have Photophacts going back to Edison's day but not to much help today.
            Sorry to ramble, I'd love to find a typical schematic for one of these.
            Jim

            Comment


              #7
              Re: Power Supply Design Description (first post is long)

              Any thoughts on active Power Factor Correction, I have had some experience with this additional piece in the PSU puzzle, but is it going to become commonplace for most PC PSU's. The units I have seen so far
              have extra devices like LT1249 and the power mosfet it controls, switches across the output of the main rectifier. ( This takes some understanding)
              Not directly though, it does use a type of snubber to absorb/redistribute
              the energy.
              It seems that when things go wrong with this part of the circuit,
              they go horribly wrong (at least in my experience).
              I'm still having some trouble comming to grips with the application/implementation some engineers have decided upon.

              http://focus.ti.com/docs/prod/folders/print/uc3853.html
              http://www.onsemi.com/PowerSolutions....do?id=MC33262
              http://www.linear.com/pc/productDeta...42,C1138,P1417

              It may eventually make a difference, to power consumption, but it is also adding a curve ball for us techs.
              Andrew

              Comment


                #8
                Re: General PSU Discussion and FAQ contributions

                Could somebody post a block diagram or schematic for a typical ATX supply. I would like to learn more about where the various caps fall in the circuit, and relate it back to Pete's Seven Step Program above.

                Comment


                  #9
                  Re: General PSU Discussion and FAQ contributions

                  bit dated but may help you


                  ATX 200W TL494

                  here SG6105 controller

                  very simple block diagram here

                  not really a block but might help with peteS explanation above

                  Here


                  pc psu site here

                  some what on the technical side PDF here

                  The FAQ here

                  seek on the net and ye shall find...well mostly

                  HTH cheers
                  Last edited by starfury1; 06-16-2007, 09:05 PM.
                  You step into the Road, and if you don't keep your feet, there is no knowing where you may be swept off to." Bilbo Baggins ...

                  Comment


                    #10
                    Re: General PSU Discussion and FAQ contributions

                    looks like the very simple bock diagram link above has died

                    So thought id fix it with another link and then decided to round it up a bit
                    So hope no one minds

                    Here is the wiki on Switched-mode power supply

                    The Wiki should help expand on PeteS's Great written explanation of
                    "How SM-PSU's works"
                    posted above.


                    Here is a photo from the Wiki of the "typical supply" outlined in this thread (large 3Meg file at Wiki)



                    Copyright notice from wiki photo....
                    Public domain I, the copyright holder of this work, hereby release it into the public domain. This applies worldwide.
                    I, grant anyone the right to use this work for any purpose, without any conditions, unless such conditions are required by law.


                    Here is a BAD example of the Above


                    Copyright Me and BADCAPS for anybody that cares

                    I should add that the outline here is of Now older technology
                    It is still the basis of how they work regardless.

                    The next evolution and mostly what you going to see is the above but with a modified Mains side to Incorporate What is know as
                    "Active Power Factor Correction" A-PFC
                    (thats putting it simply, and the construction,Design does go further then that in reality)

                    This is what your more likely to see these days
                    NOTE the small PCB and 1 large Capacitor, that should be a bit of a giveaway its an APFC type.
                    (but there are exceptions to any rule of thumb and the really big PSU's will probably have two
                    and some older ones may have 1 (I think), bottom line check!!!)



                    Liberty 620W (some of these have been known to go boom!)

                    Check this thread out for photos of the many variations and themes you find with SMPSU (thanks go to Dood)

                    Power supply build quality pictorial

                    Thanks PeteS for the Excellent work you did and others for contributing to it.

                    Cheers
                    Attached Files
                    Last edited by starfury1; 02-16-2008, 06:11 AM.
                    You step into the Road, and if you don't keep your feet, there is no knowing where you may be swept off to." Bilbo Baggins ...

