Page 1 of 2 12 LastLast
Results 1 to 10 of 11

Thread: Fault finding a switch mode power supply.

  1. #1
    Join Date
    Dec 2011
    Location
    UK S80 postcode
    Posts
    1,554
    Post Thanks / Like

    Default Fault finding a switch mode power supply.

    Hi, I have 2 Chauvet LED moving heads one works fine the other has no output on the 24V power supply. If I remove the supply and run it from 24V on my bench supply the unit also works fine. If all else fails I will replace the 24V supply, the original is obsolete and I’m struggling finding one that will fit in the space so if possible I would like to repair it. It appears that power supply issues are not uncommon on older units as there are quite a few for sale as spares or repair (will not power on).

    I’ve attached a schematic, it looks like the semiconductors are Onsemi brand (1230ic, 2907A & A1020), I’ll link the data sheets. I’m familiar with this sort of voltage and am using an isolating transformer when mains is needed.
    IMG_1336.jpg

    What I’ve done to date:

    Powered up with no load on the 24V secondary, the input section (bottom left on the schematic) works ok. Oops I missed off a capacitor on that section, 100uF 400V after the bridge. Anyway the voltage is good and stable.

    There is 320V on pin 8 of the 1230.

    The 100uF capacitor across pins 6 and 4 of the 1230 charges to about 13V but then discharges to 7V ish and then repeats indefinitely. This is understandable behaviour (Ref data sheet) and would be expected if the P2 winding or related components had failed. I’ve changed the 100uF 35V electrolytic capacitor and removed the diode on pin 4 which tests as ok.

    I left this diode out and powered across the capacitor with 12V with my bench power supply to make things safer. I get 12V on the components connected to pin 6.

    12V on the top side of the Zener diode on pin 3 and 0V on the other.

    12V on the top side of the 100K that connects to the 100R feedback resistor and pin 3 (A).

    There is 4V on pin 2, other than that everything else is at 0V.

    All the diodes, transistors and resistors in circuit tested the same as a working board with no power.

    I’ve scoped both boards, the components connected to pin 5 show short trigger pulses (with no load on the secondary output) whereas the none working board shows nothing.


    Switch mode power supplies are not really my thing, I get the general gist:

    The mains comes in through the fuse and then is inrush current limited through a PCT device, noise it removed with a choke and capacitor, this is rectified and charges the 100uF 400V capacitor I missed off the schematic.

    On power up Pin 8 is fed from this high voltage and charges the 100uF 35V up to about 13V, this powers up the 1230 which should then disconnect the HV feed on pin 8 and start pulsing from pin 5 the windings P1 on primary side of the transformer. This is happening but stuck in a 13V disconnect and at 7V reconnect loop as pin 5 is not pulsing.

    P2 should then take over the generation of the 12V supply keeping the 100uF 35V capacitor charged across pins 6 and 4. Unfortunately this is not happening.

    Any thoughts or tests I should do before ordering new parts? I have a working supply so can take measurements from voltage both high and low and can scope the low voltage parts of the circuit.

    I’d be interested in how the circuits attached to pins 3 and 5 are working if anyone can offer insight.

    Thanks.

    1230 data sheet: https://www.onsemi.com/pdf/datasheet/ncp1230-d.pdf
    2907a data sheet: https://www.onsemi.com/pdf/datasheet/pzt2907a-d.pdf
    A1020 data sheet: https://www.onsemi.com/pdf/datasheet/2sa1020-d.pdf
    Last edited by Barnabybear; 09-09-2021 at 07:01 PM.

  2. #2
    Join Date
    Nov 2009
    Location
    Brisbane - Australia
    Posts
    978
    Post Thanks / Like

    Default Re: Fault finding a switch mode power supply.

    I'd replace the low ESR 100uF on the primary side. After that the controller IC (1230).
    www.da-share.com

  3. #3
    Join Date
    Dec 2011
    Location
    UK S80 postcode
    Posts
    1,554
    Post Thanks / Like

    Default Re: Fault finding a switch mode power supply.

    Hi David, thanks for that advice.

    I’ve been through the board again a couple of times and compared the resistances and capacitances at nearly every point with no power and they are virtually identical. I changed the 100uF 35V on the primary side as you suggest this has slowed the cycling speed but not stopped it. The original part is marked with the ON logo and 1230P65, unfortunately I can’t fine one in through hole, there are P133’s available that are on a 14 day delivery and very expensive. Running at double the frequency I’m not convinced these would be a drop in replacement, it would have an effect on the transformer and without checking it may put other components close or past their upper frequency. Whilst not entirely happy I’ve opted for the 230D6 in the SOIC8 package. This means I’m going to have to use an adapter board all of which results in less clearance, not ideal but hey they make these in this package and spec them to 500V on pin 8. I might add some form of coating in that area to reduce the chance of surface tracking due to contamination as the units are quite dusty from contamination drawn in through the fan.

    Again thanks for the advice.

  4. #4
    Join Date
    Dec 2011
    Location
    UK S80 postcode
    Posts
    1,554
    Post Thanks / Like

    Default Re: Fault finding a switch mode power supply.

    Hi, well this has been a fun project – incredibly frustrating and taken up more time than the supply is worth BUT IT’S WORKING!

    Sorry for the caps but I’m so pleased that I’ve been able to work through this and find the fault. What did it take? Hours of my life I’ll never get back but I now have a limited understanding of something I didn’t. To be honest I could have worked for minimum wage and purchased two replacements for the research and testing I’ve done. But what price knowledge?

    After taking photos (front and back) flipping one to so I could draw the components on the copper side and then creating a schematic, I tested all the components as best I could whilst in circuit and then replaced the 100uF 35V low voltage power supply capacitor all to no avail.
    So I brushed up on my knowledge of PNP transistors, worked out how I thought the circuit worked and tested those parts many times and all where good.

