1-phase motor frequency control

[jmaja]

I would need to control the rpm of a 230 V 900 W 1-phase motor. I thought it would be straight forward, but it seems to be much more difficult than with a 3-phase motor.

What I have learned so far, it would be necessary to remove the starting and running capacitors. I haven't yet checked is that possible. It's a submerged water pump in a well. There is a separate control box, which may have the capacitors.

So can this motor be controlled at any 3-phase VFD? Or are there some special requirements for 1-phase motor capable VFD's?

I would need to control the water flow to 1/6-1/3 of the maximum with good efficiency. This most like means reducing the rpm by much less than that. Maybe just 50% reduction is enough?

[beanflying]

Far better you run a flow control valve with a centrifugal pump than have a poorly speed regulating abomination. Depending on your 'actual flow rate and system' there will be little to be saved with speed control over flow control. Depending on this system (head, pipe run etc) you may only need a small reduction in speed (possibly only 20-30%).

Three phase control is another concept and you can get decent turndown with an inverter.

The Curve below is just a generic 250W Submersible Drainage pump you will notice that as the flowrate drops (valve modulated option) so does the absorbed power.

The second curve is another generic showing a centrifugal pump curve with variable speed. Generally most submersible pumps don't run VFD drives so this is not based on anything other than to give an indication of how performance is changed by speed reduction.

[Berni]

Yeah if its a submersible pump you probably will not easily get at the starting capacitor. And even so single phase induction motors on VFDs made to run them are a bit iffy. They are still loud and inefficient and success partially depends on what motor it is.

But if its a centrifugal pump type then you could reduce the flow by mechanically restricting the output pipe. For centrifugal pumps the torque required to spin it is mostly dependent on the flow trough the pump, so the more you restrict the output the less flow you get and the easier the pump is to spin. Its the positive displacement pump types that can't handle this (like rotary vane, gear, piston pumps...) since restricting there output makes them harder to turn.

[jmaja]

Far better you run a flow control valve with a centrifugal pump than have a poorly speed regulating abomination.

I would only need 0.2-0.4 l/s and 11 m head. The pump delivers now 16 m and 1.1 l/s at 800 W. I don't have curves for this pump, but a much more expensive pump shows 0.95 l/s 17 m takes 528 W. 0.2 l/s 11 m would take only  49 W and 0.4 l/s 11 m 92 W.
https://product-selection.grundfos.com/fi/products/sqe/sqec-2/sqe-2-55-96510151?tab=variant-curves&pumpsystemid=1230921145

The efficiency of my pump is now about 21% at 1.1 l/s. The much more expensive Grundfos pump has 30% at 0.95 l/s, 47% at 0.4 l/s and 44% 0.2 l/s.

If I would just have a control valve on that Grundfos pump, it would take about 480 W at 0.2 l/s and 580 W at 0.4 l/s producing about 50 m extra pressure head.

This cheap pump will not likely be as efficient, but I had hoped I could get about 100 W at 0.2 l/s and 200 W at 0.4 l/s. Would this be possible with VFD.

[beanflying]

Explain what you are doing with the pump (rough vertical lift, pipe diameter and length or run)? I spent a large chunk of my 'real job' time playing with industrial pumps so maybe a better pump selection might be in order.

Also is the flow rate desired 'fixed' or do you need to vary it ? Generally a simple gate valve or even an 'oriface plate' is used so nothing electrical at all.

[jmaja]

This would pump to a heat exchanger, which is used for a ground source heat pump. This heat pump takes 2-10 kW energy from the pumped ground water depending on outdoor temperature. The water level is 10 m below surface, thus only a submerged pump (or an ejector pump) can be used. The well is 11 cm in diameter, thus only pumps designed for this purpose can fit.

The water temperature is 7 C and can be safely cooled 5 or maybe 6 C to 1-2 C. The flow control is needed to reduce the amount of water used and power needed. If the 800 W pump runs all year round, it is quite expensive. If it would run at 100-200 W, it would be OK.

