Help with Water Pump

[4pyros]

Why not use an of the shelf charger

 

[alec_t]

Not knowing the exact 12 volt current draw the battery voltage could try to go to 20 volts

I haven't totted it up but I reckon average current draw would be ~ 50mA total. The idea was that a battery would be charged elsewhere (with an off-the-shelf charger) and the trickle charge would just make up for self-discharge. In the event of power failure then the battery would need charging fully again when power was restored. I was just kicking the idea around to see how simple we could make it. I agree ronv's charger would be better.

 

[salty joe]

(Joe, don't build it yet ).

That made me laugh out loud!
I did not get any notices-you guys have been busy. Wow, thanks.
It's good to see the features on this site coming back to life.
Happy New Year!

 

[KeepItSimpleStupid]

Who knows what the notice algorithm is, but unless you make a periodic visit, the notices stop. It's one of the problems with putting the actual messages in the notice. There is no reason to visit the site sometimes.

Read the bottom of the notices carefully: There MAY also be other replies, but you WILL NOT receive any more notifications until you visit the forum again.

See, it's iffy by design.

 

[salty joe]

I just ran 10.4V through the controller and two PDMs with two pumps hooked up. Both pumps toggled reliably.

I am having a little trouble with some of the terminal points though. Some of them that I have screwed down a lot of times are twisting in the solder joint. I wonder if I should have cleaned up the terminal point posts before soldering? Or maybe clean the posts, insert them, tin them and then solder. Any advice would be appreciated. Thanks.

 

[KeepItSimpleStupid]

They should not do that. I'm going to suspect a cold solder joint.

Unsolder and clean up the pins.

I'd have the screw loose so as to minimize the thermal conductivity. Attach the wire. Heat the terminal (larger mass) and apply the solder, let cool and try to move the wire.

 

[salty joe]

Thanks, that's what I'll do. Maybe using a larger soldering tip would be helpful too.

 

[KeepItSimpleStupid]

Joe:

If your not doing it already, when you solder a lead to a PCB Pad, use the soldering iron and push the pin to edge. Then apply solder to the "Other side" of the pin, opposite the iron. So it's iron - pin - solder.

 

[salty joe]

I think you nailed it KISS-cold joint at the terminal points. Bad to me. It was not too awful to fix. I put the blunt tip on the soldering iron, cranked the heat up and held the tip to the end/bottom of the terminal point posts until the solder reflowed. Did not take long. Only a couple terminal points failed, but I re-flowed all of them. I think we're in good shape now.
Thanks a bunch.

 

[alec_t]

I just ran 10.4V through the controller and two PDMs with two pumps hooked up. Both pumps toggled reliably.

Good. Was one of those pumps a 'fussy' one?
Sim shows the PDM should cope down to 9V or so without unwanted trip operation.
Glad the re-soldering went ok.

 

[salty joe]

Yes, I used a fussy pump. And thanks.

 

[KeepItSimpleStupid]

I think what alec was getting at is: Do you think that the cold solder joints are the reason for the pumps being fussy?

 

[salty joe]

No, I don't think so. Certain pumps would never toggle more than a few times, mostly ony once. Pumps A and two others were always problematic until I installed the 2k2 resistor. The other three pumps never were a problem. But I think they were probably a problem waiting to happen. The second PDM had the exact problem with the exact pumps until I installed that 2K2 resistor. I'm not sure any terminal posts were loose enough to twist in the solder before adding the 2K2. Plus when I had cold joint trouble, the LEDs did not light up. When the pumps did not run, the LEDs lit up and the pump would hum. I could plug in any fussy pump and have trouble or any other pump and it would run. I know I can't prove it but there is no doubt in my mind that the 2k2 solved a big problem. Maybe having to make a fix for the fussy pumps was a blessing in disguise.

 

[KeepItSimpleStupid]

Happy Belated Anniversary Joe!​

On December 30, 2011, your post started this thread. It's a little over a year old now.

 

[Electroenthusiast]

Happy Belated Anniversary Joe!​

On December 30, 2011, your post started this thread. It's a little over a year old now.

Biggest thread i have ever seen on anywhere. Seems interesting to see whats throughout.

 

[DerStrom8]

Biggest thread i have ever seen on anywhere. Seems interesting to see whats throughout.

AAC has one that's well over 2000 posts long. But this is definitely the longest TECHNICAL thread I've seen anywhere, ESPECIALLY that's stayed on topic

Happy anniversary Joe!

 

[salty joe]

Thanks guys, ha ha. I really appreciate so many amazing people willing to see this through.

 

[alec_t]

Time to crack open that bottle of digital champagne, Joe . Cheers!

I think what alec was getting at is: Do you think that the cold solder joints are the reason for the pumps being fussy?

Actually I was just hoping Joe had checked that even a 'fussy' pump would start reliably at ~10V, so that using a 12V backup battery rather than 24V would be possible.

Edit: Joe, can you remind us of the continuous current ratings of your two 24V power supplies? I'm thinking one of them might also power the charging of the backup battery.
It may also be possible to mod one of the OEM controllers to act as a charge controller for the battery; or would you prefer to go with Ronv's suggested LM317 charge controller?

 

[salty joe]

The power supplies are 19-24V and 20-30V. They are both 15A.

I think I'd like to try Ronv's creation. Thanks all you guys.

 

[ronv]

You will need a pretty good sized battery if it needs to run 2 pumps for 24 hours. Maybe a 30 40 amp hour sealed lead acid.
Let us know if you have any questions.