Homebrew Dynamometer ideas

In relation to my other posts on induction motor rewinding, I have need for a dynamometer to quantify the results of my experiments. Commercially available units are way beyond my budget, so I have some ideas on how to construct my own.

The motors I will be testing will be anything from 1/4 HP to a maximum of 5 HP original frame size, and I will mostly need to test shaft output power up to about 10 HP (maybe briefly to 20 HP), RPM up to 7200, and torque up to about 40 ft-lb.

As an initial setup, I connected the shaft of a 3 phase 1 HP 850 RPM motor to the shaft of a 1 HP 100 VDC permanent magnet motor, and connected a 2200 Watt 240 VAC electric heating element across the output. I read about 43 VDC across the approximately 26 ohms of the element, which is 71.1 watts or about 0.095 HP. The calculated torque is 0.589 ft-lb.

I can easily measure the shaft speed with a non-contact tachometer, and the total power should be fairly close to the output power of the motor used as a generator, with some allowance for efficiency. I can change the load resistance to get higher power at a given speed, or use a PWM system to make it continuously variable (and programmable).

However, a direct measurement of torque may be a bit trickier. I am considering mounting the generator on a set of bearings so it will be free to rotate around the shaft, but then restricting its torsional movement by means of a set of springs. I can put a pointer on the housing of the generator and make a scale calibrated in ft-lb, which I could calibrate with a torque wrench. Another way would be to weld a hex nut to the end of the generator housing, and use a torque wrench to realign the housing to its normal rotational position. Of course, the high tech way would be to use pressure transducers on the enclosure mounting. This would be best if I want to make a good speed/torque and speed/HP graph.

Comments and suggestions welcome.

Paul E. Schoen, President P S Technology, Inc.

715 Warren Road Cockeysville, MD 21030-2824

You want cheap? Google "prony brake".

Good Luck! Rich

On Sat, 16 Jul 2005 01:36:22 -0400, Paul E. Schoen wrote: ...

I once "measured" the torque of a stepper motor, by improvising a little crank for its shaft, and had it pull a spring scale that I was holding in my hand. I ass-u-me'd that it'd be close enough to the ballpark for my purposes at the time, but it's incredibly time-intensive, like it takes all day to set it up.

Check google for a "Prony Brake" - AFAIK, it gives the next best thing to a direct measurement of torque - you have to do a little arithmetic, but I've just now done a little browsing, and I think you might find some useful information there.

If you have a whole bunch of motors, it could be worth spending some time to set up the setup. ;-)

Good Luck! Rich

I thought you were looking for cheap. The prony brake shown in the URL's I posted are mainly conceptual. You have to put a hollow drum on the end of the shaft of the motor under test, with the friction applied to the outer surface of the drum. The drum is open on the end and contains water and thus can easily dissipate a horsepower for some time.

Prony brakes are fine for classroom demos but are never used for motor testing because they will not hold anything like constant torque for more than a few seconds due to excessive heating, since all motor power is converted to heat at the friction surfaces. 1 HP = 2545 BTU/Hr; a real dyno must be able to dissipate this heat without torque change or self destruction.

Of course the generator efficiency varies greatly with changes in speed and load, somewhat less with changes in temperature, so this method is not very accurate.

The usual method would be to use a load cell on the torque arm, calibrated periodically with known weights hung from the arm at a known distance from the support bearing centerline. Torque and speed signals are then used for closed loop control of the dyno. If you intend to produce the performance curves you mention more than a few times it will definetly be worth your time & money to buy or build a load cell and implement closed loop control.

Or, if you want really cheap, put a bathroom scale under the end of a moment arm of known and convenient length.

That would seem to be the most straightforward way to measure torque. Anything you do from the output of your DC generator will require corrections for efficiency, which will change as the winding temperatures change.

You could replace the DC generator with a water pump or centrifugal fan. The water or air flow can be restricted to vary the load on the device.

Another idea is to use a generator with a field winding. Drive a resistive load (as you describe above) but vary the load level by varying the field. This would pobably be the easiest to build into a feedback loop for constant load regulation.

--
Paul Hovnanian     mailto:Paul@Hovnanian.com
------------------------------------------------------------------
Optimist:  "The glass is half-full."
Pessimist: "The glass is half-empty."
Engineer:  "The glass is twice as big as it needs to be."

7200 RPM is fast for even 10HP.I think you'll have difficulty finding something big enough to absorb 20HP that will stay together at that speed (at an affordable price). 4500 RPM is probably easier to find.

Your best plan is to find a shunt-wound DC motor (or better still a proper DC generator), run it into a fixed load and control the loading with the field (Variac and rectifier). That's how it's usually done. Finding one that won't disassemble itself at 7200 RPM may be trickier. I've seen a runaway motor put shrapnel through 1/16" machine guards.

Don't forget, if you're running up a 10 HP motor, you've got to get rid of about 7.5 kW of heat out of your dynamometer load. Also, don't go over the nameplate voltage of the generator, or the commutator will probably flash round.

