Are VFD's bad for motors? - Page 2
Close
Login to Your Account
Page 2 of 3 FirstFirst 123 LastLast
Results 21 to 40 of 46
  1. #21
    Join Date
    Dec 2013
    Location
    Vermont, USA
    Posts
    452
    Post Thanks / Like
    Likes (Given)
    6
    Likes (Received)
    112

    Default

    For a low intensity user (home, occasional...) who is concerned about risk of motor bearing pitting from a VFD, one could improvise some sort of very simple motor shaft ground wiper, such as from a piece of annealed/spring-y brass stock riding against the motor shaft - which would be totally no-go in a commercial setting but would accomplish the intended purpose in the low-intensity usage....

  2. #22
    Join Date
    Mar 2009
    Country
    UNITED STATES
    State/Province
    California
    Posts
    4,972
    Post Thanks / Like
    Likes (Given)
    89
    Likes (Received)
    910

    Default

    Quote Originally Posted by Jraef View Post
    Nobody has been using square waves on VFDs now for over 20 years, maybe longer.

    VFDs work by "tricking" a motor into thinking that it is getting an AC sine wave, by taking advantage of the fact that the motor is itself an inductor, and there is a phenomenon called the "inductive time constant" that means that current cannot change instantly in an inductor, it has a specific rate of change that cannot be different. So what a VFD does is to send a "PWM" (Pulse Width Modulation) output of little short pulses of DC into the motor, then before the inductive circuit lets the current rise all the way, it turns it off again, then on, then off, then on then off, at a rate of thousands of times per second. The timing of the length of the pulses and the length of the gaps between them is what controls the actual RMS voltage (Root Mean Squared, a form of averaging) seen by the motor and then how often the pulse strings change from positive to negative dictates the frequency.
    Describes what a square wave is, or a rectangular pulse train. The edges of the pulses have huge spikes. The inductor has to deal with them.

  3. #23
    Join Date
    Mar 2009
    Country
    UNITED STATES
    State/Province
    California
    Posts
    4,972
    Post Thanks / Like
    Likes (Given)
    89
    Likes (Received)
    910

    Default

    Quote Originally Posted by monoblanco View Post
    Thanks, Jraef -
    Good to get a response from someone who has the facts and numbers. I have VFDs on five different motors in my shop, various ages, all at 230 volts, never any problems, nor do I know of any 230 volt VFD user among my associates who has had a problem. I almost consider the subject to be an urban myth.
    Regards,
    Monoblanco
    It's also a urban myth that people kill themselves with their knives and forks.

    Your sample size is too small to form a normal distribution.

  4. #24
    Join Date
    Oct 2017
    Country
    FINLAND
    Posts
    531
    Post Thanks / Like
    Likes (Given)
    86
    Likes (Received)
    158

    Default

    Quote Originally Posted by Gordon Heaton View Post
    I agree that it is something about the VFD. I posted in response to GregSY's comments in post 9, inferring that quick reverse in and of itself is detrimental.
    That's good to hear, because while I don't think I'll be adding any VFDs to my lathe, a DC injection brake module might be on the horizon, just so I can have it stop faster for threading up to a shoulder and similar situations.

    I do plan to get one for my 2x72 belt grinder build though, but as I understand from this thread, the VFD can be configured to run properly with the motor. And also run it in delta mode (230v instead of 400V for the motor I plan to run it with, 2.2kw strömberg motor I got for 20 euros).

  5. #25
    Join Date
    Aug 2010
    Location
    Tucson, AZ
    Posts
    199
    Post Thanks / Like
    Likes (Given)
    6
    Likes (Received)
    49

    Default

    One needs to put in perspective what is considered an "old" motor and the type of duty cycle/load that it will be called to do. On a lathe a VFD replacement is dependent on many factors, but I have installed 100's on lathes built from the 60's to current production with no motor issues running 230VAC. The performance of 3 phase motor's is far better than single phase as too smoothness of the cut and longevity. Since many small users/shops only have single phase here in the US, a VFD on smaller machines is a very good option even on some older motors, although on lathes many people upgrade their motors to an inverter rated. Since you have 3 phase in Finland, no real need to go with a VFD unless there was some additional VFD features you could use.

