1. Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.
    Dismiss Notice

Help with Water Pump

Discussion in 'General Electronics Chat' started by salty joe, Dec 30, 2011.

  1. alec_t

    alec_t Well-Known Member Most Helpful Member

    Joined:
    Jul 10, 2011
    Messages:
    9,255
    Likes:
    1,218
    Location:
    Cardiff, Wales
    Then either the collector/emitter values are the wrong way round or the transistor is.
     
  2. salty joe

    salty joe Active Member

    Joined:
    Dec 30, 2011
    Messages:
    838
    Likes:
    28
    I was looking at this backwards and....sorry.

    The emittor, the number 1 pin, has 12.05V and the collector, the number 3 pin has 1.66V.
     
  3. alec_t

    alec_t Well-Known Member Most Helpful Member

    Joined:
    Jul 10, 2011
    Messages:
    9,255
    Likes:
    1,218
    Location:
    Cardiff, Wales
    We've all been there. Datasheets can be ambiguous too. It's not always clear whether they show a top view or bottom view of a component.
    If the alarm sounds while the collector (pin 3) voltage is <~2V then the fault is in the alarm module or the connections to it.

    Edit: In the alarm module, try connecting a ~10k-33k resistor from Q1 base to ground. That will reduce the sensitivity of the alarm to spurious voltage on Q1 base.
     
    Last edited: Dec 7, 2013
  4. dave

    Dave New Member

    Joined:
    Jan 12, 1997
    Messages:
    -
    Likes:
    0


     
  5. salty joe

    salty joe Active Member

    Joined:
    Dec 30, 2011
    Messages:
    838
    Likes:
    28

    The alarm module is not on the same board as the pair of PDMs. Would it be a good test to send 2V to the alarm module with a wall wart if I can find the right one in my junkpile?

    Should this resistor be permanent

    I'll get back on this tomorrow-we're gonna be at an allll day long wrestling tournament. Thanks.
     
  6. alec_t

    alec_t Well-Known Member Most Helpful Member

    Joined:
    Jul 10, 2011
    Messages:
    9,255
    Likes:
    1,218
    Location:
    Cardiff, Wales
    Depending on the particular red LEDs you have in the alarm module 4.5V (3 AA cells in series) should sound the alarm, 3V (2 cells) might, but 1.5V shouldn't. All channels of the alarm module should respond in the same way; any odd man out is your suspect.
    By rights it should, so that there is a pull-down to prevent Q1 base 'floating'. That doesn't happen in simulation but could in the real world. An oversight on my part :oops:.
     
  7. salty joe

    salty joe Active Member

    Joined:
    Dec 30, 2011
    Messages:
    838
    Likes:
    28
    I have not messed around with the alarm module-it could have a problem-as far as I know it's good.

    Since the only thing on the board with PDM 12 (the one with the alarm problem) is troblefree PDM 11 and they are isolated on the bench, I've been looking at PDM 12.

    I compared voltage at junction points, IC pins and terminal points between the two PDMs. The only difference I could find (besides ~12V at PDM 12 alarm) was IC pins 3, 8, 9 and 14 on PDM 11 was 12.05V and on PDM 12 those pins showed 11.26V.

    Any ideas, or is it time to throw in the towel? Thanks.
     
  8. alec_t

    alec_t Well-Known Member Most Helpful Member

    Joined:
    Jul 10, 2011
    Messages:
    9,255
    Likes:
    1,218
    Location:
    Cardiff, Wales
    Good thinking to compare the two PDMs. Having a 'known good' unit as a reference is usually the quickest way to track down a fault.
    If you've got a solid 12.05V on pin 14 in PDM11, but only 11.26V on pin 14 in PDM12 then there would seem to be a poor connection from the 12V supply to the latter pin.
    Never!
     