Mount the generator on an axial rotating mount, and constrain its movement with a radial arm, acting on a spring balance, or better still, a load cell. You can vary the arm length to get different multiplication factors. Use an additional end stop to limit rotation to something sensible.

That's how commercial dynamometers did it for years.

Don't forget lots of guards and E-stops within arm's length. Motor test rigs can kill you..

I can put a pointer on the

Just do it by first principles, radius and force. Torque wrenches are not that accurate.

The most important bit is the baseplate. Have a dig around the local junkyards for scrap machines with big flat tables with lots of T-slots in them. Don't even *think* of skimping on how the rig is fastened down. If you've ever seen what happens when a loose motor gets braked suddenly, you'll know what I mean.

--
"Electricity is of two kinds, positive and negative. The difference
is, I presume, that one comes a little more expensive, but is more
durable; the other is a cheaper thing, but the moths get into it."
                                             (Stephen Leacock)

posted are

the motor

drum is open

some time.

You are quite right about cheap and able to dissipate 1 hp for "some time". I have used commercial water drum and friction belt prony brakes up to 2 hp, and other variations of prony brakes have been used in sizes capable of absorbing the up to 20 hp the OP requested, although probably not in the last 50 years or so. No problem dialing in a specific torque and taking a quick reading, but even at 1 hp the torque will start to drift slowly as soon as you let go of the load adjustment, and brake lifetime will be limited by wear of the friction surfaces, which will be shedding fine dust into your work area. Closed loop control of torque on a prony brake is possible eg with servomotor control, although I doubt if this has ever been done, and I don't think the prony brake is a particulary good way to generate motor performance curves or measure steady-state thermal performance IAW IEEE STD 112 (or 114), standard test procedures for polyphase (single phase) induction motors.

But indeed if cheap is the primary criteria you can't beat the prony brake - unless you find a real deal on a used generator!

posted are

the motor

drum is open

some time.

What about my suggestion in another post that for cheap, he could also use a bathroom scale at the end of a moment arm!

He is thinking about using a torque wrench for calibration, so modest accuracy is apparently ok. I think a bathroom scale at the end of a moment arm would at least have better resolution than a torque wrench (probably automotive type).

Use a regenerative DC motor speed controller with a DC motor (of course) to put the load energy back into the power line. The DC motor's armature current can be used to indicate torque.

Luck to you. John

Thanks for the many safety tips. I am more familiar with electrical safety measures involved with large circuit breaker testing and have seen the effects of electrical power gone astray, but mechanical devices like motors are new to my experience and I don't want to learn the hard way. I'll definitely be careful and I will start with a small prototype.

See other comments below.

Paul

for

of 5

power up

to

I wonder if an AC generator head may be easier to find, less expensive, and useful for other purposes after tmy experiments are complete. I have seen new 10 kW generator heads for about $300. They are designed for small gasoline engines so they probably run at 3600 RPM. I can probably use speed reduction pulleys and belts for the smaller motors which might handle 7200 RPM at 5 HP.

Ideally, I would like to make something like a 24 pole 300 RPM motor (at 60 Hz) and take it to at most 3600 RPM. However, I think I would need a 72 slot stator, which is probably rare in motors under 10-20 HP.

Least expensive would probably be a digital bathroom scale. I'm only looking for about 5% accuracy or so, just to see where efficiency and performance may start to fall off at higher frequencies.

I have a small milling machine that might work for up to 3/4 HP motors. I know people at a local machine shop who might know about a larger scrap machine. Thanks for the safety tips.

for

to

URL's I posted are

of the motor

The drum is open

for some time.

The Prony brake would involve a lot of mechanical setup, whereas I already have a 1 HP DC motor that seems to work OK as a generator. I can also buy a

10 kW AC generator head for about $300 and then later hook it up to a gasoline engine and use it as an emergency power source. It would be easy to control the AC output power into a resistive load (like an old hot water heater) by using a triac or SCR phase fired controller.

I did find several useful threads as a result of the suggested Google search, especially one on motor winding:

Thanks,

Paul

On Mon, 18 Jul 2005 05:20:30 -0400, "Paul E. Schoen"

It is always nice to predict from theory what your experimental results will be. Useful books for understanding the basic theoretical limitations of motors include the McGraw-Hill Electric Motor Handbook and Analysis of Electrical Machines by Richard T. Smith. A math program such as MathCad or better is the only reasomable way to play with motor math models, and should allow you to predict the performance curves expected by your dyno tests with fairly good accuracy.

It may also be instructive to look for recent articles on high efficiency motor designs and to understand the impressive induction motor and inverter design recently developed and patented by Chorus Motors, described in the June 2001 Mechanical Engineering magazine and at

Mechanical water pump?

--
  Keith

something like this maybe?

A friend of mine has build the electronics and software for quite a few dynos used for cars but they are mostly based on inertia but I guess that isn't much use on something that only works at one speed. They are sometimes combined with a telma retarder but thats probably overkill for the power you are talking about.

-Lasse