    Quick reversal is not something you would typically do with threading on a lathe, you cut a thread to a stopping point and then back out the cutter and reverse. A VFD gives very quick and predictable threading stopping position. On the mill installs I set them up with 1 second acceleration/deceleration for threading with an automatic triggered depth position. I setup my installs with various acceleration and deceleration rates that are user selectable. On threading for a lathe I use 1 second braking and thread up to 500 RPM, acceleration I keep at between 3-5 seconds. I use an automated electronic stopping system on the lathe that integrates into the VFD control system and repeats the stopping position typically to +/-0.0002". This takes most of the drama out of threading be it internal/external/to a shoulder. Under normal use I keep the braking at 3 seconds for most manual lathes. If one considers the stresses on a drive system when a motor starts w/o ramped acceleration, it is far higher than will be experienced with a VFD.

    As far as DC injection, this usually plays a more significant role in stopping high momentum systems and works at the lower frequencies, I have not found it to improve the braking rates on 12-16" swing lathes running a VFD. There are a number of programming parameters that are VFD/brand specific that factor into the stopping ability and repeatability of the stopping position.

  6. Likes ptsmith liked this post
  7. #26
    Join Date
    Apr 2008
    Location
    Michigan
    Posts
    37
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    24

    Default

    Jraef;

    Your "layman's" writing style is exceptional.

    I'll always recall your insight on a 1500 HP motor failure I needed direction on.

    John

  8. #27
    Join Date
    Jun 2001
    Location
    St Louis
    Posts
    17,873
    Post Thanks / Like
    Likes (Given)
    1806
    Likes (Received)
    2984

    Default

    The insulation placement on old motors is often better than on newer ones. That usually offsets the margin provided by better insulation types on newer motors, vs the varnished paper or cloth in older ones.

    Adding an inductor will cut the spikes that are applied to the motor. That will take away one source of problems.

    Another issue is that VFD rated motors have a larger "turn-down ratio", they can be run at full current to a lower speed than older motors, due to better heat tolerance and better cooling. That actually may be a much bigger deal than the voltage rating. Older motors saw voltage spikes also, just not as many, and had to stand up to them. If a motor was rated to withstand 1000 V, it usually would withstand quite a bit more than that, because there had to be a margin to allow for the few that were not made quite as well.

    If you have an older motor, you will likely want to limit the amount you slow the motor at full current, to avoid "cooking" it.. Possibly limiting to a 2:1 or maybe 3:1 ratio, where VFD rated motors may be able to and rated to be run considerably slower than that at full current (full torque).

  9. Likes ptsmith liked this post
  10. #28
    Join Date
    Jan 2007
    Location
    Flushing/Flint, Michigan
    Posts
    7,687
    Post Thanks / Like
    Likes (Given)
    383
    Likes (Received)
    6416

    Default

    Quote Originally Posted by JST View Post
    ..
    Adding an inductor will cut the spikes that are applied to the motor. That will take away one source of problems.
    And we are to post 27 before this basic gets mentioned, maybe because JST has some audio background, knows filters and transients.
    More to the OP, yes old motors from the 30's and decades beyond do not like VFDs without a choke.
    It's just a low pass filter that sucks up the spikes and rounds off the corners.
    Bob

  11. #29
    Join Date
    Jun 2001
    Location
    St Louis
    Posts
    17,873
    Post Thanks / Like
    Likes (Given)
    1806
    Likes (Received)
    2984

    Default

    Quote Originally Posted by CarbideBob View Post
    And we are to post 27 before this basic gets mentioned, maybe because JST has some audio background, knows filters and transients.
    More to the OP, yes old motors from the 30's and decades beyond do not like VFDs without a choke.
    It's just a low pass filter that sucks up the spikes and rounds off the corners.
    Bob

    Close...