    A lesson for me and possibly worth noting for others:

    I had in my head that the fault most be on the primary side of the transformer circuitry as there was no power to the primary side of the transformer. When having tested time and time again and not being able to find a fault I started looking at the secondary side of the transformer and it’s circuitry.
    The issue was a failed Zener diode in the feedback circuit on the secondary side of the transformer.
    As far as I can work out, the secondary side is a sort of request more power circuit. So at power up the primary side sends some power over to the secondary side and says if you want more let me know, because of the failed Zener diode it never requested more power and as a result most of the circuits on the primary side appeared to be dead as they had nothing to do.

    Was this worth the time and effort? Yes, I had fun mostly. I’ve increased my knowledge, have two working moving heads and there are quite a few more listed on auction sites that will not power up.

  5. #5
    Join Date
    Dec 2011
    Posts
    7,337
    Post Thanks / Like

    Default Re: Fault finding a switch mode power supply.

    That is awesome that you solved this . Is the zener an over V protection diode ? Do you have suspicion as to what caused it to fail ? Or maybe it did not fail and served its intended purpose .

  6. #6
    Join Date
    Dec 2011
    Location
    UK S80 postcode
    Posts
    1,554
    Post Thanks / Like

    Default Re: Fault finding a switch mode power supply.

    Hi Angus, the Zener supplies the diode side of a feedback opto-coupler. As far as I can see it just drops the 24V down to 12V and is only passing about 18mA, so I’m not sure why it failed. It is a through hole glass bead type and before I removed it from the board it had broken into two parts (cracked in the centre) so it could have been physical shock. I’ve included the schematic below.

    IMG_1337.JPG

    Obviously with the failed Zener diode the opto-coupler would never provide any feedback. I believe the primary side was shutting down believing that there was a short circuit on the 24V output of the supply. From the data sheet and the circuit above it looks like the feedback is provided by driving the opto-coupler LED harder the greater the voltage of the 24V output, but if the output was shorted the voltage may never reach 24V and no feedback would be provided. To get round this there is a 125mS timer in the NCP1230 ic, once the supply is running if feedback isn’t received before this timer times out the supply is shut down, I think this is what was happening.

  7. #7
    Join Date
    Dec 2011
    Posts
    7,337
    Post Thanks / Like

    Default Re: Fault finding a switch mode power supply.

    Quote Originally Posted by Barnabybear View Post
    Hi Angus, the Zener supplies the diode side of a feedback opto-coupler. As far as I can see it just drops the 24V down to 12V and is only passing about 18mA, so I’m not sure why it failed. It is a through hole glass bead type and before I removed it from the board it had broken into two parts (cracked in the centre) so it could have been physical shock. I’ve included the schematic below.

    IMG_1337.JPG

    Obviously with the failed Zener diode the opto-coupler would never provide any feedback. I believe the primary side was shutting down believing that there was a short circuit on the 24V output of the supply. From the data sheet and the circuit above it looks like the feedback is provided by driving the opto-coupler LED harder the greater the voltage of the 24V output, but if the output was shorted the voltage may never reach 24V and no feedback would be provided. To get round this there is a 125mS timer in the NCP1230 ic, once the supply is running if feedback isn’t received before this timer times out the supply is shut down, I think this is what was happening.
    I have been reading about opto's and it seems 10mA is about the normal required . Had always thought zener diodes were primarily used as over voltage protection. Zeners are a cheap form of protection similar to a fuse . I am betting that this is a common issue to that lighting elements power supply .

  8. #8
    Join Date
    Nov 2009
    Location
    Brisbane - Australia
    Posts
    978
    Post Thanks / Like

    Default Re: Fault finding a switch mode power supply.

    Good to hear you got it working.

    SMPS can be tricky to fault find due to the feedback mechanisms involved.
    www.da-share.com

  9. #9
    Join Date
    Dec 2011
    Location
    UK S80 postcode
    Posts
    1,554
    Post Thanks / Like

    Default Re: Fault finding a switch mode power supply.

    Hi David, yes this was interesting. For the younger amongst us when I did my apprenticeship back in 1980 this would have been nearly impossible to fix, to get data sheets for components you would have to send a letter to the manufacturers containing a payment and a self-addressed envelope with stamps to send it to you. This not only had a cost but took time, now within a few seconds you can view them online.

    As the evenings are getting darker I’ve been able to do some tests tonight outside. I was concerned that even though my house is white these might not be bright enough given that there is a streetlight on the corner of my plot. I can confirm that they are going to be better than I had expected, the white is super bright and the colours work well. Issues are going to be focus and size, the gobos are going to need much smaller cut-outs than I had expected as the effects are much larger than I’m wanting to use.

    I’ll set up a better test over the weekend and post some video.

  10. #10
    Join Date
    Mar 2010
    Location
    Mentor, OH
    Posts
    1,179
    Post Thanks / Like

    Default Re: Fault finding a switch mode power supply.

    Iíve enjoyed reading this post. Barnabybear, thank you for the detailed explanation of how you troubleshooted this problem. Doing board level repair is time consuming as you indicated, however, the reward is satisfying. Saving these lights from the landfill is a bonus. Iím looking forward to your test videos and your final display setup.

    There is a wealth of knowledge from the members on this forum and I am thankful for all the sharing of information here.

    I enjoy taking a broken item and attempting to repair it. My knowledge is limited in the electronics field (electrician by trade) and following some of these posts, helps with my repairs. Learning to correctly troubleshoot is key in success.
    Chuck P
    nothing to show

Page 1 of 2 12 LastLast

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •