What's new
What's new

VFD's "burn up" a motor?

sandiapaul

Titanium
Joined
Mar 19, 2004
Location
Princeton, NJ USA
I have a Wade 8a lathe with it's original early 1950's 220 3 phase 1.5 HP motor. As you might imagine it is a big motor. I run my shop with a large RPC and will wire the Wade up to it soon. I'd love to have variable speed on the lathe so I was looking around for a way to have the machine running on my RPC and get a vari speed option. I landed on a website, https://vfds.com/ and a chat window popped up. I told the person exactly what I wanted to do and he told me a VFD would "burn up" an old motor.

1. Is this true?
2. What is the easiest way to get a vari speed on my old lathe? I really want to keep my old original motor on the machine...

Thanks!
 
Likely not. Might depend on whether it is an "antique" motor, or just a decent 1950s type.

In general, the usual sine wave type VFDs will not do damage. And, the insulation in them is often better than the newer ordinary "slammed-together and sold" motors. So probably no damage whatsoever.

Some old motors can have a lot of "iron losses" that may heat the motor more than expected. And motors with old wound cloth wire insulation are likely old enough to be more of a risk.

People talk about "burning up" motors, but I have never seen it, nor heard of it actually happening just due to use of a VFD. There are regular motor problems that may lead to toasting a motor that happens to be on a VFD, of course.
 
Agree with JST.
You hear you need an inverter duty motor or smoothing chokes in the line.
Never burnt one up or have seen such problems and I have motors on machines with WWII tags on them.

What I have seen go bad is people take a 60hz motor and try to run it up at 400hz. BOOM. This is a mechanical problem.
My limit is 90-120 and even there I worry that this is really pushing my luck.
The other side is that on a fan cooled motor at much lower speed the fan does not do much so you overheat it and melt the guts.

VFDs from the 70's and early 80's were kind of brutal in their output so I think this is where this comes from.
Weird thing here is that they would sometimes eat motor bearings.

Bob
 
Last edited:
The phenomenon that can damage motors is related to how a VFD functions to "trick" a motor into thinking it is getting a lower voltage along with the change in frequency using a technique called "PWM" (Pulse Width Modulation). Essentially the motor is getting a string of high speed DC pulses that it cannot react to very fast, so the "average" voltage the motor can reactor to is manipulated by changing the width of the pulses and the width of the gaps between them. In doing this, the wires and the difference in impedance (a type of resistance for AC coils) can cause reflections of voltages that cycle back and forth between the VFD and the motor terminals, building up like ripples in a pond crossing each other, until spikes of voltage begin to exceed the insulation value of the motor winding insulation.

For the most part though, this is a bigger problem for older 460V motors, where the "normal" insulation value was 2x the peak voltage of the motor design, so a lot of older 480V motors had 1000V or maybe 1200V insulation, whereas the spikes can reach 1400V or more. With 230V motors, the motor mfrs often use the same winding insulation because it's too expensive to carry two different types, so the 230V motors end up with a higher than necessary insulation level, which helps out when using a VFD.

If the motor is "irreplaceable", you can add a filter to the output of a VFD. That filter may cost as much as the VFD itself though, so you have to REALLY want to keep that old motor.The other strategy is to get whatever life you can out of the motor you already have, then if it smokes, buy a new "Inverter Duty" motor to replace it.

The other thing to be concerned with on older motors is what's called the "turn down ratio", which is a concept that didn't exist in 1950, so you likely can't get that info. It has to do with the motor's ability to cool itself if it is a fan cooled motor. So TEFC or ODP motors that have shaft mounted fans are often at risk of overheating because when you slow down the motor, you slow down the cooling fan too. TENV motors are not an issue, but there were not a lot of TENV motors in 1950. So to be safe, assume no BETTER that a 2:1 turn down ratio on an old motor, meaning never run it below 50% speed.
 
