RPM Measurement

RichardB · Feb 26, 2008

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

milldrone · Feb 26, 2008

Richard,

This might be way out in left field, but how about an ABB VFD using their DTC sensorless vector. And using modbus or profibus comms? I have found this to be more accurate than any analog system.

RichardB · Feb 26, 2008

milldrone,

I'm not trying to control the speed of the motor, just measure it.

It's easy on units where the shaft is visible, just a piece of reflective tape and a sensor, but units where the shaft is not visible is a PITA.

Gil47 · Feb 26, 2008

If the motor has no speed control then read the name plate
2 pole motor on 60 HZ = 3600 - slip about 3550 RPM
4 pole motor................. 1800................. 1750

If it has a VFD in sensorless vector mode it will most likely display RPM on one of its screens. just as Milldrone says.

Steve Bailey · Feb 26, 2008

How about a prox switch on the back end of the motor counting the fan blades. You can connect it to a PLC with high speed counter capabilities or to a frequency-to-voltage or frequency-to-milliamps signal conditioner if you really want the analog signal.

RichardB · Feb 27, 2008

A bit more info

There is no VFD. The unit is on a test setup where we need to verify performance. I'd like to be 0.5% or better accuracy.

We've considered counting fan blades but would like something a bit more practical. Just don't know of anything else thats feasible.

Thanks for all of the input so far.

Rich

milldrone · Feb 27, 2008

RichardB said:
we need to verify performance. I'd like to be 0.5% or better accuracy.

Just don't know of anything else thats feasible.

Rich

Rich,

The ABB drives with DTC control are absolutely amazing. I'm suggesting using the drive because of the RPM stability not because of the ability to change speed.

In this thread linked below, DickDV discusses servo vs induction motor abilities. He also shares some experence with a very demanding test cell he had a hand in making function.
http://www.plctalk.net/qanda/showthread.php?t=7602&highlight=rpm+dtc

Here is a quote from the thread

DickDV said:
In my view, the single redeeming quality of servo motors is their dynamic performance. By that I mean their ability to make fast speed and direction changes. This is accomplished by very low inertia rotors and other winding characteristics.

As for positioning and speed accuracy, a good high performance AC or DC drive can do that today if a proper positioning controller is placed ahead of the drive. What an AC or DC drive/motor can not do is make the changes quickly.

Just for illustration purposes, I have a laboratory test cell system consisting of a 200hp ABB ACS600 drive using DTC (Direct Torque Control) and a special built water-cooled AC induction motor from Electric Apparatus Company that acheives 4 rpm speed error from no load to full load at 10000rpm. And its done without a tach or encoder on the motor. But, on step changes in speed reference, it takes about 700ms to get the speed to the new level and settle down. A full servo system could likely do the step changes faster.

Reliance Electric also makes a low inertia version of their excellent RPM-AC motor which, for speeds under about 4000rpm, is capable of near servo performance on a DTC-type drive. There may be others that I am not aware of, particularly in Europe and Japan. I have run into some spectacularly high performance AC induction spindle motor designs which have almost unbelievable spec sheets but I've never actually used one.

Note that he mentions no encoder!

RichardB · Feb 27, 2008

Milldrone,

Maybe I'm a bit confused here, but I do believe in DickDV's instance he is using a Drive to Control the Motor Speed. Using DTC and a special Motor he is able to get withing 4 RPM's of his target. Thats great.

My issue is that I DON'T have a drive. I have a Pump that by varying the Load(Flow) the motor RPM's also Varies. I need to capture those RPM values.

Again, maybe I'm confused. I don't believe I can simply put a motor drive in front of the pump, have it send 60Hz all the time, not affect pump performance, and expect it to measure RPM's for me.

If I'm wrong, please explain this too me.

Rich

Webhead · Feb 27, 2008

No your not wrong Richard. I know I'm late to the conversation but it seems like people are telling you to go ahead and buy a VFD just to read the RPM. Probably not what you had in mind as far as being cost effective but it would tell you the RPM of the motor and also give you more protection of the motor via faults and alarms that would shut it down in case of a problem.

If there is no possible way of seeing the shaft and you don't want to count the fan blades I'm not familiar with any other way of doing it but use a VFD.

kamenges · Feb 27, 2008

Originally posted by RichardB:

Again, maybe I'm confused. I don't believe I can simply put a motor drive in front of the pump, have it send 60Hz all the time, not affect pump performance, and expect it to measure RPM's for me.

I think you are thinking reactively not proactively. While the drive will accurately indicate RPM in all cases, I think the big benefit of the drive is that it will put the motor speed where you tell it to be and keep it there. If you test requires that the motor speed sag as the load changes you can run proportional gain only in the velocity loop. But I woiuld think you would want to kee pthe speed as constant as possible, especially in a test environment.

Keith

RichardB · Feb 27, 2008

ok, let me get this straight.

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

I've used VFD's in cases where I wanted to control machines, but don't know how it would passively measure rpm's.

I'm still very confused here. BTW, the pump(s) under test are of varying sizes and voltages. This would mean multiple drives and a tough cost justification. I can get through that if i can confirm it'll work.

Rich

Peter Nachtwey · Feb 27, 2008

You don't always get want you waaaant.

RichardB said:
I'd like to be 0.5% or better accuracy.

I can tell you haven't thought about this at all. Cringe when I get a specification like this. What is wrong with it?

I agree with those that say to read the speed out of the drive. The the speed may be derived from a model or some other advanced technique.

RichardB · Feb 27, 2008

Peter Nachtwey said:
I can tell you haven't thought about this at all. Cringe when I get a specification like this. What is wrong with it?

I agree with those that say to read the speed out of the drive. The the speed may be derived from a model or some other advanced technique.

Peter,

Thanks for your candor. You are correct in that I haven't thought this through. I'm looking for a simple, cost effective, way to measure RPM's, although it may not be all that simple. As far as what's wrong with the specification, please enlighten me.

As for the deriving speed from a model are you talking about an RPM vs Amp Draw curve, or something similar?

Thanks,

Rich

milldrone · Feb 27, 2008

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

Rich

Not just any drive, an ABB ACS800 is the only drive I believe that will do what you want. Note: I could be wrong there may be others . The DTC concept would be my choice considering your restraints.

RichardB said:
BTW, the pump(s) under test are of varying sizes and voltages. Rich

Actually this is something a VFD does very well. As a matter of fact one application is to use a VFD to supply a motor with 240 volts from a 480 supply. Sorry can't go the other way! Overload protection is just programming. BTW you can store 2 custom VFD profiles in the programmer, so you could have a 240 setup and a 480 volt setup. And load them in about 30 seconds.

There might be one snag though. In order to get the DTC working correctly the drive needs to tune the motor by itself. This means it cannot be attached to a load during this test.

I'd suggest PMing DickDV with your problem, see what he says.

RichardB · Feb 27, 2008

milldrone said:
Not just any drive, an ABB ACS800 is the only drive I believe that will do what you want. Note: I could be wrong there may be others . The DTC concept would be my choice considering your restraints.

I definitely have to check these drives out.

milldrone said:
Actually this is something a VFD does very well. As a matter of fact one application is to use a VFD to supply a motor with 240 volts from a 480 supply. Sorry can't go the other way! Overload protection is just programming. BTW you can store 2 custom VFD profiles in the programmer, so you could have a 240 setup and a 480 volt setup. And load them in about 30 seconds.

Hopefully programmable from a PC, say serial port?

milldrone said:
There might be one snag though. In order to get the DTC working correctly the drive needs to tune the motor by itself. This means it cannot be attached to a load during this test.

I'd suggest PMing DickDV with your problem, see what he says.

Hmm...Would it need to tune each different motor? every time I wanted to do a test?

RPM Measurement

RichardB

Member

milldrone

Member

RichardB

Member

Gil47

Member

Steve Bailey

Lifetime Supporting Member + Moderator

RichardB

Member

milldrone

Member

RichardB

Member

Webhead

Member

kamenges

Member

RichardB

Member

Peter Nachtwey

Member

RichardB

Member

milldrone

Member

RichardB

Member

Similar Topics