Piping has 26 mm inner diameter and is about 30 m. Thus not a big deal at 0.2-0.4 l/s.

I haven't seen any pumps that would produce these numbers. Usually they produce 50+ m pressure head and take at least 400 W. I called Grundfos and the pump I linked was their best option, but it costs about 1500 € with their own controller, which is not very suitable for this purpose.

I would add a control system, which starts the pump when the heat pump starts and the controls the flow so that the return from the heat exchanger is safely above freezing. 0.2 l/s could be the minimum, although even less would be OK at minimum power.

[Berni]

The pump A won't magically get all the characteristics of pump B by simply running pump A at the speed of pump B.

The shape and dimensions of all the components in the pump are optimized for certain operating conditions. You obviously have a pump that is designed to run at fairly high powers. So even if you reduce its power by slowing it down or just restricting the flow rate it will still likely consume considerably more power than the other small pump that it optimized to run at those flow rates.

Also with centrifugal pump the speed mostly determines the head pressure, not the flow rate, the flowrate simply becomes whatever is possible with the given head pressure trough the restriction of the pipes and the pump itself. So getting a low flowrate out of it trough slowing it down might become a knife edge balancing act, one speed might not provide enough lift to get the water up there, slightly higher speed might make too much flow trough a very low restriction end.

Running a huge 5 liter V8 engine at 20hp output will also have crap efficiency compared to a little 1.2 liter economical inline 4 that is optimized for great efficiency there. But the V8 might also be able to produce 400hp when the little inline 4 simply can't.

[beanflying]

An offset Injector pump while an option is fairly inefficient and also noisy. For mechanical/cost reasons shaft drive is out at this small size. So back to 4" nominal Borehole Pumps as the best solution.

Ok quick calculations you need the 10M vertical and a bit over 1M more for pipe friction for 11 total. The 110mm bore being the main limiter.

So two options Mains single phase and throttle it (three phase will be prohibitive on cost) or likely a better option DC Motor and speed/power control it. Just after midnight here but I can have an ask around tomorrow locally but on a quick search there is plenty of options available in 24 or seems more likely 48V DC in 4" Borepumps. So have a search while I 

[jmaja]

The pump A won't magically get all the characteristics of pump B by simply running pump A at the speed of pump B.

Of course not, but these pumps are quite similar. The other one just has published curves also for reduced rpm. Now the pump is running at very low load, since it is designed to output up to 60 m head. All of these pumps have similar curves. Efficiency is very low at low high flowrate + low head and low flowrate + high head.

Changing rpm should move the optimum flowrate and head closer to my operating point. Flow control doesn't need to be that accurate. Main point is not to consume 800 W all the time. Also 1.1 l/s is much, since the water needs to be dumped somewhere. There seems to be enough water coming from the well. The water level didn't drop at all when I pumped one hour (4 m3).

[beanflying]

The pump A won't magically get all the characteristics of pump B by simply running pump A at the speed of pump B.

Of course not, but these pumps are quite similar. The other one just has published curves also for reduced rpm. Now the pump is running at very low load, since it is designed to output up to 60 m head. All of these pumps have similar curves. Efficiency is very low at low high flowrate + low head and low flowrate + high head.

Changing rpm should move the optimum flowrate and head closer to my operating point. Flow control doesn't need to be that accurate. Main point is not to consume 800 W all the time. Also 1.1 l/s is much, since the water needs to be dumped somewhere. There seems to be enough water coming from the well. The water level didn't drop at all when I pumped one hour (4 m3).

Just a minor thing so everyone is on the same page. Why I was suggesting a valve or plate you won't be pumping more volume than you require. You are effectively increasing the head the pump sees and as such decreasing the volume it passes. With this reduction in volume the power decreases. ** This applies to centrifugal pumps and NOT to Positive Displacement types (some of the DC pumps are Helical Screw types) so speed control is needed to modify flow or you could also return some back down the bore (inefficient).