Thanks for the replies...if I do it I guess I sort of roll the dice. The original motor is not so precious I guess. I'm the second owner(I'm pretty sure, not positive) and the first was MIT so I'd kinda like to keep it original as possible. Of course now that I said that someone will ask why don't I just stick with the original drive setup? :)
 
The phenomenon that can damage motors is related to how a VFD functions to "trick" a motor into thinking it is getting a lower voltage along with the change in frequency using a technique called "PWM" (Pulse Width Modulation). Essentially the motor is getting a string of high speed DC pulses that it cannot react to very fast, so the "average" voltage the motor can reactor to is manipulated by changing the width of the pulses and the width of the gaps between them. In doing this, the wires and the difference in impedance (a type of resistance for AC coils) can cause reflections of voltages that cycle back and forth between the VFD and the motor terminals, building up like ripples in a pond crossing each other, until spikes of voltage begin to exceed the insulation value of the motor winding insulation.

For the most part though, this is a bigger problem for older 460V motors, where the "normal" insulation value was 2x the peak voltage of the motor design, so a lot of older 480V motors had 1000V or maybe 1200V insulation, whereas the spikes can reach 1400V or more. With 230V motors, the motor mfrs often use the same winding insulation because it's too expensive to carry two different types, so the 230V motors end up with a higher than necessary insulation level, which helps out when using a VFD.

If the motor is "irreplaceable", you can add a filter to the output of a VFD. That filter may cost as much as the VFD itself though, so you have to REALLY want to keep that old motor.The other strategy is to get whatever life you can out of the motor you already have, then if it smokes, buy a new "Inverter Duty" motor to replace it.

The other thing to be concerned with on older motors is what's called the "turn down ratio", which is a concept that didn't exist in 1950, so you likely can't get that info. It has to do with the motor's ability to cool itself if it is a fan cooled motor. So TEFC or ODP motors that have shaft mounted fans are often at risk of overheating because when you slow down the motor, you slow down the cooling fan too. TENV motors are not an issue, but there were not a lot of TENV motors in 1950. So to be safe, assume no BETTER that a 2:1 turn down ratio on an old motor, meaning never run it below 50% speed.

No argument with that, load inductors aren't necessarily that expensive though - I wound a set on three matching old 24v control transformer cores, I'm like that - once you determine what inductance you need (and can measure it! Credit-card sized LCR meters are stupidly cheap now!) from the "percentage". Commercial offerings are usually between 3% and 6%, this is the proportion of motor-supply AC voltage that gets dropped across the inductor - so e.g. in my case 415v supply, I want between 12 and 24v *at supply frequency*, calculate the impedance, calculate the inductance, wind and measure the coils then encapsulate in epoxy! It's sensible though to "gap" the laminations, so an E-I stack gets de-interleaved leaving a stack of each, and a PTFE or similar shim goes between to break the magnetic circuit and reduce/prevent core saturation. The impedance of the inductor is *way* higher at the VFD switching frequency, so what's delivered to the motor has much lower dV/dT, which reduces the risk of insulation breakdown.

Cooling... I have a big (10") mains Papst fan and DIY shroud on the end of my motor, forcing air through it! Most VFDs can be set up with a relay closure output above/below "supervisory frequency", I use that to switch the fan off above e.g. 30Hz (I run as low as 10Hz) - when the motor's own fan starts moving air - so it's cooled form stationary, works for me and if there's space a pretty simple and useful mod. I've seen a lot of hobbyist lathes and mills with DC variable-speed motors fitted with PC fans on the end, similar issue as they get low speeds by slowing the motor to the point that air's not moving through, but not switched, running fulltime!

Dave H. (the other one)
 
Thanks for the replies...if I do it I guess I sort of roll the dice. The original motor is not so precious I guess. I'm the second owner(I'm pretty sure, not positive) and the first was MIT so I'd kinda like to keep it original as possible. Of course now that I said that someone will ask why don't I just stick with the original drive setup? :)
Open up the oldie and clean it up. Maybe replace bearings. Coat windings with varnish. Make everything sanitary. I did this to my oldie.
 