  9. salty joe

    salty joe Active Member

    Joined:
    Dec 30, 2011
    Messages:
    838
    Likes:
    28
    A poor connection would be an understatement. I either forgot to solder it or it was the world's worst and coldest joint. There was a loose wire in the 12V system.
    Now we have 12.05V at the PDM 12 IC and the alarm shows 0V!!! Thanks for pointing that out-I was thinking bad IC.

    Thanks, I appreciate that!
     
  10. alec_t

    alec_t Well-Known Member Most Helpful Member

    Joined:
    Jul 10, 2011
    Messages:
    9,255
    Likes:
    1,218
    Location:
    Cardiff, Wales
    Looks like we've nailed it then :). Hope all's well when you put the PDM back into the system.
     
  11. salty joe

    salty joe Active Member

    Joined:
    Dec 30, 2011
    Messages:
    838
    Likes:
    28
    Hey Alec, would you be willing to look at another PDM problem? This one is lucky PDM13.

    The trip test did not work. I tracked it down to IC pin 10 connected to the same junction as IC pins 5 & 6. Now the trip test works, including the alarm, but the alarm will not sound if PDM 13 power supply is unplugged. Did I likely damage something with the wrong hookup?

    Just in case I didn't ruin anything, and you'll have a look I took down some numbers. PDM 13 and problem free PDM 14 share a board.

    With power suppllies plugged in, PDM 13 alarm-0V,PDM 14 alarm 0V. PDM 13 Isense-0.33V, PDM 14 Isense-0.29V. PDM 13 fault-0.02V, PDM 14 fault-1.11.

    With power suppllies unplugged , PDM 13 alarm-0V,PDM 14 alarm 11.82V. PDM 13 Isense-0V, PDM 14 Isense-0V. PDM 13 fault-0.02V, PDM 14 fault-10.28V.

    I've spent many hours trying to figure this out and I am stumped. I can't find any wrong hookups and the V at coresponding junctions in PDM 13 & PDM 14 all check out. I can't find any issues with the FSM either.
     
  12. alec_t

    alec_t Well-Known Member Most Helpful Member

    Joined:
    Jul 10, 2011
    Messages:
    9,255
    Likes:
    1,218
    Location:
    Cardiff, Wales
    You do like to keep a chap busy, Joe ;).
    Not in this case, fortunately. If the trip test on PDM13 works then the IC in PDM13 is ok.
    From those useful measurements it looks like the problem is a poor connection (Isense or Vt) to the FSM channel from PDM13, or from the FSM (Fault) back to PDM13.
    Or the FSM channel itself could have the problem. You can check the FSM channel by temporarily shorting its Isense input (right at the FSM terminal) to ground. That should sound the alarm (after a brief delay) without turning the pump off.
     
    Last edited: Dec 13, 2013
  13. salty joe

    salty joe Active Member

    Joined:
    Dec 30, 2011
    Messages:
    838
    Likes:
    28
    Many thanks. Your ground Isense test confirmed a problem at FSM channel 13. It would not sound the alarm. I'll take a super close look tomorrow.

    BTW, I realized what caused the (previously mentioned) pump LED to glow for over 5 minutes. If a pump is not plugged in, that pump's LED glows for awhile after the power is turned off.

    And an odd thing, some PDMs do better with some new controllers/new pumps than others in terms of setting off the alarm and the 'flick'. It does not seem to matter where I set the PDM trimmer. So, I'll be playing matchmaker.:)


    edit: If I give the FSM 12V and ground, could the ground Isense test be used on the bench?
     
    Last edited: Dec 13, 2013
  14. alec_t

    alec_t Well-Known Member Most Helpful Member

    Joined:
    Jul 10, 2011
    Messages:
    9,255
    Likes:
    1,218
    Location:
    Cardiff, Wales
    That's because the fat cap C5 takes time to discharge via the LED.
    As I mentioned previously, Rsense in the PDM is key to PDM and FSM operation. It's only 0.22 Ohms, so even a few milliohms extra resistance (e.g. due to solder joints and wiring lengths) can be significant. On top of that there are component tolerances (10%?), pump current profile (different from the old pumps, and it's unknown exactly what the new controllers do to run the pumps up to speed) and pump build tolerances conspiring against you; so it's not too surprising.
    If the FSM is on the bench and disconnected from all else, give it 12V on both the '12V' terminal and the Vt input. Then connect the Isense input to ground ('Fault' output should rise to ~12V in ~4 sec) or to a +1.2V or +1.5V source ('Fault' output should drop to <~1V).
     