[jmaja]

All the curves I have seen for 1-phase borehole pumps shows lower power input at low head + high flow rate compared to high head + middle flow rate.

I didn't realise there are 12 and 24 V borehole pumps. There aren't any available here, but easy to order abroad. Do they last as well? There seems to available about 200 W models with about correct head and flowrate. Some are not stainless steel.

[jmaja]

I have looked at some of the 24 V borehole pumps. They seem to all be targeted for solar usage. Nothing wrong with that, but I haven't found any details yet how they operate. Most seem to have a MPPT controller. This is mostly external. They seem to have brushless permanent magnet motors. But only two wires to the pump. Doesn't a brushless motor need to have several poles and thus more than two wires? And PWM with accurate phase control needs to be used.

Is the motor controller inside and only simple DC goes into the pump? How can the rpm be controlled then?

Not much curves for these 24 V pumps. This 200 W pump seems to be able to output 0.4 l/s and maybe 17 m, which means 33% efficiency. Maybe the controller can reduce the power as I want? Quite much more expensive than many other options, but OK price, if it works well and lasts.
https://www.ebay.com/itm/3-DC-Shallow-Well-Solar-Water-Pump-24V-200W-Submersible-Off-Grid-MPPT-BoreHole/233371349632?hash=item3656043280:g:IrQAAOSw4mNehZy2

It would be nice to see a manual or datasheet.

[beanflying]

Think of the motor more like a PC fan than say an R/C BLDC one so it has no brushes and only two wires.

Typical bore pump motors use a standard NEMA bolt pattern and shaft coupling but that was always a bit hit and miss even with larger companies deviating from this standard. The DC versions seem not to be following any standard so are likely custom made for the pumps. But if you can find a manufacturer of NEMA Flanged and shafted DC motors then this is an option (likely more expensive).

Three main types of pump are Helical Screw, Diaphragm and Centrifugal. The first two are positive displacement types and more suited for higher head lower flow applications so stick with centrifugal.

In the centrifugal ranges you will also find Plastic or Stainless impellers providing you don't have to much sand or fine grit in your bore either are fine the rotating parts and bearings are still metal in both cases.

Had a quick look locally and most of the DC pumps appear to be sole imported out of China rather than the larger manufacturers like Grundfos and Lowara.

Coffee time...

[beanflying]

Made a couple of phonecalls to some people I still know in the industry. Seems none of the main players really do DC in 4" diameter The exception to this has always been Mono (Helical Screw) and Shurflo (Diaphram) which as per earlier higher head lower flow.

This Chinese based seller 'seems' to be one of the sources of Pumps at least in Oz https://kairuijidian.en.alibaba.com/company_profile.html?spm=a2700.shop_index.88.51 so they might be worth contacting directly to try and find a curve and suitable model.

Apart from that I suspect you will be importing your own if you go down the DC path.

[jmaja]

So whatever type of motor they have in their 24 V DC pump I could just skip the possible MPPT-controller coming with the pump and control the rpm with plain 24 V DC PWM?

What method for rpm control does the Grundfos SQE use? What is special about the SQE vs. all the other Grundfos 230 V 1-phase borehole pumps? It can be used directly with 230 V or via Grundfos control box, which allows efficient rpm control. There seems to be only Danish datasheet. I can understand the most of it, since it's close the Swedish, which I can. https://api.grundfos.com/literature/Grundfosliterature-1401035.pdf

It uses MSE 3 motor and the MS 3 is the same, but without rpm control. Also for it the datasheet seems to be only in Danish. It says "MSE = Elektronisk regulerbar motor med mulighed for kommunikation via CU 300 og CU 301", which means electronically regulated motor with possibility for communication via CU 300 and CU 301. So are the rpm control electronics in the motor and some data transfer method via the normal three wires for 230 V?
https://api.grundfos.com/literature/Grundfosliterature-145908.pdf

Is using VFD with 230 V 1-phase ruled out? Even if the capacitors would be in the control box and not under water? Is there a problem with this type of usage: https://www.ato.com/using-vfd-for-single-phase-motor

[Berni]

If its a brushless pump then the 24V are powering the integrated BLDC driver electronics. So PWM might not make it very happy. But you can offten run these controllers at a much lower voltage and the motor spins at a slower speed as a result (it doesn't have the voltage to overcome the back emf of the motors windings at higher speed). So it will probably run just fine at half speed on 12V

[beanflying]

The Chinese 24V examples will be DUMB motors unless otherwise mentioned so yes simple PWM control should be ok.