You want to wind the inductor so that the first turns are far from the last turns, and ideally the winding is broken up in sections. You want the least capacitance between the input and output, so that there is less way for the fast rising pulses to couple through and bypass the inductor.
 
There are limitations due to the age of the motor, in particular the speed range is limited to something like 30-60 Hz, below that the performance will most likely fall of and heating will become more an issue due to the increased work load, you are only producing 0.75Hp and 1/2 the torque due to the loss of the mechanical advantage being lost. Typically cooling issues with TEFC motor is more of an issue below 15Hz and/or continuous load, the older motors have a lot of mass so intermitant load is unlikely to cause over heating. You still may have isolated hot spots and further insulation damage if pushing the lathe hard below 60Hz. You can use a dV/dT output filter like the TCI (V1K6A01 , KLC8BE), you can pick one up for very little either used or NOS. The dV/dT filters have limitations to the maximum carrier frequency and need to be sized for the VFD output amps. You want to keep the VFD carrier frequency lower, which may cause more annoying whine from the motor, which increases as you lower the frequency. Some VFD's vary the carrier frequency so it presents more as a hum as opposed to a whine. Using the original motor, I would keep the VFD overload parameters a bit less then factory defaults which is usually 150% for 1 minute, so something like 130% with a carrier frequency of 4kHz. Alternative is to remove the factory motor and replace it with a newer 2 or 3Hp model with the VFD, and keep the old motor if you want to restore it back to origional spec. This will give you better performance over a much wider motor operating range like 20-90Hz. A newer TENV or TEFC inverter motor would cost about the same as a dV/dT filter.
 
" he told me a VFD would "burn up" an old motor...."

I belive you've been misinformed. As mentioned keep the lower speed above 30 cps and I doubt you'll have problems. For many here an 'older motor' was built in the 20s....
 
Many have solved cooling issues with an auxiliary fan ducted into the motor, and separately powered.

Given sufficient airflow, that will really help as far as increased "turn-down" ratios.

A bigger issue is faster speeds. Old motors were not made to withstand higher rpm, and you run the risk of "birdcaging" the rotor if you over speed them. That depends on motor design, and the type of rotor conductors, etc.
 
VFDs from the 70's and early 80's were kind of brutal in their output so I think this is where this comes from.
All VFD's have some pretty abusive outputs though I'm sure the modern ones have gotten a lot smoother than the ones from the 70's and early 80's used to be. There's a reason we used to have a whole lot more DC motors and drives in variable speed applications.

Its simply how the VFD's are approximating an AC sine wave - instead of having a smooth sine wave like you would get from a generator spinning around, you rectify AC input power to a DC bus, and then use high speed switching (IGBT outputs with very fast controllers) to approximate the sine wave that you are aiming for. Unfortunately that means a lot of aggressively dV/dT and a bunch of sharp little square steps that resemble or average out to more or less the smooth wave you'd get from a generator.

Between that and the ability to dramatically under or overspeed a motor, coupled with insulation and cooling facilities that may or may not be up to handling that abuse and heat well, you can absolutely kill a motor with a VFD, up to and including ones that have the "inverter duty" label on the dataplate. Lower duty cycles and oversizing motors and the cooling for them if you expect to operate very far away from 60Hz are generally good practice if you want to avoid cooking an otherwise good motor.
 
Agree with comments by Jraef. . . not likely to be any kind of problem with a 230V drive and motor and if concerned, keep the motor leads short and use a load reactor. The insulation on an older motor would have to be pretty degraded to be problematic with a 200V class drive.

I have seen many 460V motor failures on VFD’s with long (300+ ft) cables and high switching frequencies set on the drive. This is when an output filter or load reactor can keep the reflected waveforms and associated pulses under control. Some were winding turn to turn short circuit failures, others had issues with bearing fluting and eventual failure.

