A rudimentary filter and HP boost for VFD motors. Thinking out of the box

I often find myself in the position of trying to solve a problem or complete a project with a budget of small to none.
I was trying to come up with an economical reactor for VFD motor leads.
I thought of using transformer windings and this train of thought led to other thoughts.

  1. A transformer makes a good VFD filter but it is expensive.
  2. How about using a low voltage secondary winding of of a transformer as a reactor.
    This will probably cost more than a reactor and is wasteful of material.
  3. How about combining the two ideas.
    If we use a three phase auto transformer and boost the voltage, we have the advantage of the low voltage winding acting as a filter, and with extra voltage available at the motor, we can go somewhat above the supply frequency before we hit the Volts per Hertz limit.
    10 HP, 208 Volt motor.
    With a 120:24 Volt autotransformer between the drive and the motor, we can boost the 208 Volts to 250 Volts.
    Now we can maintain our 3.47 Volts per Hertz ratio up to 72 Hz.
    And at 72 Volts we now have 12.4 HP out of a 10 HP motor.
    I am interested in comments as to the effectiveness of the filtering.
    The extra HP will often be only an added safety factor.
    And the HP boost helps to justify the cost of a transformer.

Another advantage of higher voltage at the motor may be more torque boost available to start sticky loads.
I am not sure what may happen when we try to start a treadmill with a horse standing on it.
We may need more boost.
The original owner said that the horses declined to enter the box with the conveyor running.
No kidding!
He said that a DOL start with a horse on the treadmill did not work well either.
from his body language I thought that it may not be the time to ask what happened.
He is now using a tread mill with a vfd drive.
There are lots of these in service but not published information.
Well, lots of blurbs hyping the advantages but almost no technical information at all.
We’re up and ready to find out as we go along.
Stay tuned.

Will the higher initial voltage cause problems with ‘burn out’? or require a larger size conductor to reduce burn out?

Not my area…


In regards to the voltage at the base frequency, in North America standard induction motors are typically wound with conductors insulated for at least 480 Volts.
A typical motor will be rated for 230/460 Volts, so an increase from 208 Volts to 250 Volts is well within the capability of a standard motor.
The other limit on motor voltage is magnetic saturation.
If the voltage is too high the magnetic circuit will saturate and act as an air core inductance rather than an iron core inductance.
The ratio of the inductive reactance between an iron core inductance and an air core inductance is around 50,000:1
Once a magnetic circuit is pushed into saturation the excess voltage will cause excess current and burn out will be rapid. Possibly in minutes.
Fortunately the inductive reactance is frequency related and as the frequency is increased the knee of the saturation curve is raised.
250 Volts on a 208 Volt motor at 60 Hz may cause rapid burnout.
250 Volts on a 208 Volt motor at 72 Hz is fine.
The motor will run faster.
The motor will safely generate the same torque at the rated current.
The same torque at more RPM means more HP.
The issues with burn out of VFD driven motors are related to the high frequency switching and rapid rise times of the pulse width modulation.
This generates a lot of harmonics that can cause high transient voltages.
Hence inverter rated motors and filters for non inverter rated motors driven by VFDs.
Who was it that said;
“Not that the answer has to be long but it will take awhile to make it short.”
Mark Twain?
The above is fairly standard VFD and motor theory.
What is not standard and what I am hoping for comments on from the gurus is the filtering action of the 24 Volt winding of an auto transformer.
That’s the part that is “out of the box”.
Thanks for asking.

I think it’s a splendid idea - if it works.

Are you thinking the usual buck/boost transformers? I know it will grate on you to be using three instead of two. LOL

And, the initial intent is to just scrub off the sharp edges?

In this case, I’ll take three transformers. grin
I was on a SAGD site some years back.
They were using down-hole submersibles to pump the heavy oil.
The pumping rate is critical and VFDs are used.
There may be hundreds or even thousands of feet between the VFD and the motor.
On the same skid, some wells used 480 Volt VFDs and then transformers to boost up to 4160 Volts.
Other wells on the same skid used 4160 Volt VFDs and a large filter.
I know that the filtering will be much less with a 24 Volt winding compared to a 480/4160 Transformer.
I am wondering if the small transformer will be useful at all or just a waste of time, other than the small HP boost.
Wondering, wondering.

The problem is these are auto-transformers so the galvanic connection can probably still pass sharp corners. 1:1 isolation transformers would probably be much much better at edge carving. But unless you’re going to play the higher horsepower gambit it’s probably smaller and cheaper to just go with OTS reactors.

I mean lookit the prices and footprints:
3 Phase reactors

1 Like

I’m starting to realize that this is a good idea going nowhere.
Comparing a conventional transformer with an auto transformer:
Conventional transformer, 120 Volts in, 144 Volts out.
We have the reactance of a 120 Volt winding plus the reactance of a 144 Volt winding to smooth the sharp corners.
Auto transformer, we have only the 24 Volt winding effective as a filter.
There will be some second order effects from the 120 Volt winding. (or not)
I am afraid that I have discovered the weakness in my great idea. grin

Not what you’re asking, and probably outside the scope due to size, but I’m aware of at least one inverter manufacturer that requires either a filter on the output of their inverter, or uses a transformer to provide both isolation and filtering in the one unit. They do have specific criteria for their transformers to be able to handle the higher harmonic content, and they also still need a reactor as part of the filter arrangement.

Thinking further about it, as they’re motor drive VSDs adapted for energy storage applications, they likely need the filter as there’s no motor load on the end of the drive, so they need to compensate somehow.