Testing with VFD

[EnigmaPaul]

Hi Folks -

I need to setup a VFD in a workshop to test the accuracy of my AC voltage and current metering on the line side. VFDs often disrupt the measurements of RMS due to harmonics, conducted, and radiated noise.

I'm going to get a low cost PWM VFD and need to apply a load to it.

I'd like to use a motor to most accurately simulate an inductive load but I was wondering:

1) Can I connect a resistive load bank instead so I can control the current? What can I use for an inexpensive 3-phase load bank?

2) If I use a motor, is it possible to just use a simple friction brake to load down the motor? Where would I get a suitable brake without having to make one?

Thanks!

 

[Fred Abse]

Hi Folks -

I need to setup a VFD in a workshop to test the accuracy of my AC voltage
and current metering on the line side. VFDs often disrupt the
measurements of RMS due to harmonics, conducted, and radiated noise.

I'm going to get a low cost PWM VFD and need to apply a load to it.

I'd like to use a motor to most accurately simulate an inductive load but

A motor is nowhere near an inductive load.

I was wondering:

1) Can I connect a resistive load bank instead so I can control the
current? What can I use for an inexpensive 3-phase load bank?

Depends on the design of the VFD, as to whether you can use a purely
resistive load. Modern VFDs monitor the current in the phases (usually
two, and derive the third), then do some quite sophisticated DSP math, to
arrive at motor excitation. They expect to see a real motor.

2) If I use a motor, is it possible to just use a simple friction brake to
load down the motor? Where would I get a suitable brake without having to
make one?

Any brake (dynamometer) has the problem of dissipating the energy it
absorbs, usually as heat. A frictional brake will get very hot.

 

[EnigmaPaul]

A motor is nowhere near an inductive load.







Depends on the design of the VFD, as to whether you can use a purely

resistive load. Modern VFDs monitor the current in the phases (usually

two, and derive the third), then do some quite sophisticated DSP math, to

arrive at motor excitation. They expect to see a real motor.









Any brake (dynamometer) has the problem of dissipating the energy it

absorbs, usually as heat. A frictional brake will get very hot.

Yes, but the testing would only be for short periods of time.

 

[RobertMacy]

Yes, but the testing would only be for short periods of time.

We used to load a motor by driving a generator and then power a radiant
room heater with that, or that type of load, you know the giant ceramic
resistors that can take 100W each.

By loading the generator it was easy to vary the load on the motor.

 

[Neon John]

1) Can I connect a resistive load bank instead so I can control the current? What can I use for an inexpensive 3-phase load bank?

No. Three phase VFDs look at the back-EMF as part of their
determination as to when to step to the next phase rotation.

2) If I use a motor, is it possible to just use a simple friction brake to
load down the motor? Where would I get a suitable brake without having to make one?

I did that for a Science Fair project many moons ago. The research
was to determine the effects of various porting techniques (2-stroke)
on power output. The absorber was a disk brake from an Opel with a
hole bored in the back on axis with the spindle and a 1" shaft welded
in. This assembly was mounted on pillow blocks and equipped with a 1
ft long arm that pressed on a scale.

That is NOT the way to do things today. The cheapest absorber that I
know of that you can make without a machine shop is an ordinary
induction motor (squirrel cage) with DC applied to its input
terminals.

This causes the motor to become a generator with the electricity
induced into the rotor's shorted bars. Almost all the absorbed energy
is dissipated in the rotor so some good forced cooling is necessary.

You can easily demonstrate this effect. Pick up any old AC motor
(even a house fan) and put a 9 volt battery across the AC in. Observe
how difficult the motor is to turn.

Not too many years ago I build a dyno using this method that used a
5hp single phase motor as the absorber. I unbolted the foot and
mounted the motor body between 6 ball bearings, 3 at each end. A 1 ft
long (from the shaft centerline) pressed on a load cell that fed my
data acquisition system. A simple Variac and rectifier varied the DC
voltage. An optical pickup registered the shaft speed. Once you have
RPM and torque in ft-lbs, hp = (torque * RPM)/ 5252.

With a high capacity blower blowing air into one end bell and through
the rotor, this absorber could handle 50 HP for several minutes. I
used it to test small EV motors.

Check with your local electric motor shop. It's not that uncommon to
have a phase open up on a 3 phase motor and burn out 1 or more of the
windings. Chinese made motors are too cheap to rewind so they get
discarded for their scrap value. You can probably pick up such a
motor for the scrap value. One winding should be enough, though 2
would be better.