Also in low atmosphere conditions, coronal arcing can take motors out in a similar manner.

This article has a few good graphics on inverter generated sine waves and what they look like to the motor.

 
I have a Wade 8a lathe with it's original early 1950's 220 3 phase 1.5 HP motor. As you might imagine it is a big motor. .... I landed on a website, https://vfds.com/ and a chat window popped up. I told the person exactly what I wanted to do and he told me a VFD would "burn up" an old motor.

That's a blanket falsehood. I bet the next thing the chatwindow was going to do, was offer you something else... like a 'perfect motor'.

As the guys suggested, there are 'some things' that can result from 'some motors' and 'some VFD settings', but you're never gonna see that happen in your circumstance. The US made industrial-grade motors made in the '50's were the best of the best, hands down. There's modern 'inverter duty' motors made that have some 'design features' that help suppress some of the bad things that SOMETIMES happen, but your Wade's motor is not going to suffer from anything like that, especially if you're using it as a personal-use (non industrial) environment.

The worst thing a guy could do, is run a VFD'd motor at an extremely low speed, without some sort of constant flow cooling. Take the mechanical fan off the shaft, and make a shroud for the motor, and install a constant-speed motor that's running anytime the VFD is powered up. My 10EE has a mid '50's Allis-Chalmers 1800rpm motor that I nixed the shroud, and made a new one (just happens a 5-gallon plastic bucket fit perfectly), and a 10" high-powered computer enclosure fan flows through that motor wonderfully.

I run the spindle very slow... all the time... 3:1 reduction, toothed belt drive, and no problems. Same with the setup on my Johannsen radial drill, my Bridgeport BRJ, my Whiteman power trowel, my Cinci #2 horizontal... I've got VFDs on my tool grinders, one of my air compressors, a table saw, a very large bandsaw... and soon... a 20hp electric sawmill... not a single one of 'em has a 'special' motor.... and none are 'burned up'.
 
Dave,
I kinda figured that. I was installing a drive belt to the spindle this weekend and spent a lot of time looking at the motor. It's freakin gigantic, a Peerless. The motor is 2 speed reversing. In low speed it is nearly dead silent. Once I get the stock drive sorted I can later look at a VFD. Right now it's just cool to see it turn. I started the restoration in 2005!
 
Those old motors have insulation where it needs to be, placed there. New machine assembled motors are almost literally "slammed together".

I've tested old and new for corona discharge. Old motors stand much more voltage in general before you get detectable corona.
 
I run my Bridgeport J crazy slow fairly often on the vfd with the belt in 2nd highest spindle rpm; well less than half motor speed, to the point where its easily stalled- some jobs don't need backgear or belt change. I've never even felt it get warm. Sometimes I run the Nichols that slow also... though thats coupled with belt changes for the torque- similarly no signs of distress at all. The Nichols has a fancy 1hp vector drive vfd w/ 3/4 hp motor- the Brigeport has a funny but effective KB Electronics vfd I bought from Plaza Machinery a million years ago.
 
Last edited:
Greg!

Your KBVF still going strong? I got mine in one of the tool grinders, on account that it's small enough to tuck inside the pretty cast base. I put a speed pot through an existing hole in the original steel front panel. I won't say the KB has the jump-and-run of the AB1305 or 1336's, but tiny, handy, and means the 1/2hp 230v 3-ph will run off a 120v wall outlet... ;-)
 
Hiya Dave,

Yep- working fine, haven't been inside the box for probably 10 years.. mine is 240v, set up to run the J head or the shaper head via a toggle switch, speed control via a pot also on the box. Its a handy design- no parameters to set.. though clearly can't be tailored as precisely but plenty good for a Bridgeport. I was just running the spindle at something 20-ish rpm tonite, really slow for an easy stall running a small taper reamer- worked :)
 








 
Back
Top