Do I need a magnetic contactor with a VFD?

  1. skotl

    skotl Forum Supporter

    Messages:
    6,483
    Location:
    Edinburgh, UK
    I've been following Clough42's videos on sorting out a VFD with a lathe and trying to figure out what I need, electrically, before and after the VFD.

    It looks like (from the link below) that he's using:

    1. MCB
    2. Magnetic contactor
    3. AC reactor (he reckons this ain't needed in a home shop)
    4. Fast acting fuse
    5. Input noise filter
    6. VFD
    7. Output noise filter
    8. Motor
    I have the MCB built in to the supply.
    The contactor seems very much like what we would call an NVR in the UK - should I bother wiring in either an NVR or a contactor, or could I just go with the 13A fused plug and switch to supply the VFD (with the MCB behind it)?

    I do plan to have low-voltage switches for forward/stop/reverse and frequency on a pot. I'm thinking I can power the whole lot either directly from a switched 13A BS1363 socket or maybe put an NVR in series.

    Thoughts?

    Reference link:
     
  2. mike os

    mike os just a little insane.....

    Messages:
    5,833
    Location:
    North Wales
    Nope vfd needs supply side protection like any machine, can be mcb, fuse etc.

    Vfd already has an equivalent nvr/contactor built in so won't restart after power outage

    No switch gear should be between the vfd and motor, control is via the vfd control panel.... which may be wired into the machine controls
     
    mtt.tr and skotl like this.
  3. mike os

    mike os just a little insane.....

    Messages:
    5,833
    Location:
    North Wales
    Additionally fitting nvr/contactor before the vfd may result in damage to the unit on overrun and will also remove the braking function as it needs power to stop the motor
     
  4. m_c Member

    Messages:
    76
    Location:
    East Lothian
    The VFD manual would be my main guide, not some OTT youtube video.

    @mike os most VFDs can be programmed to automatically restart after power loss, so I wouldn't generically say they act like NVR, although that's how I'd expect them to act by default.
    Also killing power to the VFD shouldn't cause any issues, but it can lead to unpredictable motor stopping depending on if the VFD control circuitry loses power before or after the motor has stopped.
     
    mike os likes this.
  5. mike os

    mike os just a little insane.....

    Messages:
    5,833
    Location:
    North Wales
    Yes by default was implied ....

    Most manufacturers seem to dislike the idea of disconnection under load, although that is obviously the default in the event of power failure. Not come across one with sufficient power to apply electronic braking once power is disconnected
     
  6. Anguz

    Anguz Forum Supporter

    Messages:
    433
    bedfordshire
    All of our panels at work with vfds have a contactor supplying the power to the drive... wired to the emergency stop button.
    Contactor drops out if you hit the stop
     
  7. skotl

    skotl Forum Supporter

    Messages:
    6,483
    Location:
    Edinburgh, UK
    Fantastic - thanks guys.
    I figured there was a difference between US and UK best practise.
     
  8. Pete.

    Pete. Member

    Messages:
    8,355
    Location:
    Kent, UK
    I just stick a plug on mine. Fused plug - wire - vfd. My HLV has a second fuse within the cabinet and if I was hard-wiring or using a commando socket I'd certainly put an inline fuse in.

    If you're running it from a 3-pin plug at home why do you need a contactor? Just switch it off and/or pull the plug. Might not be 'best practice', probably isn't but then does your kettle have one? What about the hoover/strimmer/flymo/washing machine?

    You CAN set some to be run enabled when powered up but you have to set them that way, I never saw one where it was the default setting, doubt you could find one purely from a liability aspect.
     
    skotl and eLuSiVeMiTe like this.
  9. eLuSiVeMiTe

    eLuSiVeMiTe Forum Supporter

    Messages:
    8,963
    Location:
    Bedfordshire England
    Did exactly that. There's a fused switch. No drama.
     
    skotl likes this.
  10. Hopefuldave Intergalactic pot-mender

    Messages:
    933
    Location:
    The Shed of Danger, surrey, England
    I went a bit OTT with mine...

