RPM Measurement

[jrexrode]

I'll second milldrone's opinion on the ACS800 drives. I have quite a few in operation, and they are dead on without any feedback whatsoever.

Milldrone, you don't have to have the motor uncoupled for the drive to get an accurate motor profile. While having it uncoupled and performing a full ID run will give you the best perfromance, I have several drives that I perfromed just the motor magnetization tune. The motor doesn't have to be uncoupled, and it won't rotate during this. I can't tell any difference between this and the full ID run when it comes to our applications.

Hope this helps.

 

[kamenges]

Originally posted bu RichardB:

I can simply put a VFD in to passively measure RPM's, with no feedback sensor.

If by 'passively' you mean measure RPM without controlling the motor itself, then the answer is no. The drive will be controlling the motor speed. If by 'passively' you mean measure RPM without a feedback device, then the answer is yes.

While there may be other drives that use DTC as a control method I think ABB is the best and most experienced at it. If it works out that this is the way you want to go I wold definitely talk to the ABB guys.

What HP range are you talking about for the pump motors. If it is 1HP - 100 HP you will need multiple drives. If it is 5HP to 20 HP you can PROBABLY use a 20HP drive and current limit the drive for 5HP operation. DickDV or another ABB rep will need to answer whether you loose so much resolution in the 5HP case that you can't accurately track speed.

Assuming you have the drive software you would need to do a motor identification run for each motor type at least once. After that you MAY be able to just download the saved parameters for each motor model you use. DickDV or another ABB rep will have to answer whether that gets you close enough. For the jobs I have done in the past with other drives I can get away with this.

Barring this as an option I would look to a high speed retroreflective laser PE, put a piece of reflective tape on one fan blade and measure speed that way. Banner, Keyence and others make models that will do the trick for you.

Keith

 

[jrexrode]

Thinking about this some more, if you have access to the fan on the back of the motor, perhaps you could extend a shaft out the back and mount an encoder and display RPM on a Red Lion meter or some other similar device.

 

[Peter Nachtwey]

Peter,

As far as what's wrong with the specification, please enlighten me.

First how does one measure speed that accurately? One can measure distance traveled over a few seconds and perhaps be able to get a measurement that has that kind of resolution. I would bet you want to know MUCH more rapidly than once every couple of seconds. Think about it. 0.5% is one count out of 200 over how much time? At 1800 RPM is 30 revs a second. If there are only 6 bolts or fan blades to count on each revolution that is still only 180 counts per revolution this means you would need to wait a minimum of about 1.1 seconds to get the desired resolution. Is this fast enough? What about while accelerating or decelerating? Calculating the speeds during these times is very challenging. Counting counts over 1.1 seconds will only give you the average for that period. Not the true rate at the end of the period.

Now what happens if you don't have 6 bolts or fan blades to detect.

A good drive has a model and some math were it is calculating ( estimating ) the current state ( position, velocity, acceleration ). This state is necessary for commutating fields,PID control, calculating feed forwards and handling the internal torque loop. This estimated state is what you want to read out of the drive. It should be very accurate and be updated at the drives internal loop scan which should be faster than 1 millisecond.

If the current drive is not a smart one when it may be cheaper ( time, money, performance, cost ) to get a smart drive rather than go through all the hassle of mounting prox switches and writing ladder code etc. which will NEVER be as accurate as the information from the smart drives.
Now you just need to find a drive that will easily communicate that information to your PLC. What kind of PLC is that, RichardB?

 

[RichardB]

Peter,

Thanks for the explanation. Time isn't a huge factor for us. We can hold a flow point for several seconds and measure the RPM's. In reading alot of your previous posts on this forum, I have a pretty goods idea of where you're coming from. I always try and understand your posts and learn something, which is why I was happy you were responding to this.

The PLC is an automationDirect 400 Series with minimal ladder. Most of the control is done using a custom VB.net application running on an XP based PC.

 

[RichardB]

If by 'passively' you mean measure RPM without controlling the motor itself, then the answer is no.

This is what I meant by passively. During the test I need to also collect voltage, current and Watt data.

Barring this as an option I would look to a high speed retroreflective laser PE, put a piece of reflective tape on one fan blade and measure speed that way. Banner, Keyence and others make models that will do the trick for you.

We have several of these already for use on units where we can see the shaft. I was hoping we wouldn't have to take the fan shroud off of every unit. This does seem like the easiest option.

Thanks Guys, If anyone has any further ideas they're all welcome.

Rich

 

[John Soltesz]

Guys,

I'm looking for a way to measure RPM's on close coupled pumps with no access to the shaft. We have used vibratory reed tachs in the past, but they keep getting dropped and then don't work so good.

I was hoping for a sensor magnetically attached to the motor could give me an analog RPM value or something I could relate to an RPM value.

Looking for Ideas.

Thanks,
Rich

Have you ever used a strobe for measuring RPM? We used to set the spinner of a spray dryer at 10,000 rpm using a strobe. All you need to see is some part spinning, shaft or fan and adjust the strobe until the image is still. Quite simple.

 

[DickDV]

Now watch this folks! DickDV is going to recommend against a drive!

It sounds to me like one of your tests would be to set a measured flow rate and then determine how much the motor/pump system droops in speed. If that is one of your tests, then, for all the wonderful things an ACS800 does, its not a good choice for this simply because it is too much hassle to do a motor ID every time you change motors and, further, too much hassle to constantly be adjusting the speed loop gains for tight and loose regulation.

Instead, I would take the test motors to your friendly motor shop and have them remove the backside cap over the shaft and have them install a stub shaft for a DC tachometer. Adapt a mounting surface to the rear endbell and you can read speed although the tachs are usually only rated 1% accuracy. If you have to be better than that, do the same thing but use an encoder instead. You will need somthing to read the encoder pulse rate to get speed. I suppose a low performance oscillascope would work for that.

 

[DickDV]

Now watch this folks! DickDV is going to recommend against a drive!

It sounds to me like one of your tests would be to set a measured flow rate and then determine how much the motor/pump system droops in speed. If that is one of your tests, then, for all the wonderful things an ACS800 does, its not a good choice for this simply because it is too much hassle to do a motor ID every time you change motors and, further, too much hassle to constantly be adjusting the speed loop gains for tight and loose regulation.

Instead, I would take the test motors to your friendly motor shop and have them remove the backside cap over the shaft and have them install a stub shaft for a DC tachometer. Adapt a mounting surface to the rear endbell and you can read speed although the tachs are usually only rated 1% accuracy. If you have to be better than that, do the same thing but use an encoder instead. You will need somthing to read the encoder pulse rate to get speed. I suppose a low performance oscillascope would work for that.

 

[DickDV]

Oops! sorry for the double post.

 

[Jesse Benefiel]

Today's tachometers usually employ a 1/tau (measures internal counts between pulses) measurement rather than timebase (counting pulses over time), so accuracies of .01% can be achieved. Now if only you can develop the pulses!

Good luck in finding a signal, should the drive solution not resolve your issues.

 

[Peter Nachtwey]

Just curious

Today's tachometers usually employ a 1/tau (measures internal counts between pulses) measurement rather than timebase (counting pulses over time), so accuracies of .01% can be achieved. Now if only you can develop the pulses!

What is the frequency that is used to count the time between pulses?

I would not use 1/tau to describe the method of using the time between pulses. I had to stare at your post for a second to figure out what you are really trying to say. 1/tau sounds like a time constant. That is what confused me for a second. 1/delta_t is clear.

 

[Jesse Benefiel]

Point taken Peter. We always called it 1/tau or "time interval" rate indication back when I was in technical support. That was, well, 16 years ago(!), so I don't remember how the engineers actually explained it to me when I wrote it in my notes.

I don't know the answer to your question RE the internal frequency - and the plant is closed so I can't ask. But since I'm responsible for the HMIs and not the panel meters, Richard is lucky that I don't try to talk him into a PLC and one of our HMIs.

 

[Gil47]

These are off in left field, well maybe even several paddocks over from left field.

I put this search term into Google "motor slip measurment"
and these 3 exerps caught my eye on the first 2 pages

http://www.patentstorm.us/patents/6042265-description.html
Most motor installations are not equipped with speed sensors so a direct measurement of the speed/slip is not possible. However signals appear in the current that may be used to estimate speed quite precisely. Virtually all motors have some degree of rotor asymmetry, which results in the appearance of frequencies in the motor current related to the slip. The most prominent of these is found at two slip frequencies below the line frequency and its magnitude is often used as a measure of the severity and number of broken rotor bars. Examination of this spectrum, which may be automated, therefore yields the required slip frequency.


http://ieeexplore.ieee.org/Xplore/login.jsp?url=/iel5/7381/20066/00928884.pdf

An induction motor speed measurement based on current harmonic analysis with Chirp-Z Transform
Aiello, M.; Cataliotti, A.; Nuccio, S.
Instrumentation and Measurement Technology Conference, 2001. IMTC 2001. Proceedings of the 18th IEEE
Volume 1, Issue , 21-23 May 2001 Page(s):578 - 582 vol.1
Digital Object Identifier 10.1109/IMTC.2001.928884
Summary:The paper presents a new method to measure motor speed by means of frequency estimation of rotor slot harmonics present in the supply current of squirrel single cage induction motors. The novelty of the method is the use of Chirp-Z Transform as supply current harmonic analysis. The advantages are an improved spectral resolution and accuracy. Moreover a shorter sampling time window is required reducing errors related to not stationary current signal. Experimental results are presented to validate the proposed method and to perform a comparison with FFT based one


http://ieeexplore.ieee.org/Xplore/login.jsp?url=/iel5/7381/20096/00929542.pdfNon-invasive slip measurement of induction motors
Micheletti, R.; Pieri, R.
Instrumentation and Measurement Technology Conference, 2001. IMTC 2001. Proceedings of the 18th IEEE
Volume 3, Issue , 2001 Page(s):1958 - 1961 vol.3
Digital Object Identifier 10.1109/IMTC.2001.929542
Summary:The paper deals with the accurate slip measurement of induction motors. The proposed procedure uses a non-invasive slip measurement scheme based on digital filtering and dynamic parameter estimation. The slip measurement is carried out without a speed sensor and is deduced utilizing the magnetic field in proximity of the induction motor. First the waveform of the induced e.m.f., taken from a search coil, is filtered using algorithms based on the discrete Fourier transform. Then the stator frequency and rotor frequency are obtained by comparing the filtered voltage with a mathematical model using an optimization procedure. The model's parameters are varied until an adequate match is obtained with the filtered voltage. Experimental results are presented to validate this method

 

[RichardB]

Instead, I would take the test motors to your friendly motor shop and have them remove the backside cap over the shaft and have them install a stub shaft for a DC tachometer.

Dick,

Thanks for the input, but I'm trying to do this on customer motors. Some may appreciate the stub shaft, although some/most probably wouldn't.