7805 voltage regulator misbehaving?

Discussion in 'The Projects Forum' started by mrwilson, Jun 8, 2013.

  1. mrwilson

    Thread Starter New Member

    Jun 8, 2013
    8
    0
    My home security system is powered from a 240VAC to 16VAC transformer, with a 12VDC backup battery. The system rectifies the 16VAC providing an output DC voltage running to the control panel at around 15.5VDC.

    The system has been switched off for over 2 years, and I'm getting it going again. Having replaced the backup battery, I fired it up and found that running off the battery (ie. transformer switched off at the mains power outlet), the system works fine. However, when I run it off the transformer, the control panel stops working.

    The control panel contains a 7805 voltage regulator (spec'd to output 5VDC from input voltages up to around 36VDC). When running off the backup battery only, the input voltage at the 7805 in the control panel is 12V, and the output voltage (as expected) is 5V. When powered from the transformer, the input voltage at the 7805 in the control panel is 15.5V, however the output voltage measures at 1.3V (which is obviously not what is supposed to happen, expecting 5V output irrespective of the input voltage).

    The question is: is the problem likely to be caused by a faulty regulator? ...(if so, why does it work with 12V at the input?) ...or is it more likely a result of the filtering/smoothing capacitors on the circuit board not the regulator itself? ...(if so, please explain the likely cause)

    If you have a solution / thought on the cause of the problem, I'd also like recommendation(s) on what / where I should measure on the circuit board, or what else I can do to confirm the cause.
     
    Last edited: Jun 8, 2013
  2. GopherT

    AAC Fanatic!

    Nov 23, 2012
    6,058
    3,820
    I would imagine your filtering capacitors or bridge rectifier is bad. Measure both AC and DC voltage of the voltage going into the voltage regulator when the battery is disconnected. If AC is more than a volt or 2, your capacitors may be weak.

    The other option is that the regulator is going into protection mode because the power dissipation is too high (15v input - 5 v output)/amperage. The 7805 should handle an amp or more but that depends on the voltage drop and heat sink used. Do you know the current draw? Are you using a heat sink?
     
    mrwilson likes this.
  3. mrwilson

    Thread Starter New Member

    Jun 8, 2013
    8
    0
    GopherT - thanks for the fast response ...

    RE DC voltage input with the battery disconnected - that was measured at 15.5VDC. (I had the battery disconnected for all measurements using the transformer).

    I don't believe power dissipation is too high - doesn't even get a chance to get hot (won't even light up when the power source is the transformer). The current draw through the panel at standby is supposed to be 60mA per the installation manual (security system is a GE/Caddx Nx4). The regulator is not attached to a heatsink.

    I think you might be right RE the problem actually being either weak capacitors or a bad rectifier circuit. This is where my knowledge get's poor, so suggestions on what I need to measure / test here would be much appreciated. :)

    Fyi, I'm in Australia, so heading off to bed shortly and will measure the AC voltage at the regulator input tomorrow, along with anything else that needs to be measured.
     
  4. tracecom

    AAC Fanatic!

    Apr 16, 2010
    3,869
    1,393
    I know you said the input to the 7805 is 15.5VDC from the wall wart, but did you measure that with the wall wart connected to the circuit or just measure the output of the wall wart without it being connected? What I am suggesting is that perhaps the wall wart is bad, and when put under any load, it sags. Thus the 7805's output also sags.
     
  5. GopherT

    AAC Fanatic!

    Nov 23, 2012
    6,058
    3,820
    I didn't think it was a wall wort, he said a 16vac transformer.
     
  6. mrwilson

    Thread Starter New Member

    Jun 8, 2013
    8
    0
    16VAC transformer (connected to a mains power outlet), provides about 18.8VAC (measured) to the alarm control sub-system, where it is rectified to DC. All other system components (sensors, control panel - also known as the 'code pad', strobe light, siren, etc) are powered from DC output via terminals inside the alarm control sub-system. ...see attached wiring diagram from the installation manual.

    The DC voltage measured across the terminal output points (terminals 7 and 8 in the wiring diagram) for the code pad is about 15.2DC (I remeasured it again this morning). It measures the same at the input to the voltage regulator inside the code pad.

    With the system powered from the transformer, I just measured the AC voltage with disturbing results...

    1. At terminals 7 and 8 and the input to the code pad: 32.9VAC
    2. At the regulator inside the code pad: 31.5VAC.

    ...And just to make sure there's not something weird happening in the alarm circuit: 0VAC measured at all points while powered from the battery

    I assume this means that it's likely the rectifier circuit not doing its job...so, recommendations, please, on what and where to measure next to confirm the cause or even the bad component?
     