    Last edited: Dec 14, 2013
  15. salty joe

    salty joe Active Member

    Joined:
    Dec 30, 2011
    Messages:
    838
    Likes:
    28
    Thanks for the FSM bench test. All four chanels behaved exactly the same. After about 3 seconds, fault rose to 10.44V. The wall wart puts out 12.05V.
    I also added a 22K between the alarm module transistor base and ground. Thanks for that.

    Last night when I did the ground Isense test on channel 13 with everything hooked up I got no alarm. I'm pretty sure there was good contact, IDK, I was holding a wire to the Isense and ground terminal points. I did get an alarm at channel 14.

    I replaced the fault, V+ and Isense wires between PDM 13 and FSM channel 13 before last night's Isense ground test. No help there.
    I also checked all fault, Isense and V+ connections on PDM 13 for resistance as well as visually-everything looked good.
    For now, the FSM is still on the bench. Any ideas would be greatly appreciated.
     
  16. alec_t

    alec_t Well-Known Member Most Helpful Member

    Joined:
    Jul 10, 2011
    Messages:
    9,255
    Likes:
    1,218
    Location:
    Cardiff, Wales
    Well the FSM looks healthy then.
    I've lost track of what's where :). Is channel 13 in the wave system or in the tidal system? Did it have its Vt input high (12V) during the test? (If Vt is low the FSM channel is disabled by design). I should have asked that question in post #2351 :rolleyes:.
     
  17. salty joe

    salty joe Active Member

    Joined:
    Dec 30, 2011
    Messages:
    838
    Likes:
    28
    These last four PDMs and new pumps are part of the tidal system. PDM 11 & 12 share a common V+ and 13 & 14 also share a common V+.

    Yes, last night I made sure the other two tidal PDMs (7 & 8) that run with PDMs 13 & 14 had their LED lit up when I tested. I can re-test to confirm.
     
  18. salty joe

    salty joe Active Member

    Joined:
    Dec 30, 2011
    Messages:
    838
    Likes:
    28
    Since the FSM seems good, I hooked it back up and ran the Isense ground test again while the pumps were running. The alarm sounded for channels 11, 12 & 14 but not for channel 13. I double checked the fault, Isense and V+ connections between PDM 13 and FSM channel 13.
     
  19. alec_t

    alec_t Well-Known Member Most Helpful Member

    Joined:
    Jul 10, 2011
    Messages:
    9,255
    Likes:
    1,218
    Location:
    Cardiff, Wales
    Hmmm. Now it's all hooked up again, check the voltages in PDM 13 at the D10/D11 junction, and at both input pins (12,13) of U1d, both with and without the Ch 13 Isense terminal grounded.
     
  20. salty joe

    salty joe Active Member

    Joined:
    Dec 30, 2011
    Messages:
    838
    Likes:
    28
    The anodes of D10/D11 showed 10.84V and sounded the alarm when the DMM probe was touched, both with channel 13 Isense grounded and ungrounded.
    Pin 13 showed 11.85V both with channel 13 Isense grounded and ungrounded.
    Pin 12 showed 0.82V with Isense ungrounded and 2.33V with channel 13 Isense grounded.
    Thanks.
     
  21. alec_t

    alec_t Well-Known Member Most Helpful Member

    Joined:
    Jul 10, 2011
    Messages:
    9,255
    Likes:
    1,218
    Location:
    Cardiff, Wales
    Ok. Now we're getting somewhere. Pin 12 isn't doing its thing. Which of pins 12/13 is Vt and which is Fault'?
     

Share This Page