Help with Water Pump

[alec_t]

While you're doing the fan test at 24V, use your DMM on the current range (the 10A range if you have one), in series with the fan, to measure the average fan current for future reference. I'm guessing it would be ~ 40-60mA for a 24V fan model.

 

[salty joe]

The fan ran for an hour at 24V and got the slightest bit warm. Just a shade above room temp.
With the DMM set at 10A and one probe on 24V+ and the other probe on the fan's +wire, the fan ran and the DMM registered 0.09.

The fan driver and second 'on backup' warning circuit are done. I hope to hook them up tomorrow.
Thanks.

 

[alec_t]

Looks like the fan is ok on 24V then, drawing ~90mA. So it can be run from +Vext, and if you're using the FET it won't need a heatsink on the FET itself.

 

[salty joe]

I finally got a chance to hook up the new circuits. The individual ‘on battery’ red LEDs for each old PS work perfectly. Thanks for giving me separate warning lights for the filter pumps’ power supplies. It’s a beautiful thing.

The fan driver and charging system seem to be giving me a little bit of a hard time. The green ‘charging’ and ‘fan running’ LEDs stay on and the fan runs even when the system is on battery backup.
On the plus side, the fan keeps the LM317 very cool.

Pin 13 of LM324 adjusts from 15.09-15.72V. Pin 12 is 15.77V. (With R1 jumped and battery disconnected) IIRC, we need pin 13 – pin 12 = 50mv.
One change is the addition of D8 and D9 in the filter pump power backup (FPPB-Mk3). I don’t know how I missed installing them in the first place.

I connected the power ground from the fan driver to the ground from one of the ‘on battery backup’ circuits, instead of running two separate grounds. Super easy to fix if it matters.

I could not find any errors in the fan driver circuit. Any ideas on where to look?
Thanks guys!

 

[alec_t]

Pin 13 of LM324 adjusts from 15.09-15.72V. Pin 12 is 15.77V. (With R1 jumped and battery disconnected) IIRC, we need pin 13 – pin 12 = 50mv.

Correct. At the moment pin 12 is > pin13 (i.e 'wrong way round'). Ifyou can't adjust it as required the fixed resistor which you added in series with the trimmer may need to be moved to the 'other end' of the trimmer.

 

[salty joe]

https://www.electro-tech-online.com/attachments/fppb-mod_detail-gif.75294/

Thanks for pointing me to the trimmer.

The trimmer has two legs on one side and the third leg is offset on the other side. Am I correct to assume the offset leg is the wiper?

I have the offset leg connected to a 1K5 and 33K resistors in series going to ground.
Of the other two legs, one is connected to pin 13 of LM324.
The third leg is connected to a 10K resistor that goes to VinU1.
Should I switch the connections so that the offset leg is connected to pin 13 of LM324?

OK, now I think I understand what you mean. Remove the 1K5 that is in series with the 33K and install the 1K5 in series with the 10K. Is that right?

 

[alec_t]

Am I correct to assume the offset leg is the wiper?

Don't know for sure. Probably. You can check by connecting your DMM (on the appropriate Ohms range, circuit power off) between the suspected wiper leg (the offset one) and each of the other legs in turn. If both readings are ~1k (with the trimmer set to its mid-point) then the offset leg is the wiper. That should connect directly to pin 13 of the LM324.

Of the other two legs, one is connected to pin 13 of LM324.

If that 'other' leg is not the wiper then it should go to the 1k5/33k resistor string.

Remove the 1K5 that is in series with the 33K and install the 1K5 in series with the 10K. Is that right?

Yes, if you can't get the adjustment you need with the connections I've just described. You may not even need the 1k5 (it all depends on manufacturing tolerances).

 

[salty joe]

Sure enough, the offset leg is the wiper-thanks for telling me how to find it. Connecting the wiper to pin 13 helped a lot, but not quite enough.
Pin 13 adjusted from 15.17-16.02V. Pin 12 is 15.76V.

Jumping the 1k5 that is in series with the 33k pushed us in the wrong direction. Now, pin 13 adjusts from 15.00-15.87.
I gotta hit the jobsite and start earning a living. Thanks!

 

[alec_t]

Jumping the 1k5 that is in series with the 33k pushed us in the wrong direction.

Ok. So something higher than 1k5 needs to replace it. Try 2k2 (I think that's in your kit).

 

[salty joe]

I temporarily hooked up a 2k2 and the max value at pin 13 was 16.10V, so I tried a 3k3 and got a max value of 16.19. With a 4k7 hooked up, I could get 16.26 at pin 13.
But the fan and green LEDs did not come on when the battery was charging. (LM317 got warm and there was12.90V at the battery) I adjusted the trimmer until the fan and lights came on. Then, with battery disconnected and R1 jumped pin 13 showed 15.83-15.89.
Your thoughts on this would be greatly appreciated.

 

[alec_t]

I adjusted the trimmer until the fan and lights came on. Then, with battery disconnected and R1 jumped pin 13 showed 15.83-15.89.

