Variable speed drive for asynchronous motor

[Macro]

on a small pump, the motor may not be asynchronous but what is called a universal motor with coals. in this case, the voltage variation lowers the speed. that's what there is on all portable drills.

On the heating circulators they are indeed asynchronous motors ... But their power is ridiculous and very often they are almost useless ... They are simple centrifugal circulation accelerators ...
The simple fact of clamping them at the outlet (old method of regulating heaters) gives exactly the same result. A cenrifuge pump which turns on a closed tap consumes almost no electricity the consumption increases with the flow of circulation.

[Macro]

I would like to vary the speed of our solar pump in order to optimize the operation of the system.

The pump is a single-phase 650W guinard pump.

What exists as an apparatus to vary the power of such an engine and at what cost (about?). Thank you

If it is a centrifugal its consumption is proportional to the flow that you ask. Bridle the exit and the consumption will drop ...

[Christophe]

Yes the clamping is what had been advised by André I do not know in what subject ... On the other hand he said that it was within a fairly reduced margin (-20% ...). That's why I never tried.

Clamping the fluid is similar to increasing the pressure drops: there is not so much less electrical power consumed when the friction is increased ...

Running a centrifugal pump (not a circulator which is intended for) whose outlet would be completely closed is not very advisable ... it seems to me ...

In this extreme case: the power is not zero but the efficiency is ...

[Macro]

On the industrial installations on which I work the pumps have calculated to have a minimum pressure at the several outlet points and effectively we calculate the pumps so that the pressure rises to the discharge of about 20% when we are at minimum. But on a 250 m3 / h pump we oscillate with an output flow between 15 and 240 all day.

Here I am going to measure the currents and measure the results ...

[Macro]

On centrifugal pump with asynchronous 400V tri 55KW motor
Max flow 250m3 / h Hmt 63m

She
At start-up 200 (limited by altistar starter)
has flow 0 consumption on a 46 amp phase
at 15m3 / h 51A
at 120 66A
at 240 97A

[Christophe]

Great for measurements!

The flows are regulated how? Just on the downstream pressure drop as you said above? Is the mano height constant for all flow rates?

At 0 flow (valve closed?) We therefore consume roughly 50% of the nominal power ... for nothing ... zero efficiency. It must heat up well

But we also see that the performance of this clamping method is not very good: passing from 15 to 240 m3 / h or 16 times more fluid power, we consume electrically only 2 times more, ratio of 8!

Knowing that at 240 m3 / h, nominal power, the efficiency is probably around 60 to 70% I let you calculate the efficiency at 15 m3 with this clamping method!

It should heat the pump body, the fluid, the pipes, the valves, the motor ...

However in my case it is precisely for the purpose of optimizing the yield, so exit this method! We may consume less but we waste a lot more!

Otherwise, I also made measurements on my solar pump! It works better than I thought. I'm writing the record.

I feel stupid enough not to have tested earlier ... and above all I don't understand that I didn't know it ... (never learned)!

[Christophe]

Here are mine!

Measurements made on the solar pump, following conditions:

- self-priming pump Guinard Kiétis 4000 650W

- in stabilized hydraulic load (ie when the water is raised and lowered = primed hydraulic circuit, not to be confused with primed pump). It takes at least 1.2 bar to prime.

- for each supply voltage I made a measurement of the mBar at the pump output (approximate because the mano has already lived and is not very precise, but it gives an idea)

We have this:



It seems consistent!

In terms of performance:

- The speed variation is very noticeable (unlike the vacuum engine tested this morning, see: https://www.econologie.com/forums/abaisser-l ... 11126.html )

- The circuit self-primers (i.e. it needs to go up to 1.2 bars) up to 140 V AC, below it does not happen anymore but the pump always manages to self-prime ( up to 100V and even 90V).

- At 140V, the consumption in rising water column is 560W and it takes about (to be measured) twice as long as at 2V / 230 W, so I don't know if it is interesting to stay in 650V on this phase! I will do measurements later. The pumping efficiency seems clearly less good on this phase of pumping at 140V than at 140V!

- The pump starts to make a "squealing" from 100V, at 80V the circuit is deactivated, even when the pump is still running (power supplied to the fluid insufficient, the consumption of the pump is then 210W, hydraulic efficiency zero).

- At 70V the pump stops.

- Without pressure, it restarts slowly between 80 and 90V (I did not stay long so low: I do not like to put watts, therefore calories, in something that does not turn!)

- For the moment 110V seems to be the best compromise: the noise is very low (it is very very impressive at the pump room! Limit we hear more! It would be interesting to measure with a sound level meter), the circuit does not defuse, the pressure is still at 300 mBars ... remains to be seen the pumping efficiency ... and therefore if the flow is sufficient! (to see by measuring the T ° and estimate a COP ... if I get there)

The following ?

- Here I set the rheostat to 140V: this is the minimum voltage for self-priming of the circuit. Consumption in steady state is 440W (against 650W at 230V). I will leave for a few days (it will be nice this weekend) to see how it heats!

- I have 2 solutions for the rest:

1) leave on 140V and finalize the assembly, either with the rheostat, or with an AC divider bridge (does it exist?)
2) if energetically and thermally at 110V it is ok then switch to 110V in primed mode + higher speed for priming (voltage to be defined) with a time relay.

- So it remains to see the "thermal" performance but I think they will be good since it is recommended a pumping power of 3 to 5W per m² of solar panel in drain and as I have 65 m² of surface (therefore power between 200 and 330 W) ... remains the unknown of the performance of the pump under speed (surely not very good) ...

- Subsidiary question: is there not a risk that the air cooling of the engine will be insufficient given the drop in RPM (the RPMs, therefore the air flow rate and therefore the cooling power, decrease faster than the electrical power consumed so the power to cool ...)?

To be continued!

ps: I'm still blushed that we can vary the power of an asynchronous motor in this way ... there is surely a hidden defect in the method

[Macro]

Do you measure your consumption at the input of the autotransfo ???

[Christophe]

Yes with a plug econometer (EM230) which only works at 220-230V AC of course, I suspect there are some losses in the transformer. When empty it takes about 10W.

I could measure the amperage that arrives at the engine.

I suppose the problem of the loss of hydraulic efficiency at reduced RPM is more important.

[Macro]

Not on ... I have pumps that were mounted at 1250 rpm or 2450 ...

I think this technical solution is not used for a simple cost story ...

It seems to me that the electrotec had this kind of varios in the workshop next door when I was in high school they were huge ...

On our side we did much the same with a CI (from memory a TCA something) and triacs ... For a rudimentary cost and ridiculous bulk ...

But the best is to have a pump adapted to the required use. If you really have two very different flow rates to apply, a large one to fill your installation, a small one to circulate ...