Page 15 of 24

Re: home improvement turns into science fair projects?

Posted: Fri Feb 28, 2020 6:23 pm
by terkio
Sensing line cord current to restart, is fine if pump is wet, but could be damaging to motors running dry. It seems a brief couple second test every few hours should be well tolerated, but probably only of modest utility.
I am aware, it is bad to run a water pump dry, the user manual usually says so.
What is wrong actually ?
The only thing, I can see is cooling.
Water does cool the motor by the axle and housing. However, when running dry, the motor spins with little torque, hence draws little electric power.
These are two opposite effects about heating. The outcome is not clear to me.
No water has no effect on the motor bearings.
As far as I know, they run dry, whether the pump is pumping air or water.
May be water has an effect on the water seal at the shaft. May be this seal needs lubrication from a wanted tiny leak. I know there is an escape hole to make sure water doesn't go into the motor in case of a leak at the shaft seal.
So, running dry, may be is bad for the shaft seal. Recent pumps use ceramic seals.
Should we fear of the running dry syndrome ?

Re: home improvement turns into science fair projects?

Posted: Fri Feb 28, 2020 6:57 pm
by JR.
that's my official excuse and I'm sticking to it. :lol:

I expect it would be well tolerated, but for now easier to just worry about turning it off...

Next question is can I do this without a low voltage power supply...?

Still soaking this in beer, and I will need to make some measurements with current sensor after it arrives.

JR

Re: home improvement turns into science fair projects?

Posted: Tue Mar 03, 2020 11:38 am
by JR.
Current sensor arrived and testing under way...

It is a 1000:1 split core.... I have terminated it with 6.8 ohm apparently they prefer a dead short load and warn about operating it open circuit (i.e. do not use fuses).

I have it crudely wired into my sump pump power source. I definitely perceive a difference with my scope between pumping full water solid 60 mV p-p AC, and after beginning to pull air... it cycles down between 40-50 mV p-p... I hoped for a larger drop, and it is still moving some water, so I may run it longer until it is pulling only air...

With a micro, I could easily detect these two states, but for a KISS discrete approach I am looking for more definitive current drop to parse status.

Testing will continue... with water flow down to only a trickle it is till registering a pretty solid 40mV p-p. So a measurable difference but as much as I hoped for.

JR

Re: home improvement turns into science fair projects?

Posted: Tue Mar 03, 2020 12:26 pm
by terkio
I suggest you look at the phase.
Phase of I relative to V.
I presume the motor is an asynchronous motor, this is usually what is used for water pumps.
I remember the torque directly relates to speed slipping.
I do not remember wether torque relates to phase. Who knows, may be the phase gives some information about the torque.

Re: home improvement turns into science fair projects?

Posted: Tue Mar 03, 2020 1:55 pm
by JR.
terkio wrote: Tue Mar 03, 2020 12:26 pm I suggest you look at the phase.
Phase of I relative to V.
I presume the motor is an asynchronous motor, this is usually what is used for water pumps.
I remember the torque directly relates to speed slipping.
I do not remember wether torque relates to phase. Who knows, may be the phase gives some information about the torque.
Right now I am getting a floating transformer isolated current. To identify phase I would need to compare that current to applied voltage that is not isolated.

With a micro I could grab mains phase from an unregulated power transformer winding, in the power supply required to power the micro.

This does not feed into my KISS approach.

I guess I might be able to crudely parse relative phase by squaring up the mains voltage and comparing that to a squared up current output.

===
I am now suspicious that looking at the voltage across the current winding may not be accurate...(winding impedance is in parallel with load and who knows what else..?) It is intended to be used into a virtual earth.

Shunting a real VOM across the sense winding reveals a pretty well expected 1mA (=1A) while pumping mostly air... But now I no longer have enough water in my crawlspace to get full water flow testing...

I'll need to wait several hours and try again.

JR

Re: home improvement turns into science fair projects?