    Type C MCB rather than a fuse for the power in.

    As per Anguz, contactor set up as a NVR on the mains input to the VFD, spare contacts on it in series with the low-voltage common to the Fwd/Rev switch and also added an extra "neutral switch" on the lathe in series with the contactor "on" button so the lathe fwd/rev has to be in the neutral position before I can power the VFD up, prevents surprises!

    Baccy-tin sized 20A mains filter so the kids and I can listen to the radio.

    Home-brewed output inductors between VFD and load: a trio of identical (scrapped) control transformers dismantled with all the E and I laminations at the same ends and gapped with a shim of PTFE, rewound with 1.5 mm enamelled copper (about 120 turns, Some Experimentation Required*) to preserve the irreplaceable vintage motor and VFD as the motor is 3-speed with switched windings.
    There's an interlock on the speed-change lever to put the VFD into "coasting" (breaks the fwd/rev common) before switching, only way I could think of doing it!

    I also added varistors/MOVs/VDRs across the speed-change Frankenstein switch (2KV threshold voltage) to minimise any spikes, on the motor side of the switch.

    MY ABB VFD runs its control circuits from the bus capacitors as far as I can tell, dynamic braking pushes charge back into 'em (and the external braking resistors), so it maintains its control circuits in an E-stop.

    Dave H. (the other one)

    * The norm for the motor-side load inductors is between 3 and 5% of supply voltage dropped at full load, base frequency - this is the reactive impedance Z you're looking for so e.g. 240v 5A at 50Hz, 5% will be a voltage drop of 12V: Note the voltage and current multiplied give 60 VA so this is the sort of size of transformer core to use! Bigger won't hurt!

    Some calculations.... eek.

    Using V = I*R
    12 = 5 Z
    Z = 2.4 Ohms (reactive)
    Z = 2*Pi*f*L - rearrange this to L = Z/(2*Pi*f)
    L = 2.4/(2*3.14*50) = 2.4/314 = 8 milliHenry (mH) for a "5%" inductor, scale back to 4.8 mH for a 3%.

    The experimentation bit is measuring the inductors, either a cheap LCR meter from a Chinese Gent on EvilBay or setting up test equipment to measure from first principles!

    Note that the impedance of the inductors scales with frequency, that 2.4 Ohms at 50Hz scales to 720 Ohms at 15KHz (e.g. switching frequency), around 2K at the 45 KHz harmonic, etc etc, the high frequencies being the rising edge of the VFD output waveform which is what can cause breakdown in ancient insulation - what I wanted to avoid.

    It's legitimate to add capacitors on the motor side of the inductors (NEVER the VFD side), which can turn the inductors into a sine-wave filter, but that can lead to unexpected effects like resonances, a lot of calculations necessary! I didn't.

    There you go, intricate and tedious detail (I'm good at that), none of which was wanted :)
     
    daleyd, skotl and eLuSiVeMiTe like this.
  11. m_c Member

    Messages:
    76
    Location:
    East Lothian
    @mike os sorry if I came across a bit harsh this morning, it was just a quick reply before I headed to work.

    @Hopefuldave I would not have bothered with a neutral interlock. Any VFD I've installed, by default will not automatically run when powered up, regardless of fwd/rev terminal status. If Fwd/Rev is connected, then they need disconnected and reconnected before the VFD will start moving the motor.
    Also changing direction, the VFDs deceleration/acceleration parameters should be honoured, so again I wouldn't bother with any interlock. If it does fault out, then it's because a parameter needs changed.

    Try stopping the motor too fast, you'll trigger an overvoltage condition, due to the VFD trying to dump the motors energy back into the VFDs internal DC bus (this is where braking resistors come in, as it gives the VFD somewhere to dump the energy to).
    Try starting the motor too fast/under too much load, you'll either trigger an overcurrent, or undervoltage fault (which one will mostly depend on the quality and/or internal monitoring of the VFD).