  7. tracecom

    AAC Fanatic!

    Apr 16, 2010
    3,869
    1,393
    I suppose that because you can choose to run the 5V regulator circuit from either the 12V battery or the mains supply, that would make it possible to substitute another mains supply. Is that correct, and do you have another source of rectified, filtered 15VDC? If so, you could definitely isolate the problem to the old mains supply.

    Is the mains supply a "module" that consists of the transformer, the bridge, and the filter caps, or are these components attached to a single PCB for the whole unit?

    Can you post some pictures of the mains supply?

    It sounds like GopherT may have called it right at the very beginning: bad bridge and/or bad filter caps. Often electrolytic filter caps are swollen or bulged when they go bad. Do you see any swelling of the sides or bulged ends? Likewise, do you see any signs of high heat near the diodes (or bridge module) or anywhere else for that matter?
     
  8. mrwilson

    Thread Starter New Member

    Jun 8, 2013
    8
    0
    Tracecom: I suppose that because you can choose to run the 5V regulator circuit from either the 12V battery or the mains supply, that would make it possible to substitute another mains supply. Is that correct, and do you have another source of rectified, filtered 15VDC? If so, you could definitely isolate the problem to the old mains supply.

    Is the mains supply a "module" that consists of the transformer, the bridge, and the filter caps, or are these components attached to a single PCB for the whole unit?

    ...

    Often electrolytic filter caps are swollen or bulged when they go bad. Do you see any swelling of the sides or bulged ends? Likewise, do you see any signs of high heat near the diodes (or bridge module) or anywhere else for that matter?


    >> I only have a choice on the power supply while I'm trying to diagnose the fault. The normal mode of operation is that the transformer (plugged into a power outlet to convert from 240VAC - being in Australia) provides the power to the alarm system. The battery is only there as a backup in the event of loss of mains power.
    >> I do not have another source of rectified, filtered 15VDC ...though perhaps I could connect 2 x 1.5VDC batteries in series with the 12VDC battery to see how the system handles clean 15VDC compared with apparently noisy DC. ...but not sure if that's what you're interested in.
    >> No swollen caps or signs of high heat near the diodes - that was the first thing I looked for when I started trying to work out what was wrong with my system.
     
  9. tracecom

    AAC Fanatic!

    Apr 16, 2010
    3,869
    1,393
    So, is the transformer a separate wall wart? And the rectification and filtering is inside the unit? I am trying to understand the configuration in order to think of a doable troubleshooting method.
     
  10. mrwilson

    Thread Starter New Member

    Jun 8, 2013
    8
    0
    Yes. Transformer is a "wall wart" (had to do a bit of hunting to work out what that terminology meant - not used downunder). The rectification circuit is on the circuit board for the alarm control system. The regulator is on the code panel circuit board - some distance from the control system, mounted near our front entrance.

    See attached pics. One shows the 4 diodes that I believe form part of the rectifier circuit. The others show the backup battery, transformer ("wall wart") - "hard wired" into the alarm control system, and circuit board with the mess of wires connected that run to the 4 sensors, sirens, smoke alarm, code panel, etc.
     
  11. MrChips

    Moderator

    Oct 2, 2009
    12,446
    3,362
    Try putting about 470uF cap after the rectifier diodes.
     
  12. mrwilson

    Thread Starter New Member

    Jun 8, 2013
    8
    0
    MrChips - thanks for your recommendation. Would you mind please explaining your reasoning and the expected outcome before I go to the effort of trying to solder a cap onto the circuit board? (which will be a first time for me - so really not keen unless I have some strong confidence that it will likely fix the issue)

    How will the addition of the 470uF cap remove the approx 32VAC that's appearing after the diode bridge?

    Any ideas as to what exactly is causing the 32VAC after the bridge?
     
  13. mrwilson

    Thread Starter New Member

    Jun 8, 2013
    8
    0
    For those of you that have responded thus far - thanks very much for your help. I'm running out of time to fix my alarm system and have concluded at present that the simplest thing to do (other than getting the system board replaced and reprogrammed) is probably to connect up a 12VDC power supply (240VAC to 12VDC converter) to the circuit instead of a 16VAC power supply. (Knowing that the 12VDC battery backup power supply runs the system with no problems.)