Good. If you've got 15.76V at pin 12 then ideally you need ~15.81V at pin 13.
So you now have the required adjustment range with the trimmer (actually you did.....just about... with the 1k5 back in post #2188). You can set that with finer resolution if you have the DMM +probe on pin 13 and the -probe on pin 12. Adjust for 50mV reading. Probably a bit fiddly. If the 50mV setting results in the fan not turning on when charging is in fact occurring that could be because the power supply voltage drops slightly under load, affecting the pin 12 voltage. We may then have to increase the setting to, say, 70mV or so.

 

[salty joe]

The charging system was trying to do its' thing but I was interfering...I had the volts off by a factor of ten. Note to myself-mill means the third place from the decimal point. Duh.

The charging system works beautifully, thanks!

Will this 5A fuse work at the old PSs? **broken link removed**

The other missing part is the 68k NTC. I have leads connected between R9 and terminal points.
I selected 68K, surface mount and cut tape. I'm unsure about the rest. Can you please have a look and let me know if there is a good choice for this project? **broken link removed**

So far, my computer gets along MUCH better with this site since the upgrade. Big difference. I like your dragon, Alec. Did you draw it?

 

[alec_t]

I had the volts off by a factor of ten

That would explain things

The charging system works beautifully

I'm pleased that's sorted.
The fuse looks fine.

Re the 68K NTC, I just restricted your search by specifying the 1206 package size (the largest) and got product ID 541-1148-1-ND (cost ~$1). So, a make-a-hairpin type assembly like you did with the resettable fuses in the PDMs?

So far, my computer gets along MUCH better with this site since the upgrade.

Guess we'll never know why. I'm slowly finding my way round the new site.

I like your dragon, Alec. Did you draw it?

Nah. Just a bit of clip art from somewhere. Seemed appropriate as the Welsh flag has a dragon (more like EricGibb's avatar....not sure if Eric has Welsh connections). Time you uploaded your 'Salty' avatar?

 

[alec_t]

....or why not use a (cheaper) through-hole part for the 68k NTC :-
https://www.digikey.co.uk/product-detail/en/NTCLE100E3683JB0/BC2552-ND/773992

 

[KeepItSimpleStupid]

The US doesn't seem to like thru hole thermisters. They don't seem to exist for the most part.

I've been trying to understand them recently and Digikey and Vishay and Murata threw me in a tizzy. Instead of ratios between R25/Rt, there is is B value something which I haven't looked at until this morning.

I suppose your only going to look at a narrow range of temperature from 25 C and therefore the slope at R25 doesn't really matter?

 

[alec_t]

Hi KISS. I don't think the B value is critical. As you say, the temperature range is narrow. The NTC is there to reduce the charge rate slightly if battery temperature increases, but it's guestimation as to what the reduction should be for a particular battery. Do you have any figures for that? I couldn't find any with a limited amount of googling.

 

[KeepItSimpleStupid]

What kind of battery will we be using? I just looked up any old 12 V 7.2 AH capacity battery and it said it's designed for trickle use and constant voltage charging. https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&ved=0CEkQFjAA&url=https://www.panasonic.com/industrial/includes/pdf/Panasonic_VRLA_LC-P127R2P.pdf&ei=kRgyUoulPJTI4AOJxICgDQ&usg=AFQjCNGCp0UBsciV8v1qZ2q1BeqzigV3LA&bvm=bv.52109249,d.dmg

This https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&cad=rja&ved=0CE0QFjAB&url=https://www.panasonic.com/industrial/includes/pdf/Panasonic_VRLA_Overview.pdf&ei=uRgyUqT7Lc-u4AOPpIG4CA&usg=AFQjCNE3NFgtggXHlo44pYBfF_g7FXFl0w&bvm=bv.52109249,d.dmg

deep discharge battery looks more ike CC-CV with an overt temperature cut out on discharge and I would consider that for charge too. There is some info where the capacity changes with temperature, but it looks more important to do low voltage and a high temp cut out for long life as well as the CC-CV charging.

It looks like "the battery" may matter.

Here is some more "general info" indirect coutesy of jpanhalt at another thread.

 

[ronv]

Here you go.

**broken link removed**

 

[alec_t]

Thanks Ron. The curve shows there's quite a tolerance in the compensation, but in general we're looking to drop the charge voltage ~0.3V for a 10degC change in battery temp around normal room temp. IIRC from simulation the NTC resistor chosen will shift it about half that amount; but at least it's a shift in the right direction.

 

[salty joe]

Thanks Alec for the 68K NTC and the OK on the fuse.

I probably should have figured that battery choice would not be absolutely straightforward...

The sticker on the 'old' filter pump says 1.3A, 30W, 24V. When they are running on the battery, would they only consume about half as much power? If so, would a 70ah battery run two old filter pumps for about 54 hours?

The battery will be in the basement, so the temp will be between 50-70F, 10-21C.

"Newer, more intelligent charging systems are now incorporating automatic temperature compensation. "(lifted from the article,thanks Ron) This really is a high powered system, not just the battery backup. I'll try to get it mounted in such a way to do it justice. Thanks you guys-you're the best!