    The inductor, assuming its value is reasonable compared to the motor inductance, will simply "block" much of the high frequencies. That tends to reduce the peak voltage of the transients. The transients still exist, but the voltage now is SHARED between the inductor and the motor inductance. The plan is to keep the motor's share down to what it can easily tolerate

    Some details that mean inductors are not perfect blockades for transients....

    1) Unless the inductor is carefully wound to cut this effect, there is likely to be a small capacitance right through the inductor, since there are wires wound next to one another, and they have a capacitance from one to the other. So very high frequencies can get right through with much less attenuation. The good news is that the frequencies are very high, and those frequencies are normally not as large in terms of transient voltage, because the VFD does not switch on and off perfectly ( not instantaneously), so the "threat" is lower.

    2) the inductor needs to be made to take the transient voltages itself, or IT will suffer the damage. This is normally OK, since the inductors are a new thing, and are made specifically for this purpose.

    3) the inductor needs to be something reasonable, like a 3% or more inductor at the voltage and current rating. Otherwise ot may not block enough of the transient voltage.

    4) The inductor also needs to be a LOW enough value so that it does not put too much impedance in series with the motor at 60 Hz. There will always be SOME power loss from a series inductor, but you don;t want it to be too much.

  12. #30
    Join Date
    Jul 2007
    Country
    UNITED KINGDOM
    Posts
    4,758
    Post Thanks / Like
    Likes (Given)
    2512
    Likes (Received)
    2309

    Default

    I use an L-C sine wave filter specifically manufactured for use on VFDs. The motors are happy with it, with no whistling noises from the windings.

  13. #31
    Join Date
    Feb 2018
    Country
    UNITED STATES
    State/Province
    Texas
    Posts
    165
    Post Thanks / Like
    Likes (Given)
    30
    Likes (Received)
    28

    Default

    Quote Originally Posted by Mark Rand View Post
    I use an L-C sine wave filter specifically manufactured for use on VFDs. The motors are happy with it, with no whistling noises from the windings.
    What exactly do you have? I'm interested in one.

    I added a TCI dV/dT output filter to my Lagun knee mill with no audible improvement.

  14. #32
    Join Date
    Feb 2018
    Country
    UNITED STATES
    State/Province
    Texas
    Posts
    165
    Post Thanks / Like
    Likes (Given)
    30
    Likes (Received)
    28

    Default

    I've read that a low carrier frequency is easier on the drive (slower switching) but harder on the motor (more winding vibration), and the opposite is true, that a higher carrier frequency is easier on the motor and harder on the drive.

    Our forum experts have stated in many threads, including this thread, the opposite of that. I'm not knowledgeable about VFDs and this confuses me. Maybe someone can make sense of it for me.

    I also read somewhere that bearings are only an issue at higher voltages, 480V and above. At ~ 230V it's not a concern.

    And I thought I'd mention that you can sometimes find dV/dT output filters and load reactors on ebay cheap. I think I paid $25 and $35 shipped for 8A and 12A TCI dV/dT output filters like the one in the photo.
    Attached Thumbnails Attached Thumbnails s-l1600-1-.jpg  

  15. #33
    Join Date
    Sep 2016
    Country
    UNITED STATES
    State/Province
    New Mexico
    Posts
    16
    Post Thanks / Like
    Likes (Given)
    3
    Likes (Received)
    2

    Default

    Quote Originally Posted by ptsmith View Post
    I've read that a low carrier frequency is easier on the drive (slower switching) but harder on the motor (more winding vibration), and the opposite is true, that a higher carrier frequency is easier on the motor and harder on the drive.

    Our forum experts have stated in many threads, including this thread, the opposite of that. I'm not knowledgeable about VFDs and this confuses me. Maybe someone can make sense of it for me.

    I also read somewhere that bearings are only an issue at higher voltages, 480V and above. At ~ 230V it's not a concern.