    Question: What would happen if I was to connect a 12VDC supply to the terminals that are currently connected to the 16VAC power supply (after removing the 16VAC power supply)? ...Is it possible that what ever is causing the fault in the rectifier circuit could also cause problems given a DC input? (ie. could the problematic caps - assuming this is the problem, also cause problems with a DC power supply?) ...What are the risks (if any) in connecting a DC power supply at the input to a rectifier bridge? ...Am I likely to cause additional damage to the circuit board? (if not, then I'll give it a go and see if it works)

    If there are any risks / expected issues with connecting the power supply as above, then I think my other options are (1) connect it to the backup battery terminals or (2) connect it to the circuit board at the connection points for the backup battery.

    Thoughts?
     
  14. MrChips

    Moderator

    Oct 2, 2009
    12,446
    3,362
    You ask many reasonable questions to which we cannot provide reasonable answers for lack of information. Any answer and possible outcome is purely speculation.

    I assume you do not have access to an oscilloscope for this would allow us to see actual waveforms in your circuit. Any AC supply after rectification will have significant voltage ripple. The purpose of the capacitor after the bridge is to smooth out the ripple. I would assume there is already a cap in the circuit serving this purpose. What is the value of this cap?

    The 470uF addition is to increase what is already there. There is no guarantee that it will solve your problem.
     
    Last edited: Jun 9, 2013
  15. Evil Lurker

    Member

    Aug 25, 2011
    117
    23
    Can't really get a good look at it, but I think I see three wires coming in off the wall wart. If that is the case then it is indicative of the wall wart having a center tap transformer in it providing a + and - rail. And, if that is the case, and the black wire is coded for earth or 0V, then that could possibly explain your 36v AC... a simple miswiring problem.
     
    Last edited: Jun 9, 2013
  16. tracecom

    AAC Fanatic!

    Apr 16, 2010
    3,869
    1,393
    Good catch!

    I think you are onto something. Looking at the wiring diagram in post 6, the black is supposed to be on terminal 15, and the AC on 16 and 17. It looks in the photo like the black is on 17.
     
  17. Evil Lurker

    Member

    Aug 25, 2011
    117
    23
    Its about the only thing that makes sense really. If you take the OP's voltage measurements into consideration, it's almost exactly twice what it should be.

    Typically from my understanding when a linear transformer takes a dump the voltage goes down, not up. And with a center tap transformer if you leave the center tap disconnected and hook it to a full bridge rectifier just like a normal transformer you will get the most voltage. With the center tap connected as 0v you can either have double the amperage at half the voltage, and/or a negative rail. Furthermore to my understanding, many of these sort of devices need a negative rail to properly operate the op amps connected to the sensors. Plus if it uses an LCD display I think there has to be some sort of negative voltage as they operate on AC current.
     
  18. tracecom

    AAC Fanatic!

    Apr 16, 2010
    3,869
    1,393
    He was just about to give up; maybe he'll check this thread one more time before chucking the whole thing.
     
  19. mrwilson

    Thread Starter New Member

    Jun 8, 2013
    8
    0
    Guys - haven't given up just yet ...however I think I still need to either replace the main board or (first) look at the 12VDC transformer option as a power supply (as I proposed in post #13) ...still waiting on some feedback to my questions related to that idea.

    Terminal 17 is connected to a black wire (neutral)
    Terminal 16 is connected to a white wire (active)
    Terminal 15 is connected to a yellow wire with green stripes (earth)

    This is the way it's always been connected (10+ years), since being installed by a licenced installer, with no problems.

    The alarm system was working fine up until 2008 when I moved overseas for work and rented the house out. While tenanted, due to not knowing how to reset the alarm after it had gone off, the tenant turned off the mains power to the system, running the backup battery dead. The alarm system has stayed off since then to my knowledge.

    When I returned to Australia a few months ago, and turned the alarm system back on, it set the alarms off straight away (still in the "latched" state from when it was set off in 2008) and so I turned it off again until I could find the instructions for to reset it. I then turned it on again, but this time the alarm didn't go off and the code panel didn't even light up ...thus began my investigation to work out what was wrong.

    What I'm wondering now is whether a capacitor or other component(s) may have failed when I switched the alarm system back on having been off for so long (I've read that some capacitors can fail when used after a lengthy period of non-use).

    If this is likely, is it possible that if I connect up a 12VDC power supply to terminals 16 and 17, that I could cause additional damage to the circuit board due to a short circuit? ...or damage the 12VDC power supply due to a short circuit on the board?
     
  20. tracecom

    AAC Fanatic!

    Apr 16, 2010
    3,869
    1,393
    It looks like I was completely wrong about the wiring; I apologize. Mr. Chips got it right when he said we are just guessing. I'll back off and hope someone else can help. Good luck.
     
Loading...