Help with Water Pump

[salty joe]

I apologize for not looking at post 1915 closely. I thought it was the same as the original schematic for blending the new pumps and new controllers and new PS with our controller. I missed the diodes-sorry about that. I have the diodes hooked up incorrectly. I'll fix it. Thanks.

On second thought, I do have the diodes hooked up as shown in post 1915. I lined eight 1N4148 diodes up and soldered a bare wire across the cathodes and sent that to 24V in our controller. Each PS is connected to its own dedicated anode.

 

[salty joe]

The PDMs have separate 24V and power ground lines. They share signal ground, 12V, and Vt lines.

With everything hooked up using the original schematic for blending new pumps, new controllers and new power supplies with our controller (no PS backup, 24V from one PS to our controller) each pump seems to run on its own PS. If the PS that does not have 24V connected to our controller is unplugged, only that pump stops. If the PS that has 24V connected to our controller is unplugged, both pumps stop.
The only change made was to take 24V from each PS and connect them to individual anodes of 1N4148, then running a wire from the connected cathodes to our controller. (as in post 1915) Then, as soon as a single PS is connected to either PDM, both pumps run.
I have been working with two new PS, two new pumps, two new controllers, two PDMs and our controller.
Thanks for looking at this.

 

[KeepItSimpleStupid]

@Joe

Where is +12 coming from in both cases?

Put it another way: What is the value of Vt when the pumps are stopped but supposed to be running?

--

Part of the reason I haven't been trying to answer lately is that I have been having trouble following.

 

[salty joe]

The 12V is coming out of our controller.

The pumps have not been stopped when they are supposed to be running,

 

[KeepItSimpleStupid]

@Joe

If the PS that does not have 24V connected to our controller is unplugged, only that pump stops. If the PS that has 24V connected to our controller is unplugged, both pumps stop.

If the PS that has 24V connected to our controller is unplugged, both pumps stop

without the 12 V on Vt, the pump should stop, right?

So, I'm asking what is Vt above?

 

[salty joe]

https://www.electro-tech-online.com/attachments/tidalsystem-gif.71768/

KISS, this is the original schematic for blending new pumps, new controllers and new power supplies with our controller and PDMs. When the PS that has 24V going to the controller is unplugged, the controller has no power and all pumps stop. I'm concerned that failure in that particular PS shuts down all pumps. We are trying to overcome this with the PS backup. So, in this original schematic everything works as expected. I'm not sure I understand your question, but I'd have to assume the Vt would be 0V at that point.

 

[alec_t]

With everything hooked up using the original schematic for blending new pumps, new controllers and new power supplies with our controller (no PS backup, 24V from one PS to our controller) each pump seems to run on its own PS. If the PS that does not have 24V connected to our controller is unplugged, only that pump stops. If the PS that has 24V connected to our controller is unplugged, both pumps stop.

So far so good

The only change made was to take 24V from each PS and connect them to individual anodes of 1N4148, then running a wire from the connected cathodes to our controller. (as in post 1915)

Ok so far.

Then, as soon as a single PS is connected to either PDM, both pumps run.

That shouldn't happen; only one pump should run. Seems there could be a stray connection between the 24V+ inputs of the two PDMs. Check the wiring in that area. Check also that all the power grounds are connected together at a common point, i.e at the power ground terminal of our controller.

 

[salty joe]

I think I found the problem. I disconnected a PS 24V from anode number 1 and there is 24V coming out of that anode! (PS hooked to anode 2) I checked the other anodes and there was also 24V coming out of anode number 2. (PS hooked to anode 1) Those are the anodes that I’ve been working with. Anodes 6-8 showed 0V, except for one that has 2.7V.

I hooked up the two PS 24V lines to anodes that had no current and everything seemed to work as planned. Truly a thing of beauty.