    And I thought I'd mention that you can sometimes find dV/dT output filters and load reactors on ebay cheap. I think I paid $25 and $35 shipped for 8A and 12A TCI dV/dT output filters like the one in the photo.
    I know from experience designing small motor drivers that switching too fast can blow the H-bridges (I warned the engineer and he did it anyway - made a nice little light show), so I'm pretty much with you on this one. If you switch them too fast the transistors never have time to switch fully on or off, look like resistors, and cook. But I would assume that VFD carrier frequencies are limited to a "safe" value (mine are 15kHz) to keep that from happening. I have mine maxed out so I don't have to listen to them whistle, and haven't had any issues. All this new inverter tech is based on IGBT's, and they've gotten pretty good in the last ten years or so. Seems like they're typically rated for 30khz or higher switching frequencies. Earlier ones were slower.

  16. Likes ptsmith liked this post
  17. #34
    Join Date
    Jul 2007
    Country
    UNITED KINGDOM
    Posts
    4,758
    Post Thanks / Like
    Likes (Given)
    2512
    Likes (Received)
    2309

    Default

    Quote Originally Posted by ptsmith View Post
    What exactly do you have? I'm interested in one.

    I added a TCI dV/dT output filter to my Lagun knee mill with no audible improvement.


    The specific ones I got might not be of any help to you, since they are a German firm, but they were these:- Sinusoidal Filters - REO (UK) Ltd

    You could brew your own from the numbers in their data sheets...

    Edited to add, they've got a US branch.

  18. Likes ptsmith liked this post
  19. #35
    Join Date
    Mar 2009
    Country
    UNITED STATES
    State/Province
    California
    Posts
    4,972
    Post Thanks / Like
    Likes (Given)
    89
    Likes (Received)
    910

    Default

    Quote Originally Posted by ptsmith View Post
    I've read that a low carrier frequency is easier on the drive (slower switching) but harder on the motor (more winding vibration), and the opposite is true, that a higher carrier frequency is easier on the motor and harder on the drive.

    Our forum experts have stated in many threads, including this thread, the opposite of that. I'm not knowledgeable about VFDs and this confuses me. Maybe someone can make sense of it for me.

    I also read somewhere that bearings are only an issue at higher voltages, 480V and above. At ~ 230V it's not a concern.

    And I thought I'd mention that you can sometimes find dV/dT output filters and load reactors on ebay cheap. I think I paid $25 and $35 shipped for 8A and 12A TCI dV/dT output filters like the one in the photo.
    The definition of power is P = V * I. The VFD is switching on and off in the bridge area to create whatever PWM waveform it is operating with.
    During transitions the change is not instantaneous. There is a slope to the voltages and currents. During the switching interval the power is calculated
    as P = integral of V * I over that time. The more time it takes to switch the higher the power, so the drive heats up. A drive with a higher switching
    output can operate at a higher frequency for the same power dissipation.

    Another issue is the spikes at the leading and trailing edges of these square waves. Square waves are not square and probably look uglier in cheap ChiComm drives.
    The spikes can be very much larger than the intended level and have smaller/larger settling times. This has to be put up with by the windings in your motor.
    If the motor is old all it takes is one weak point to fail the motor.

    I use between 8KHz - 10KHz with no filters. Instead I use a quality drive and quality VFD rated motor.

    With cheap drives and/or old motors the chances of something wearing down is greater.

  20. #36
    Join Date
    Jun 2001
    Location
    St Louis
    Posts
    17,873
    Post Thanks / Like
    Likes (Given)
    1806
    Likes (Received)
    2984

    Default

    That's the full deal.... takes away all the noise, and lets through pretty nuch only the sine wave 60(50) Hz. Does not get any better than that

    Those can confuse anything past a regular V/hz control system, though

  21. #37
    Join Date
    Feb 2004
    Location
    peekskill, NY
    Posts
    24,305
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    4449

    Default

    With all the discussion one should not lose sight of the idea that machine tools are not holy relics. They're
    tools made to do a job, and if a part wears out or breaks, it gets replaced along the way. Motors are in this
    catagory of consumable items.