Posted: Wed Mar 04, 2020 10:41 am
by mediatechnology
How about a simple impeller-based flow sensor on the discharge hose to detect loss of flow?

I did a monitor for a friend that detected loss-of-flow with the tach output from the impeller retriggering a Velleman PIC-based software-controlled power relay.
The Velleman sensed loss of tach signal and after a short time out it alarmed. (One of its many operating modes.)
The whole gizmo was completely off-the-shelf.
No interface just flow sensor, relay board, wall wart and relay closure.
I think you could use the relay board where it would start on button press and turn off after tach signal loss.
It might take two of the Velleman boards but they were cheap.

Re: home improvement turns into science fair projects?

Posted: Wed Mar 04, 2020 11:28 am
by JR.
Thanx...
Image

I just ordered one of these ($27). Hall effect, 1.5" so fits my discharge hose diameter... we'll see, another angle of attack...

After the rains overnight (serious hail falling right now... makes a wonderful noise on my metal roof :lol: ) I have no shortage of water for my sump pump current testing, but the current transformer sensor result looks almost the same as no load...

I decided to go old school and put my VOM ammeter in series.... It has a 10A fuse so no worry. This morning with full water flow meter only registers 2.4A, so the current sensor was not very far off.... Now I need to wait until I pull air again to see how low the actual pump current drops.

This current transformer may still work, I just need more sensitivity... I could have used the smaller 5A transformer but my pump is specified for something like 9A.

===

Looking around I have some switching devices from my old time of day DIY heater controller project... A 16A three quadrant snubberless triac, and opto-fet to switch the triac on/off with opto-isolation.

I am thinking of basing this on my old first generation tuner platform that has 5V regulator and quad op amp.... No desire to throw a micro at this yet. I can use the op amp to process the current transformer output into something more manageable, ASSuming the dry current is low enough to reliably parse.

JR

Re: home improvement turns into science fair projects?

Posted: Wed Mar 04, 2020 11:32 am
by mediatechnology
Read my "Workbench" contribution in Radio World regarding the Velleman K8015 relay.
The delays can be as short as 2 seconds to as long as 12 days. Wayne says the K8015’s “blinking” mode is ideal for on-air warning lights. With the addition of a coupling capacitor on the trigger input, a tachometer signal can be used to sense loss of rotation or air flow.

https://www.radioworld.com/news-and-bus ... ems/337098

Re: home improvement turns into science fair projects?

Posted: Wed Mar 04, 2020 11:50 am
by JR.
mediatechnology wrote: Wed Mar 04, 2020 11:32 am Read my "Workbench" contribution in Radio World regarding the Velleman K8015 relay.
The delays can be as short as 2 seconds to as long as 12 days. Wayne says the K8015’s “blinking” mode is ideal for on-air warning lights. With the addition of a coupling capacitor on the trigger input, a tachometer signal can be used to sense loss of rotation or air flow.

https://www.radioworld.com/news-and-bus ... ems/337098
If I ever do get clever with micro control... I would adaptively adjust resampling time based on previous run time results...

I could program up the 8bit micro already on my first generation tuner board, but 8 bit programming hurts my head... I need to check the footprint I might be able to pop a newer generation pic onto the old footprint.

JR

Re: home improvement turns into science fair projects?

Posted: Wed Mar 04, 2020 12:32 pm
by JR.
The sump pump finally started pulling air... it does not run completely out of water, but doesn't pull water continuously, it cycles or surges intermittently. While the VOM current display is not stable, it jumps around between near full current 2.4A, and down as low as 1.4A.... I suspect a smoothed current readout could reliably identify this "almost dry" mode.

I could experiment with letting it run like this longer to see if it ever runs out of water? Cycling it off for a few hours when it makes noise, then on again later is quieter, but the motor is unlikely to overheat when pulling partial water... so maybe letting it run longer is not crazy.

I feel another experiment coming on....

JR