John

John DeArmond
http://www.neon-john.com
http://www.fluxeon.com
Tellico Plains, Occupied TN
See website for email address

 

[josephkk]

Hi Folks -

I need to setup a VFD in a workshop to test the accuracy of my AC
voltage and current metering on the line side. VFDs often disrupt the
measurements of RMS due to harmonics, conducted, and radiated noise.

I'm going to get a low cost PWM VFD and need to apply a load to it.

I'd like to use a motor to most accurately simulate an inductive load
but I was wondering:

1) Can I connect a resistive load bank instead so I can control the
current? What can I use for an inexpensive 3-phase load bank?

2) If I use a motor, is it possible to just use a simple friction brake
to load down the motor? Where would I get a suitable brake without
having to make one?

Thanks!

Each to their own. Just the same i would buy, assemble, or rent a
motor / brake assembly. That way you could test its performance over the
range of loads from no load to locked rotor.

?-)

 

[John S]

Hi Folks -

I need to setup a VFD in a workshop to test the accuracy of my AC voltage and current metering on the line side. VFDs often disrupt the measurements of RMS due to harmonics, conducted, and radiated noise.

I'm going to get a low cost PWM VFD and need to apply a load to it.

I'd like to use a motor to most accurately simulate an inductive load but I was wondering:

1) Can I connect a resistive load bank instead so I can control the current? What can I use for an inexpensive 3-phase load bank?

2) If I use a motor, is it possible to just use a simple friction brake to load down the motor? Where would I get a suitable brake without having to make one?

Thanks!

About 20 years ago I worked at a company who produced AC and DC motor
speed controllers (polyspede.com). Some of our DC drives were
regenerative. To test VFDs, we had DC motors mechanically coupled to the
AC motors and used the DC regenerative drives to return the energy from
the DC motor to the power line. It was possible to vary the amount of
regeneration so as to control the load on the AC motor. BTW, just
because you return the energy to the line doesn't mean that the power
factor is good.

It is also possible to do the same with two coupled AC induction motors,
but, unless times have changed quite a bit, you will probably not find
many regenerative AC drives that return the energy to the line. Most of
them burn up the energy with some sort of load on the bus (called
dynamic braking, back then). Ours had a regenerative bridge.

For our customer, the VFD had to drive an induction motor which was
coupled to a hydraulic motor. Then, when the hydraulic motor drove the
induction motor, it had to load the hydraulic motor. The amount of
energy meant that it was prohibitive to just dissipate (I think it was
about 100Hp or so). That's the reason we put it back into the line even
if the power factor was not good; it gave it a place to go.

Just thought for you. Good luck.

 

[Fred Abse]

Not too many years ago I build a dyno using this method that used a
5hp single phase motor as the absorber. I unbolted the foot and
mounted the motor body between 6 ball bearings, 3 at each end. A 1 ft
long (from the shaft centerline) pressed on a load cell that fed my
data acquisition system. A simple Variac and rectifier varied the DC
voltage. An optical pickup registered the shaft speed. Once you have
RPM and torque in ft-lbs, hp = (torque * RPM)/ 5252.

That's basically what I'd do, except I'd use a shunt wound DC motor as a
generator. Generated voltage into a fixed resistive load, control by
varying DC field current. The hot load can go somewhere outside, where it
won't cook the operator. There must be lots of redundant DC shunt motors
lying around, off retrofitted machine tools, cranes, etc.

 

[Neon John]

That's basically what I'd do, except I'd use a shunt wound DC motor as a
generator.

The problems with that method are a) you have to come up with some
method of converting the electricity to heat and disposing of it.
That's a non-trivial or at least non-inexpensive proposition for
higher powered motors.

b) The armature, being solid, has no real way to be cooled and so it
can't handle current much over its rating for very long. The rotor on
an AC motor, being perforated with cooling holes, CAN handle the
overload. All that is required is more cooling air.

The beauty of the induction motor-as-a-load is that all it requires is
a relatively small Variac, a bridge rectifier and a blower for the
absorber.

John

John DeArmond
http://www.neon-john.com
http://www.fluxeon.com
Tellico Plains, Occupied TN
See website for email address

 

[Fred Abse]

The problems with that method are a) you have to come up with some method
of converting the electricity to heat and disposing of it. That's a
non-trivial or at least non-inexpensive proposition for higher powered
motors.