    Filtering is always good for a debate.
    For domestic installations, legally an input noise filter should be fitted. Personally it's something I always add, as they're only a few pound, and reduce potential headaches.
    Output noise filters, I've never personally installed (I'd always use suitably grounded CY cable though), but I do have one CNC machine which came with and uses un-shielded wires wrapped through some basic toroid inductors just after the VFD. I'm guessing it's enough to kill any troublesome noise, but probably still letting some noise reach the motor (it does have an Inverter rated motor, so the VFD input should never be a problem).

    AC reactors help improve power factor (aka reduce power bills...), but on a low power VFD, any benefit is likely to be minimal. Every switchmode power supply in your house is already likely to be reducing your power factor, so a KW or so occasionally isn't going to make much difference.


    Personally, for a manual machine, I'd just plug the VFD into a suitable socket.
    If you do need more than 13A, be aware that if you do wire it in and you switch both live and neutral at the same time, you can trip any RCDs upstream due to the VFD input filter capacitors. That is why manuals will often state VFDs should be protected with 200mA RCDs, but most domestic RCDs will be 30mA.

    There is also the issue with MCBs, as the high frequency switching can falsely trip them (hence the mention of Type C, which allow for higher surges).
    However, for both those issues, I'd try running it and see if you have any problems. I've only ever heard of problems where there have been other issues, like a MCB that is already heavily loaded, and the VFD switching is enough to push it over the edge to tripping.
    Or in my case, trying to run too many VFDs/Servo drives of the same RCD (I can only power up 5 drives at a time in the workshop, any more and darkness happens!)

    Fusing is another questionable topic. A suitably sized fuse should never blow, so if the VFD does something to blow the fuse, the chances are the VFD's magic smoke will already have been released before the fuse blows. A fuse is really a last line of defence to stop the magic smoke turning into not so magical flames, but an MCB should do that equally as well.
     
    skotl likes this.
  12. rikrobson

    rikrobson Member

    Messages:
    3,744
    Location:
    Perth, Scotland
    Mine are just plug vfd motor
     
  13. R-D-R Member

    Messages:
    951
    Location:
    Derbyshire - England
    I went with a 12vcontactor to turn on the power to mine as per recommended installation by the manufacturer.

    it forms part of the safety loop if there is a power loss. Not supposed to trust software settings to return the machine to a stop on power returning etc.

    that said I think you’d be fine with a plug I just enjoy doing things a bit OTT for a home gamer.


    Here’s the start of my VFD install as I say you defo don’t need it I did a cooling fan and external brake resistor etc.

    https://www.mig-welding.co.uk/forum/threads/harrison-m250-restoration.83255/page-18#post-1230655
     
    skotl, Hopefuldave and mtt.tr like this.
  14. Hopefuldave Intergalactic pot-mender

    Messages:
    933
    Location:
    The Shed of Danger, surrey, England
    Hi M_C,

    The neutral interlock cost nowt (microswitch from the junk box, a few feet of cable that was going to the fwd/rev anyway), can't make it LESS safe.
    I would rather have hardware stop it spinning up unexpectedly than trust to software in the VFD (I used to write machine firmware, know how many bugs get through to the finished product!). The other interlock switch (for speed changes) was already there, when pulling the speed lever out of the gate it would have dropped the power to the Frankenstein switch to prevent switching windings under power - pretty much what it's used for with the VFD.