                    Comment


                      #11
                      Re: General PSU Discussion and FAQ contributions

                      Just a quick question:

                      When the PSU's "-12v" is reading "0.97v" is that a sign of impeding death????
                      CPU: Sempron 2500+ / P4 2.8E / P4 2.6C / A64 x2 4000+ / E6420 / E8500 / i5-3470 / i7-3770
                      GPU: TNT2 M64 / Radeon 9000 / MX 440-SE / 7300GT / Radeon 4670 / GTS 250 / Radeon 7950 / 660 Ti / GTS 450

                      Main Driver: Intel i7 3770 | Asus P8H61-MX | MSI GTS 450 | 8GB of NO NAME DDR3 RAM (2x4GB) | 1TB SATA HDD (W.D. Blue) | ASUS DVD-RW | 22" HP Compaq LE2202x (1920x1080) | Seasonic S12II-620 PSU | Antec 300 | Windows 7 Ultimate with SP1

                      Comment


                        #12
                        Re: General PSU Discussion and FAQ contributions

                        I read it could be due to failing capacitors and a PSU change is called for ...apparently those bundled in CPU packages tend to be less than top quality
                        Good i nice to you
                        spam removed!

                        Comment


                          #13
                          Re: General PSU Discussion and FAQ contributions

                          could anyone share a what to do list for repair power suply.. starting from cheking the suspicious broken part...

                          and now I'm share some of my experience.. in repair PSU

                          PSU short.... chek rectifier diode... and replace with similar parts

                          no power but 5vsb..... chek primary power transistor or nfet..do search in replacement data sheet.. and replace it with same similar parts that has similar value in datasheet....

                          strange sound on psu.... replace primary (biggest one) capacitor......

                          that's all
                          Last edited by prongs12; 07-23-2013, 05:31 AM.
                          Micron Scrap

                          Buying Electronics scrap since 2005

                          Comment


                            #14
                            Re: General PSU Discussion and FAQ contributions

                            Hi I have a 715t2463-3 psu for a aoc 32" tv, the fault it's having is that it fires up the shuts down after 5 secs to a flashing orange light, I have changed several of the caps but still do the same has anyone else coming across this problem

                            Comment


                              #15
                              Re: General PSU Discussion and FAQ contributions

                              plus tôt changé les deux condos . C877 ET C878

                              Comment


                                #16
                                Re: General PSU Discussion and FAQ contributions

                                hello yes capacitor must be change

                                Comment


                                  #17
                                  Re: General PSU Discussion and FAQ contributions

                                  Hello,

                                  The fan in my ATX PSU stopped spinning. It is clean however, not jammed, and works properly when connected directly to +12V. The PSU's type as it is printed on the PCB is FSP400-60GLY (labeled Gigabyte 460W on the outside of the casing, figures... ). When turned on, the fan spins a few times, then it stops. Unfortunately I cannot find a shematic for this model, found another for a similar one (FSP500-60GLN) on this forum, but it seems to be incomplete. Can you please give me any hint on this ?
                                  Thanks in advance, Peter

                                  Comment


                                    #18
                                    Re: General PSU Discussion and FAQ contributions

                                    I have a Samsung Plasma (ps51d6900dm) turning on issue:- The problem is the tv turns on and gets to the logo but then turns off! tries 3 times and for safety reasons stops trying to turn the tv on i believe.
                                    I did a few checks and by no means am i very knowledgeable when it comes to fixing tv's but i checked all my voltages and they seem to be fine and disconnected other boards yet still came to the conclusion its the power supply.

                                    I then heard about heating the board with a hairdryer to see if it turns on! which it did turn on. So i isolated it to a part of the board and found when i pushed on the IC the tv stayed on when not heated. I then believed it was a cold solder joint. It worked after re soldering this area for a week! then the same problem happened again! Is this part likely to be causing this problem and if so where can i get this part as i have had no luck looking myself!

                                    Photo of IC i believe is faulty! in link
                                    http://www.avsforum.com/forum/attach...7&d=1408930516

                                    Comment


                                      #19
                                      Re: General PSU Discussion and FAQ contributions

                                      Does anyone have some tips on how to remove or clean off the white stuff that is covering a lot of the PSU components that appear either to be used for a sort of tamper proof, or to hold the components more securely?

                                      Comment


                                        #20
                                        Re: General PSU Discussion and FAQ contributions

                                        Originally posted by Roving Geek View Post
                                        Does anyone have some tips on how to remove or clean off the white stuff that is covering a lot of the PSU components that appear either to be used for a sort of tamper proof, or to hold the components more securely?
                                        A lot of companies use glue to hold components in place to improve production times. If the glue is white, you can leave it (If there's a lot of it though I try to remove most) but if the glue is tan or brown, you need to remove it as it will go conductive over time. I usually remove the components covered in glue to access the glue easier for cleaning, and just heating the board helps a lot with getting it off. I use pliers too

                                        Comment

                                        Working...
                                        X