Help with Water Pump

[KeepItSimpleStupid]

Water inside the IC. Nope. A bad solder joint you disturbed, yes.

FWIW, if you rinse with alcohol, methanol or IPA (Isopropryl alcohol), it will evaporate very quickly and displace the water. Methanol does a little better at cleaning than IPA. Acetone (nail polish remover) will also work for flux.

 

[alec_t]

The cathodes of D4/D10 are connected to pins 2 and 4 of U1a

?? Pins 2 and 4 belong to different gates, so can't both be U1a. What do pins 1 and 3 connect to?
I don't see how the FSM can give the symptoms you describe, and I would have thought any water on the pcb would have evaporated long ago.
When the 'Trip test' button is pressed and the alarm sounds D4 should pull R16 low to keep the FET off for ~30sec. Perhaps there's a dry joint at the D4/R16 junction or at the D4/U1 junction?

 

[salty joe]

IDK about the 'like-unlike' button. I like it so I push the 'like' button and it says unlike. Does that mean I don't like it? Anyway, I like all the posts and very much appreciate all your input.

Sorry about the U1a thing.
U1a is 4-5, 6. U1b is 3-1, 2.
Pin 1 connects to R8/Q2.
Pin 3 connects to pins 8 and 9.
I'm working from the PDM-Mk15 schematic. I found it with a search, but the link was no good, and I can't seem to copy and paste from my documents. The anode of D4 is connected to D5/R1/R2. The R16 I have is in the snubber between the pump leads.

I'm going to pull the PDM from the array and re-flow the D4/D10/pin2/pin4 junction. Worst case this goes nowhere and 1 pump is unprotected. Fingers crossed and thanks.

 

[KeepItSimpleStupid]

Like/unlike probably means you can change your mind.

 

[alec_t]

The R16 I have is in the snubber between the pump leads.

My mistake; I meant R15 which, in the latest mod of the PDM, is a 2k2 which goes between D4 anode and the R1/R2 junction. IIRC R15 was added to address a stability issue, but may not be needed since the SpeedCon input is not now used.

Edit: I'll keep out of the like/dislike debate, except to say that I think it's too Facebook-like and seems very childish.

 

[salty joe]

Well then I put R15 in the wrong spot. I have it installed between the cathode of D5 and Q1/R7. I did not mention R15 because as I have it, it's not connected to D4. I'm virtually certain that every PDM is the same. Anyway, that PDM is pulled loose from the array, if I'm lucky I'll have some time tomorrow to work on it. Thanks for helping to get this little bug worked out.

It takes a fair amount of time to remove/replace the pair of PDMs from the array. Would it be feasable to use a 12V 1A wall wart to feed 12V to V+ and the 12V line, and connecting the negative line from the wall wart to power ground and signal ground and running that to a mains ground? The only thing we need to see is the pump and pump LED turn off for 30 sec when the trip test is pushed. Make sense?

 

[alec_t]

I have it installed between the cathode of D5 and Q1/R7.

Ah. I suspect we're not singing from the same hymn sheet. I think your schematic is slightly earlier than the one I was looking at. But hey, if the other PDMs work leave them as is.

Would it be feasable to use a 12V 1A wall wart to feed 12V to V+ and the 12V line, and connecting the negative line from the wall wart to power ground and signal ground and running that to a mains ground?

If you mean just for testing the PDM on its own (+pump) and not connected (other than through ground) to the Wave or Tidal system (which has the 7812 regulator) then there would be no power supply conflict.

 

[alec_t]

Before you use the wall wart, make sure it outputs no more than 12V (some poorly regulated ones output way above their rated voltage). You could probably use a 9V one.

 

[salty joe]

Thanks for the heads up on the wall wart. Yes, I meant to use the wall wart to test the isolated PDM. I'll get that set up. Thanks again for the heads up. What a drag it'd be to fry stuff at this point.

 

[salty joe]

I re-flowed the D4, D10, pin 2, pin 4 junction-no help there. D4 and D10 both check out good.
I did not re-flow at the 4093. When I installed the IC, I inserted 20g, might have been 22g wire into the holes where the pins came out and soldered by holding the iron against one side of the wire/pin and feeding solder to the other side for solid neat joints. (Thanks KISS) I left all the wires long and hooked them up as I went. It'd be almost impossible to get in there now without re-doing the entire IC.I checked to make sure there was no continuity between pins 1 & 2, 2 & 3, 3 & 4 and 4 & 5.
I double checked that pin 1 goes to Q2 (pin 3)/R8 (33K) junction and the other side of R8 goes to signal ground.

I connected the temp 12V rig to the other PDM on that board and the trip test works.

Any ideas where to look next?

 

[alec_t]

Check C3 has good connections. Check R14 isn't shorted.
I'll post my version of the PDM schematic to ensure we're looking at the same thing.

 

[ronv]

Maybe you could measure the voltage on both sides of D4 before and after you push the trip.

 

[alec_t]

What is the voltage on C3 before you push the trip?

 

[salty joe]

Yes, that's the schematic I'm working from, with a few differences.
R15 is between D5 and the R7/Q1 junction. The unused SpeedCon is also at that junction.
A snubber is installed between the pump leads.
C3 is a radial cap.

I can't find any problems at C3 or R14. There's virtually no resistance between the positive leg of C3 and pins 5 & 6, R10, cathodes of D9 & D8, R14 and anode of D7.
With pump running, I got 11.65V at the anode and 12.05V at the cathode of D4.
Holding the trip button down, I got 1.24V at the anode and 10.81V at the cathode.
Thanks guys.

Running back downstairs to check C3.

 

[alec_t]

There's virtually no resistance between the positive leg of C3 and pins 5 & 6, R10, cathodes of D9 & D8, R14 and anode of D7. With pump running, I got 11.65V at the anode and 12.05V at the cathode of D4.

All as it should be there.

Holding the trip button down, I got 1.24V at the anode and 10.81V at the cathode.

Not right; should be about 0.6 at the cathode. What is C3 voltage about 1 sec after the trip is pressed?

 

[salty joe]

With the pump running C3 has about 7.80V. When the trip is held down, it shows 10.45V. When the trip is released, the V drops gradually to just below 7V then cimbs a little to 7.80 or so.

This time with the pump running the V at C3 dropped to below 6V and about 1 sec after the trip is pressed there was about 8V. That's holding the trip down.

 

[alec_t]

the V at C3 dropped to below 6V and about 1 sec after the trip is pressed there was about 8V.

That's what I'd expect.

With the pump running C3 has about 7.80V.

Hmm. Shouldn't be that high. That voltage ought to drive U1a output (pin 4) low and sound the alarm/switch off the pump. Can you re-flow the joints at U1a pins 5 and 6?

 

[salty joe]

Re-flowed pins 5 & 6. Rats, no change. Re-flowed the entire D8/R10/D7/D9/R14/C3 junction too.
I'll check U1a now.

 

[alec_t]

Can you confirm how U1a output changes (if at all) when the trip test is/isn't pressed for at least 1 sec?

 

[salty joe]

U1a is 4-5,6.
With pump running, 12.04V at pin 4. With the trip held down after about 1 1/2 or 2 sec. V drops to 10.73. When the trip is released, V is 11.23 after a sec or two. Then V climbs for another 10-13 sec to 12.04.