    If the VFD makes the tool work better, use it. If the motor breaks down then either replace it, or get it re-wound
    at a motor shop if it's truly unobtainium.

    Running machine tools is bad for them. Wears them out. So, don't run your machines. They're holy relics, right?

  22. #38
    Join Date
    Feb 2004
    Location
    peekskill, NY
    Posts
    24,305
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    4449

    Default

    Obligatory story:

    Years ago I purchased a mint hardinge split bed lathe from a shop that had gone into receivership. The shop made
    wire drawing dies, and had been in operation since the mid 1920s as far as I could tell. The previous owner died, and
    the new owners were selling and scrapping the machines.

    While I was there collecting the goods, I wandered through the shop. There was an aisle behind the shop, full depth
    of the building. It was lined for its length with used motors, stacked about four feet high. That's how many motors
    were used up in the running of that shop - the owner had it in his mind (I guess) that he would get them rewound or
    sent out for scrap money. He expired before the motors were ever sent on to their ultimate destination.

    As an aside, the shop also had every typewriter ever used in the office, and three TV sets of various vintage in the office.

    This proved to me once and for all, motors are just consumable items. Use 'em up, wear 'em out.

  23. #39
    Join Date
    Aug 2014
    Location
    bainbridge island
    Posts
    1,136
    Post Thanks / Like
    Likes (Given)
    217
    Likes (Received)
    267

    Default

    Of about 8 hvac inverter fan motors I've acquired, one had a fault due to phase to phase short, one was debonded magnets the rest were inverter failures.

    The short occurred about half way from the inverter to the Y neutral, where the wires cross over each other. Care could be taken to ensure the wires do not touch, but these are mass produced machine wound motors.

    Where the two windings crossed over like an X the peak voltage would have been about half what the drive normally produces due to the distributed nature of the windings, or.. maybe even a voltage spike randomly occurred there due to various resonances of the inter-turn capacitance, phase to ground capacitance, leakage inductance of each coil.

    Each phase is usually 4 hair pin coils in series, connected in Y. The inverter is directly connected to the back of the motor, so there are no long wire leads.



    Regarding sine wave filters, someone from Costa Rica pm'd me regarding a sine wave filter he built, using a 3% automation direct line/load reactor. It got too hot, but still worked. The laminations appeared insulated and were .020" thick.

    I bought an MTE RL-00802 reactor (3mH 8 amp rated) and with and without capacitors, running a 1/2hp bench grinder at 240v 65Hz the wattage was 100 vs 86 watts. that particular reactor is rated for something like 14 watts dissipation at full load so yes it would run hotter with capacitors.. but nothing like the 100C temperature that the man was reporting from a 3% automation direct reactor (Running a 1/2hp well pump from a 1hp reactor and drive).

  24. #40
    Join Date
    Jun 2001
    Location
    St Louis
    Posts
    17,873
    Post Thanks / Like
    Likes (Given)
    1806
    Likes (Received)
    2984

    Default

    Quote Originally Posted by johansen View Post
    Of about 8 hvac inverter fan motors I've acquired, one had a fault due to phase to phase short, one was debonded magnets the rest were inverter failures.

    The short occurred about half way from the inverter to the Y neutral, where the wires cross over each other. Care could be taken to ensure the wires do not touch, but these are mass produced machine wound motors.

    Where the two windings crossed over like an X the peak voltage would have been about half what the drive normally produces due to the distributed nature of the windings, or.. maybe even a voltage spike randomly occurred there due to various resonances of the inter-turn capacitance, phase to ground capacitance, leakage inductance of each coil.

    ....
    And that is an example of the insulation placement difference. High voltage spikes create corona voltages in susceptible spots, and that gradually breaks down insulation. Eventually something goes "pop".

    Well-placed insulation may degrade, but it is a slower process. Gives time for planned/scheduled maintenance, so that the motor you have no replacement ready for does not go "pop"on Saturday just before the overnight shift when you have a big order to get out.


Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •