1 Phase 240V Power (PF) measurement on 3HP Induction Motor

Status
Not open for further replies.

dan198741

Member
Location
usa
I'm trying to measure the power consumption of a 1 Phase 230V 3 horse power induction motor. Basically I want to see if my harbor freight 3HP motor is actually able to provide 3HP of mechanical power at the axle (merely academic interest).
But first I need to be able to measure the electrical power coming into the motor as at the end I'm also interested in evaluating the efficiency of the unit.
What I've done so far is to measure the input current with a clamp meter and the voltage with a multimeter.
How do I go about to evaluate/measure/estimate the Power Factor?

For equipment I have multimeters, RPM meter, oscilloscope (voltage probes only) and an old power meter that I have not taken the time to check or read the specs.
If I use the power meter, would I need to take any special precautions when connecting it to the motor? I'm not sure if the inductive nature of the motor could damage the power meter...?

After this is figured out I will apply a Belt Brake to measure the actual mechanical power delivered by the motor, hopefully I get a figure close to 3HP.

I'll appreciate any suggestions.

Regards.

Dan
 

gar

Senior Member
141003-2028 EDT

dan198741:

As an electrical engineer you basically know what to do, but may not have experience with with specific instruments.

All you need is a suitable power meter, and possibly some auxiliary components. You also need torque and speed measurement. How do you plan to do this?

Describe the power meter that you have.

I have a moving coil Simpson 75 W full scale 120 V nominal meter. I could make this meter do your measurement.

I also have several Kill-A-Watt EZ meters. 120 V nominal and 15 A max. This would be harder to adapt to your measurement.

A TED 1000 or 5000 energy (power) monitor could be used but it has gradually increasing errors as power factor moves away from unity.

Ideally you probably want a power meter with 5000 W full scale designed for a nominal 240 V single phase input with short time current overload capability of 250 A (20 * 12.5).

.
 
Last edited:

dan198741

Member
Location
usa
Output Power ( Watts)/ Input Power ( volt Amps)= Power Factor. Your Output power should be 2238 watts ( 746 watts x 3 HP)


I think Power out/Power In = Efficiency not Power Factor. PF would only be measured at the input (electrical side).
PF=Real Power/Aparent Power
 

dan198741

Member
Location
usa
141003-2028 EDT

dan198741:

As an electrical engineer you basically know what to do, but may not have experience with with specific instruments.

All you need is a suitable power meter, and possibly some auxiliary components. You also need torque and speed measurement. How do you plan to do this?

Describe the power meter that you have.

I have a moving coil Simpson 75 W full scale 120 V nominal meter. I could make this meter do your measurement.

I also have several Kill-A-Watt EZ meters. 120 V nominal and 15 A max. This would be harder to adapt to your measurement.

A TED 1000 or 5000 energy (power) monitor could be used but it has gradually increasing errors as power factor moves away from unity.

Ideally you probably want a power meter with 5000 W full scale designed for a nominal 240 V single phase input with short time current overload capability of 250 A (20 * 12.5).

.


For the speed measurement I have a RPM meter, this is not a problem. For the torque measurement I plan to use a Belt Brake, basically is a way to slow down the motor with a known force I included the reference file.
I also have a kill-A-watt but those are 120V single phase, I don't want to test what happens if I use it in a 240V line.
The other power meter (the old one) I need to check, but I'm running against time to finish my deck right now so I haven't got time to check it out.
 

Attachments

  • s Handbook 28th Edition2.pdf
    60.4 KB · Views: 0

Jraef

Moderator
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
OK, let's first eliminate your assumptions so you can deal with the realities.

I'm trying to measure the power consumption of a 1 Phase 230V 3 horse power induction motor. Basically I want to see if my harbor freight 3HP motor is actually able to provide 3HP of mechanical power at the axle (merely academic interest).
But first I need to be able to measure the electrical power coming into the motor as at the end I'm also interested in evaluating the efficiency of the unit.
OK, two SEPARATE issues here.
Mechanical HP = TORQUE x RPM ? 5252
You cannot evaluate the mechanical output of the motor without knowing the speed and the mechanical torque output. Amps, volts and power factor have to do with the ELECTRICAL power, but are only directly related to the mechanical output. The best you can hope to do with the electrical information is to evaluate whether the motor is CAPABLE of the mechanical HP output you are looking for, but it will not tell you exactly how much physical torque it is actually putting out.

Then the only way to know the throughput efficiency is to know EXACTLY how much mechanical power the load is requiring. Just because you have it hooked up to a load does not mean that load is REQUIRING the full HP capacity of the motor, the load will only demand what it needs from the motor. So unless your load is a dynomometer where you can dial in an exact load to be expressed on the motor shaft, this entire exercise is relatively pointless.

What I've done so far is to measure the input current with a clamp meter and the voltage with a multimeter.
How do I go about to evaluate/measure/estimate the Power Factor?

For equipment I have multimeters, RPM meter, oscilloscope (voltage probes only) and an old power meter that I have not taken the time to check or read the specs.
If I use the power meter, would I need to take any special precautions when connecting it to the motor? I'm not sure if the inductive nature of the motor could damage the power meter...?
What is your "power meter"? A kW meter? If so that is the exact device that you needs, the other meters add nothing.


After this is figured out I will apply a Belt Brake to measure the actual mechanical power delivered by the motor, hopefully I get a figure close to 3HP.

I'll appreciate any suggestions.

Regards.

Dan
So how are you planning on ensuring that your belt brake is going to express exactly 3HP worth of load on to the motor? Because if you can somehow know that, then that is the end of the experiment, right? Put exactly 3HP on the motor, measure the electrical kW, compare that to 2.238kW, the difference is your efficiency.
 

gar

Senior Member
141006-1602 EDT

dan198741:

Of the various friction brakes shown in your attachment I like figure 5 because it would probably introduce the least error. You need accurate perpendicular force and radius arm measurement. You may have a problem controlling friction load constant. Because I have an old power steering pump I might use that as a load with a solid coupling and the lever arm mounted to restrain the floating pump. Then restrict oil flow to control torque. Use enough oil so that temperature does not rise too rapidly.

How to use the Kill-A-Watt. You are correct that it won't tolerate 240 V input. To solve the voltage problem use a 2 to 1 isolation stepdown transformer. 25 VA is adequate. You might also use a resistive divider. The Kill-A-Watt is a load of about 0.3 W and 5.6 VA at 120 V. Thus, if a resistive divider is used, then the resistance needs to be low. Two 1500 W heaters of equal resistance in series would probably be adequate. The heaters will draw somewhat over 10 A (the big box store heaters may not be close to their rating). !0 A vs the Kill-A-Watt 0.05 A (200 to 1 ratio) should result in a fairly good voltage signal. Whether a transformer or resistive divider is used you will need to calibrate the Kill-A-Watt reading by some scale factor.

The Kill-A-Watt uses a low resistance resistive shunt for current measurement. If you do not expect over 15 A in your steady-state tests, then nothing needs to be done to the shunt. 15 * 240 = 3600 VA.

The shunt is between the Kill-A-Watt input neutral plug pin, and the Kill-A-Watt output neutral socket. Thus, the connection will be such that the Kill-A-Watt neutral will be in series with one 240 V hot line. Be careful. The other 240 hot goes directly to the other side of the motor.

The above calibrated on a known resistive load will provide good results on a very poor power factor load because of the use of a resistive shunt.

A way to use the Kill-A-Watt by just plugging into a normal 120 socket is to use a current transformer to measure load current and couple this to the Kill-A-Watt shunt. This will provide electrical isolation and the ability to work with much larger loads. However, the current transformer introduces phase shift and resulting errors. You can make your own current transformer with a high quality tape wound toroid. I made one some years ago with a 4" OD x 2.5" ID x 1" width (possibly Supermalloy). 10 turns on secondary and 1 turn primary. I really never checked for phase shift or burden. This was used to increase the range of my 75 W Simpson. Really a way oversized core.

To check for non-unity power factor problems put a resistive load on the meter of approximately the load you want to measure. Using good quality capacitors (polypropylene) add in parallel about an equal current in capacitive load to the resistive load. Ideally the power reading would remain constant. Next remove the resistive load and the power reading should drop to 0, but if it is in the range of a few % of the previous resistive load, then you are probably getting good results.

With a 120 V source, and only a 70 ufd capacitor as a load:
A Kill-A-Watt read 1.6 W and 361 VA.
A TED 1000 read 28 W, doesn't read VA.
A TED 5000 read 30 W and 360 VA.

.
 

iceworm

Curmudgeon still using printed IEEE Color Books
Location
North of the 65 parallel
Occupation
EE (Field - as little design as possible)
No real new information - jus mostly summerizing:
...For equipment I have multimeters, RPM meter, oscilloscope (voltage probes only) and an old power meter that I have not taken the time to check or read the specs. ...
As said, the power meter is the first choice. Read the book

...If I use the power meter, would I need to take any special precautions when connecting it to the motor? I'm not sure if the inductive nature of the motor could damage the power meter...?
Just connect it per the book. Some of the older stuff I've worked with (University power lab, 1971) had 120VAC coils. As gar noted, you will need a pt. You won't need an instrument grade. Use the DVM to get accurate calibration factors. And of course, check the current input. 240V, 1ph is about 18A. And you likely will overload to get the full range of the torque/rpm curve. Depending on the meter specs - you may also need a CT. If you go clear to pullout, current could easily be 200%.

... You cannot evaluate the mechanical output of the motor without knowing the speed and the mechanical torque output. Amps, volts and power factor have to do with the ELECTRICAL power, but are only directly related to the mechanical output. ...
I think Dan understands that.

...After this is figured out I will apply a Belt Brake to measure the actual mechanical power delivered by the motor, hopefully I get a figure close to 3HP.

I suspect your plan is to connect the power meter, rpm indicator, and shaft brake (belt loose).
Start the motor
Incrementally tighten belt and take W (power), VA, rpm, Torque readings.
After the data is taken, calculate HP at each belt tension

You should get a pretty standard Torque/ RPM motor curve. Somewhere around 5% - 10% slip - just past max torque - I would expect the motor to pull out and stall.

Only thing I would caution is a harbor freight motor likely won't take much overload for very long before the smoke comes out. I would consider fuses ahead of the power meter/motor. They likely won't save the motor, but should cut down on the flying burning shrapnel - should things not go as per plan.

Heck of a science fair project. Let us know how it comes out.

ice
 

kwired

Electron manager
Location
NE Nebraska
There is a reason we put overload protection in motors/motor circuits. Your motor will deliver more then it's rated 3 hp, but if truly designed for a maximum of 3 hp output it will operate at elevated temperature if it continuously delivers over 3 hp and will suffer degrading of winding insulation because of this excessive heating.

You can not measure output by simply measuring input values, unless you know power factor and efficiency, but you appear to be questioning those so you need to maybe connect this motor to a dynamometer so you can measure actual output performance. When loading your motor you would need to monitor it's operating temp as well to know what the output is when it reaches maximum operating temp.

You can build your own dynamometer by driving a generator and supplying resistive load banks from that generator. You do need to consider power factor and efficiency of the generator if doing it that way to be accurate though.
 

dan198741

Member
Location
usa
Thanks Jraef, Gar, iceworm and kwired, definetly very valuable feedback that will help guide the testing.

(Before this post, I didn't know if my power meter would work so I was looking for options to measure the power factor without a power meter)

I just looked at the power meter specs, it's a Voltech PM1200 here are the specs
http://www.avalontestequipment.com/datasheets/VOL_PM1200.pdf
It seems the unit is a good fit for the measurement, I just got it from a factory discarded pile but it seems to work just fine.
I don't have a dynamometer but I plan to use the Belt Brake described in the pdf I enclosed before. I will try the one in figure 4 first and if it doesn't work I'll try the one in Figure 5.
What I'll do is start the motor (no load) conected to the power meter and then add a weight to the arm for just enough time to make the measurement and avoid heat building up, then repeat adding more weight each time until I reach the specified 12.2A or just a little over. At the same time I need to measure RPMs, power, PF and with that I should be able to draw some nice curves.
I'll post the results when I'm done in case there is interest to see it.
Thanks!
Dan.
 

Jraef

Moderator
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
Thanks Jraef, Gar, iceworm and kwired, definetly very valuable feedback that will help guide the testing.

(Before this post, I didn't know if my power meter would work so I was looking for options to measure the power factor without a power meter)

I just looked at the power meter specs, it's a Voltech PM1200 here are the specs
http://www.avalontestequipment.com/datasheets/VOL_PM1200.pdf
It seems the unit is a good fit for the measurement, I just got it from a factory discarded pile but it seems to work just fine.
I don't have a dynamometer but I plan to use the Belt Brake described in the pdf I enclosed before. I will try the one in figure 4 first and if it doesn't work I'll try the one in Figure 5.
What I'll do is start the motor (no load) conected to the power meter and then add a weight to the arm for just enough time to make the measurement and avoid heat building up, then repeat adding more weight each time until I reach the specified 12.2A or just a little over. At the same time I need to measure RPMs, power, PF and with that I should be able to draw some nice curves.
I'll post the results when I'm done in case there is interest to see it.
Thanks!
Dan.
If that meter works, it displays watts, so I would not waste time plotting out amps and volts and PF, id just look at the watts, it tells you everything you need to know. But that's just me, maybe you want to check the accuracy of the meter's ability to measure watts? Whatever.

So to gars earlier point, what you will really need to know is if the motor can put out 3HP worth of mechanical power without overloading itself then. The problem is, again, that unless you can accurately measure the MECHANICAL power, you still will not really know using your method. You are going to increase load with the belt until you read 2238W going to the motor, then ensure that the current being drawn at that point is less than or equal to the FLC rating on the nameplate.

But still, that means the motor is CONSUMING the equivalent of 3HP worth of electrical power, but it does NOT mean the shaft is putting out 3HP worth of MECHANICAL power. It will be close, maybe close enough, but not exact, because there are losses inside of the motor that consume electrical power, and will be seen by the circuit, but do not equate to mechanical output. If your motor has an efficiency rating at full load, that can be applied to your findings to be more accurate, maybe as best as you can get without a dynamometer. I'm just bringing it up so that if, for example, you find that the motor is looking too good to be true, it isn't.
 

gar

Senior Member
141007-1729 EDT

dan198741:

I don't think you can expect to know the coefficient of friction very accurately. A small change in lubrication makes a big difference. In building differentials, if the pinion nut troque runs low, then run the nut face on some sandpaper, and if high rub on some wax. Also the dimensions of figure 4 may be hard to control. Figure 5 eliminates the coefficient of friction from the calculation, and the dimension is much easier to determine.

.
 
Status
Not open for further replies.
Top