So now I wonder if I got defective diodes. I don’t think I cooked them to death, but I can’t rule it out. Or maybe we need heavier duty diodes. There’s no solder slopped around-it’s neat and tight-so I don’t think that’s an issue. I’ll be glad to post pictures if you want.
Thanks much, your thoughts very much appreciated.

 

[KeepItSimpleStupid]

Well, you do have that diode test on your meter.

Testing the end of a 24 VDC wire with and without a diode, you might expect a different result. The fact is, when you draw only a tiny bit of current from a diode, the diode acts more like a resistor. In order for a diode to show an appreciable drop, you have to be flowing some current through the diode. A few uA might look like a resistor, whereas a few mA will show a drop.

If there is current flowing through a diode, you can measure the voltage across it.

 

[salty joe]

I got those first diode V measurments by grounding the ground probe on my DMM and touching an anode with the other probe.

It did not even occur to me to use the diode tester-thanks. Diodes 1 and 2 registered 001 and 003 respectively. Diode 5 registered 1. I guess those three doides are toast. The other 5 diodes were anywhere between 618 and 641.

 

[KeepItSimpleStupid]

Yep - toast. Don't forget to reverse the leads and test as well.

 

[alec_t]

Sounds as though a duff diode was the problem. Apart from an initial brief surge at power-up those diodes shouldn't be passing more than a few tens of mA to feed the 12V regulator. They're rated at 200mA, so I don't think beefier ones are needed.

 

[salty joe]

What if everything is wired up except no PS is plugged in. As soon as a single PS is plugged into a PDM, the diodes with no PS connected are between 24V and the pump. Is it possible at that point the diodes are overwhelmed and give the pump what it wants?

 

[alec_t]

As soon as a single PS is plugged into a PDM, the diodes with no PS connected are between 24V and the pump.

They shouldn't be (at least not directly); they should be between 24V at our controller input and the 24V inputs of the PDMs. And apart from the one connected to the PS they should be reverse-biased (so non-conducting) in that situation. 24V is well within their 100V reverse-voltage rating. If a diode does break down and go short-circuit then yes, the pump would incorrectly get ~24V (via its OEM controller).

Edit: Double-check the diode polarities are as they should be.

 

[ronv]

Is it possible that the 24 volts from one of the controllers is also connected to one of the PDMs?

 

[salty joe]

They shouldn't be (at least not directly); they should be between 24V at our controller input and the 24V inputs of the PDMs. And apart from the one connected to the PS they should be reverse-biased (so non-conducting) in that situation. 24V is well within their 100V reverse-voltage rating. If a diode does break down and go short-circuit then yes, the pump would incorrectly get ~24V (via its OEM controller).

Edit: Double-check the diode polarities are as they should be.

You put that much better than I did. Thanks for putting my mind at ease in terms of what these little diodes can handle. Polarity is good, thanks.

Is it possible that the 24 volts from one of the controllers is also connected to one of the PDMs?

The new controllers are connected to the pump+ and pump- of the PDM. Our controller and the PDM get 24V from the new PS. 24V goes through a diode before getting to our controller, as shown in post 1915.
I'll replace the three bad diodes and fire up all 8 new pumps.
Thanks.

 

[alec_t]

I'd add in the pumps one at a time rather than all 8 at once, checking at each stage that things run ok.

 

[salty joe]

Will do-thanks.
I replaced the 3 bad diodes and will test run tomorrow.
3 bad diodes out of 8, that's a lot-I must have cooked them.

 

[alec_t]

Check the diodes again in situ, using the diode test function of your DMM, before powering-up the system (just in case you were sold a batch of dodgy diodes).

 

[salty joe]

Diodes look good-thanks.

For the revised wave, am I correct to assume that each PDM will need a seperate Vt? My PDMs are two to a board and all PDM pairs have shared Vt. Minor surgery, I'm just making sure it's required beforehand.

https://www.electro-tech-online.com/attachments/wavecontroller_pairedpumps-gif.72571/