The SQE MSE based pump on reading the manual is not a synchronous motor and has some form of powerline data to and from the motor to allow some additional functionality from their proprietary controller to some motor mounted electronics. Costs on these was always a large premium even when fitted to above ground Pumps for water supply. MSE spec here https://product-selection.grundfos.com/sg/products/ms-3/mse-3-96160539?tab=variant-specifications

The MS3 motor spec is here and is also 10k RPM+ https://product-selection.grundfos.com/sg/products/ms-3/ms-3-96160535?tab=variant-specifications Interestingly the motor is described

MS 3
The motor is a single-phase motor of the permanent magnet rotor type ensuring optimum efficiency within a wide load range. The motor is fitted with a replaceable end cover with socket.

As to VFD control of a synchronous 4" (your existing?) pump motor without factory information I wouldn't. As the heating issue with speed turndown is largely negated by the water cooling maybe but 

[jmaja]

The Chinese seem to be brushless. Doesn't that mean they can't be dumb motors and must have an internal controller?

1-phase VFD always has a heating problem? There is very little information about it. I understood that the worst problem would be with starting capasitor remaining applied when the centrifugal switch doesn't operate at lower rpm. But if I can remove (haven't checked that yet with the current pump), that can't be the problem.

It would be quite cheap and easy way to try it. I can get a good second hand VFD for 100 €. If I still need to buy a new borehole pump, I don't mind burning the one I just bought (139 €), if there is a chance for this to be a good option.

[richard.cs]

1-phase VFD always has a heating problem? There is very little information about it. I understood that the worst problem would be with starting capasitor remaining applied when the centrifugal switch doesn't operate at lower rpm. But if I can remove (haven't checked that yet with the current pump), that can't be the problem.

If you take the capacitor off you could run it with a 2-phase VFD, or possibly with 2 phases of a 3-phase VFD. I'm not really sure how now capacitor and 1-phase vfd would work - where's the rotating field for start-up.

[jmaja]

This guide connects all the three phases.
https://www.ato.com/using-vfd-for-single-phase-motor

I have seen other ones, which use only two phases. But then you do not remove the capasitors.
http://www.gohz.com/1-hp-single-phase-output-vfd

Both of these are with a 3-phase output VFD and two or three phases are used. I haven't even seen a one phase output VFD. There are plenty of one phase in three phase out VFD's.

[jmaja]

I called Mitsubishi VFD distributors technical support. They said it should be OK to use a 50 Hz 1-phase motor down to 15-20 Hz even with the capasitors. If you try lower Hz, it will supply current, but the motor won't run. You just need to turn the output phase loss protection off.

[jmaja]

I finally opened the control box, which had tri-wing screw and I didn't have these special screw bits with me previously.

There's one big capicitor and 3+1 wires to the pump. Thus I probably can use these 3 wires with 3-phase VFD. Likely there is still a small running capacitor inside the motor.

[sam_sam_sam]

As general rule if you are going to single phase VFD drive that does not have 3 phase input you have to double the amperage of the motor current but even doing this does not guarantee that you would not fry the drive doing this but there single phase VFD but most of them are a minimum of 240 volt input you might find a 120 volt version but will take a lot of researching and they are going to cost a lot more than a regular VFD 3 phase will cost

First of all you have to use a 3 phase motor to be able to use a 3 phase VFD drive

Now I have seen some where I can not remember what company has these VFD drive that are single phase VFD drive which would work but you would have to have the right motor for this purpose 
but like said earlier unless the drive is setup as single phase VDF drive it will not work correctly

Now depending on your current draw on the motor you have there might be a solution to your problem

[WattsThat]

As general rule if you are going to single phase VFD drive that does not have 3 phase input you have to double the amperage of the motor current but even doing this does not guarantee that you would not fry the drive doing this but there single phase VFD but most of them are a minimum of 240 volt input you might find a 120 volt version but will take a lot of researching and they are going to cost a lot more than a regular VFD 3 phase will cost

First of all you have to use a 3 phase motor to be able to use a 3 phase VFD drive

Now I have seen some where I can not remember what company has these VFD drive that are single phase VFD drive which would work but you would have to have the right motor for this purpose 
but like said earlier unless the drive is setup as single phase VDF drive it will not work correctly

Now depending on your current draw on the motor you have there might be a solution to your problem

Not sure you’re stating that correctly. I’ll take my whack at it.

If you use a standard three phase input VFD with a single phase input, the output amperage rating of the VFD should be (at least) two times the amperage required by the motor. It does not mater if it is a single phase or three phase motor, the same derate applies. Amps are amps.

The doubling of rating is an approximate rule of thumb and does not always work with all brands of drives.

The reason for the rating increase is due to the input current as seen by the input rectifiers goes up by the square root of three and the bus capacitors must increase in capacitance due the reduction of input frequency by way of losing two phases. Instead of a rectified 300 Hz pulse rate, the dc bus capacitors only see 100 Hz (with a 50 Hz system). Most drives measure the ac ripple on the dc bus and this causes a fault trip when a high value is seen by the VFD. Many times the fault shows up as an input phase loss. The drive doesn’t actually know anything about the incoming phases, it just knows what the dc bus ripple is high - so the fault doesn’t occur until a high load with a corresponding high bus ripple occurs. This fools many users into thinking the drive bad when it’s just undersized.

There are plenty of VFD’s out there up to about 2.2KW that are single phase in, three phase out. Those can typically made to work with a single phase motor so long as you can disable the output phase loss fault. Some of the really cheap drives don’t have output phase loss because they just measure DC bus current by way of a shunt and extrapolate the AC motor current.

The only manufacturer of a single phase in, single phase out VFD that I am aware of in the industry is Invertek out of the UK, known as the purple drive guys. Check their website for more info, their drive reliably runs a single phase motor without removing caps or other modifications (although I don’t think the drive could possibly work with a centrifugal switch on the starting cap).

One of the things never mentioned in these discussions is the risk for motor bearing failures caused by the eddy currents created in the stator by the PWM switching energy. It’s well known in the industrial world.

Another issue is that VFD’s will cause a rise in stator temperature, again due to PWM switching and the resulting harmonic distortion. I doubt it would matter on a submersible pump with a duty cycle. But the potential for stator heating is real.

If you have any VFD questions, ask away. No specific single phase motor experience but I know a few things about drives. It’s all 3 phase industrial but yeah, that’s the day job. Drives up to 700V and 5MW.

[jmaja]

I have no problems using three phase VFD, but the motor is one phase. We have 230 V system here in Finland. Thus I can use 230 V single phase or 380 V three phase as input to VFD.

There is just one capacitor (two wires to it, thus not a combined run/start) in the control box. Four wires going from the box to the motor.

There is no current to the capacitor when the motor is running  according to my true RMS clamp meter. About 3 A to the motor while plug power meter shows 800 W.

Should i remove the capasitor or not? I have a LCR-meter. Maybe I can use it to find out more about the motor and possible run capacitor. What to measure?

I'm looking for a second hand VFD. There are a few available here in the 50-100 € range. Should I go for three or single phase input? What would be the minimun power or current rating?

I almost bought a Mitsubishi FR-740 with 3.7 kW 8A rating for 100 €, but someone was faster than me.

There's also FR-S520S 1.5 kW (single phase input) and Danfoss VLT 5002 2.2 kW (three phase).