    Agreed, the VFD takes care of changing direction, and I fitted 1500W rated braking resistors - the 10" 4-jaw with work in it stores quite a bit of energy to get rid of when up to 1000+ RPM, 1500W is only half the motor power that would go into accelerating the spindle and chuck anyway, luckily the Holbrook has a physical brake on the spindle drive for general use when disengaging the high/low speed clutches for a normal stop so a quick reverse only has the motor, first belt and gearbox input shaft to deal with :)

    The load inductors aren't there for interference, they're to reduce the dV/dT of the VFD output waveform before it reaches the motor - as per my post, the motor is a) older than me, b) 3-speed with Dahlander windings for 2, a separate winding for the 3rd, c) irreplaceable at sensible prices!

    Domestic switch mode supplies of any power will be equipped with integral power-factor correction in the chipset - they've been available and used by reputable makers for 15 years or more. £1.99 (including shipping...) Chinesium wall-warts etc. won't have 'em, additional parts count, and if they were that bothered we wouldn't see deaths by phone charger...

    I have to disagree re single vs double pole switching tripping RCDs - single pole is more likely to trip as the charging current going into the filter Y capacitors is being pulled from only one leg, rather than being equal and opposite in the L and N - exactly what an RCD is supposed to trip on!

    DECENT VFDs should be happy on a type-B MCB, they have soft-start circuits (big resistor on the DC link, bypassed once the bus voltage is up to working), cheap ones it's worth having a type-C as there's a much bigger initial surge current - they have the input rectifier connected direct to the bus capacitors, one reason why they're not as durable as a brand-name and more expensive inverter. My lathe has its own 32A socket with isolator fed from its own 20A type-C MCB in the Shed of Danger's own RCD protected consumer unit straight off the meter tails, so a very low source impedance, hasn't tripped yet (nor has the rotary converter on its own type-D 45A MCB... Ouch)

    Mention of fans in R-D-R's post - something that gets forgotten with VFD installs is that running the motor at low speeds means next to no cooling airflow - an auxiliary fan controlled by the VFD is a Very Good Idea if you're going to work it hard at low speeds, most VFDs can switch an internal relay, ideally to OPEN when the VFD frequency goes ABOVE e.g. 30 Hz - then it'll cool the motor at low speeds and between "runs", the motor's own fan can take over when up to a sensible speed. Look for "supervisory frequency" in the index to the book of destructions :)

    Dave H. (the other one)
     
    skotl and daleyd like this.
  15. rikrobson

    rikrobson Member

    Messages:
    3,744
    Location:
    Perth, Scotland
    @skotl
    You're more than welcome to pop up one weekend and see how my lathe and Mill are setup
     
    skotl likes this.
  16. skotl

    skotl Forum Supporter

    Messages:
    6,483
    Location:
    Edinburgh, UK
    Might take you up on that - 3 phase motors are a bit of a black art to me!
    Cheers :thumbup:
     
    123hotchef likes this.
  17. rikrobson

    rikrobson Member

    Messages:
    3,744
    Location:
    Perth, Scotland
    its scarier than it actually is. The First one took some programming, but the last one was pretty much plug and play. I've a wee set of controls i made for the lathe, but the one for the mill had a Pot on the VFD so it is mounted next to the mill and i use the front of the VFD as the controls


    The thing to watch is the Max Speed (its all in Hz) and the run up and run down times. They can limit the current ceing drawn. a quick run up time can pull too much current, but there is an ammeter built into the VFD
     
  18. Seadog

    Seadog Forum Supporter

    Messages:
    4,575
    Location:
    NE London - UK

    Not necessarily. Some can be configured to resume the pre-power-failure state.
     
  19. northwest Member

    Messages:
    576
    Location:
    Manchester UK
    I was very leery of them at first, White Man Magic.

    Won't have a machine now if I can't get a VFD on it, absolute doddle once you get your head around them and I am not the brightest of folk so if I can program one so can anyone.
     
    skotl and eLuSiVeMiTe like this.
  20. eLuSiVeMiTe

    eLuSiVeMiTe Forum Supporter

    Messages:
    8,963
    Location:
    Bedfordshire England
    All mine will go that way eventually.
    Quieter smoother and a lot more control. No brainier really
     
Advertisements