The energy, or most of it, goes into the *load*, which can be fan- or
water-cooled. A resistor bank, external to the generator, and even the
building, if you want.

b) The armature, being solid, has no real way to be cooled and so it can't
handle current much over its rating for very long. The rotor on an AC
motor, being perforated with cooling holes, CAN handle the overload. All
that is required is more cooling air.

The armature will generate no more internal heat than it did running as a
motor at full load. The current is the same, motoring, or generating. The
armature resistance is the same. If it didn't need cooling as a motor, it
won't as a generator. Those really big ones that did had a snail fan built
on.

The rotor of an induction motor is solid, too. Sometimes having cast-in
aluminum bars.

The beauty of the induction motor-as-a-load is that all it requires is a
relatively small Variac, a bridge rectifier and a blower for the
absorber.

*All* the absorbed energy has to be dissipated in the rotor bars.

 

[Lasse Langwadt Christensen]

Fred Abse brought next idea :










In the past I have seeen tanks of water (outside of course) useed as

dummy loads just with plates hung in them and salt in the water (added

carefully)

One installation the alternator was 250 KVA so it was quite big tank

although it may never have taken the full load. l-)

I guess if had to do something like that all day you could get a grid tied inverter and stuff it back into the grid

I know a guy that has a "rolling road" in the garage for tuning cars, most of the load is inertia from the big rotating drum but it also has a brake
afaict it is a rotating electro magnet between two water cooled iron discs

The cooling water is actually plumbed in to the house water heating


-Lasse

 

[Phil Allison]

"Enigma Paul = PITA "

I need to setup a VFD in a workshop to test the accuracy of my
AC voltage and current metering on the line side.

** Is that a VFD that operates from single phase power?

VFDs often disrupt the measurements of RMS due to harmonics, conducted,
and radiated noise.

** Only if there is no filter electro after the rectifier - only possible
with VFDs that take in 3 phase power.

I'm going to get a low cost PWM VFD and need to apply a load to it.

** So it IS a *single phase input* VFD with 3 phase output.

The input current wave will be just like with any rectifier & capacitor
oad - ie 120Hz half sine shaped pulses.

The problems you anticipate only happen when monitoring on the MOTOR side.



.... Phil

 

[Fred Abse]

In the past I have seeen tanks of water (outside of course) useed as dummy
loads just with plates hung in them and salt in the water (added
carefully)

They used to use salt water rheostats to control the starting rotor
current of big 3-phase slip-ring induction motors. Steaming by the time
the motor was up to speed and the rheostats were shorted out with the
blades at the bottom of the tank.

 

[josephkk]

The armature will generate no more internal heat than it did running as
a motor at full load. The current is the same, motoring, or generating.
The armature resistance is the same. If it didn't need cooling as a
motor, it won't as a generator. Those really big ones that did had a
snail fan built on.

Um. that is not the case when being forcibly turning way slower than
rated speed. Especially not during "locked rotor" testing.

The rotor of an induction motor is solid, too. Sometimes having cast-in
aluminum bars.



*All* the absorbed energy has to be dissipated in the rotor bars.

Well at least somewhere in the motor. Much can be dissipated in the
stator windings as well.

?-)

 

[Fred Abse]

Um. that is not the case when being forcibly turning way slower than rated
speed. Especially not during "locked rotor" testing.

What I was describing is a DC motor, used as a generator, into a fixed
resistive load, and it's torque reaction controlled by controlling the
field excitation. With the right choice of load resistance, and a limit on
the maximum field current, it should be impossible to exceed either
armature rated current or voltage.

If you want to try to lock it with a dead short, you're on your own.

 

[josephkk]

What I was describing is a DC motor, used as a generator, into a fixed
resistive load, and it's torque reaction controlled by controlling the
field excitation. With the right choice of load resistance, and a limit
on the maximum field current, it should be impossible to exceed either
armature rated current or voltage.

If you want to try to lock it with a dead short, you're on your own.

Not so much the DC motor but the induction motor fed by the VFD (which
must be able to cope for at least a little while). Meanwhile it is
feeding tons of hash back into the mains which need to be measured.
Lots of higher than mains frequency stuff there. Measuring all useful
values while this is occuring is part of OPs question the way i see it.

?-)