Hello Guest it is August 17, 2019, 07:26:00 AM

Author Topic: We live to learn from small steenky explosions. Now what?  (Read 247 times)

0 Members and 1 Guest are viewing this topic.

We live to learn from small steenky explosions. Now what?
« on: March 29, 2019, 10:53:00 PM »
Looking at Youtube for a bit of disassembly info to change out the bearings in my Chinese 2.2kw spindle motor I noticed how quickly the guys spindle would start. I wanted the same. So I set the inverter to make the spindle go waaarrrp!!! Starts in a few seconds. Nice.

Ok so the problem with that, is the huge surge of current, and I may be wrong here so correct me if I am, but once the motor reaches a speed, it no longer needs all that current being delivered so it surges for a small amount of time back into the supply phase? Anyway, the PC, the CNC breakout board, power supply and all 4 drivers went up in smoke. This because all of them were plugged into the same socket. Oh, and it had that steenky burnt electrical smell.

I have 3 phases in my workshop so have dedicated the inverter to its own phase now.

I need to replace my gear. Question. I currently have a normal DB25 breakout board for my 3 axis router. Looking at replacements they are $10. Does anyone think I should upgrade to a better breakout board? Like, is it worth while doing this or do I just stick with the cheap reliable same same?? Be good to hear thoughts please.
Re: We live to learn from small steenky explosions. Now what?
« Reply #1 on: March 29, 2019, 11:08:20 PM »
Hi,

Quote
Ok so the problem with that, is the huge surge of current, and I may be wrong here so correct me if I am, but once the motor reaches a speed, it no longer needs all that current being delivered so it surges for a small amount of time back into the supply phase

Yes, that is correct. Its called an 'inductive voltage spike', where the voltage can for brief periods shoot up to several times
normal.

If you have three phase power in your workshop you should use a three phase VFD, the current is spread over all three phases
and less likely to cause the fault you have observed. In industrial situations with large VFDs it is necessary to use a line
reactor, they look a bit like large iron transformer coils on each input phase. It vastly tames the input surge and noise
associated with VFD's. Hobbyists don't like them because they cost.....but then hobbyists blow more stuff up to!
I would recommend say 5mH on each (ideally three) phase.

You would be advised to run your PC and BoB power supply at least through a line filter. It may be enough to stop
a recurrence of PC and BoB failure. Ideally you would have a line filter for your stepper power supply as well, but it would
have to be a reasonable size (and therefore cost) to handle the required current.

The bottom line is that if you do not do anything to protect your gear you may have a repeat failure. Despite the cost
a good line reactor will save you money in the long run.

Craig
My wife left with my best friend...
     and I miss him!

Offline reuelt

*
  •  444 444
    • View Profile
Re: We live to learn from small steenky explosions. Now what?
« Reply #2 on: March 29, 2019, 11:54:50 PM »
You should look for a good quality BOB that has full Opto-isolation.
Also
Please answer YES in Poll - how you ever blown up your PC?
https://www.machsupport.com/forum/index.php?topic=39723.0
"the gift of God is eternal life through Jesus Christ our Lord"
Re: We live to learn from small steenky explosions. Now what?
« Reply #3 on: March 30, 2019, 12:03:08 AM »
Hi,

Quote
Please answer YES in Poll - how you ever blown up your PC?

Come on reuelt, that poll was to establish if a signal backfeed to the BoB not intercepted by an optioslator
caused a PC fault.

In this instance an inductive voltage spike came in on the SUPPLY SIDE and took out the whole shooting match.
Even if the BoB had opto isolators the whole lot would still go bang because of a SUPPLY SIDE VOLTAGE SPIKE.

To OP....if you don't stop your VFD from producing huge voltage spikes you will continue to blow up PC's, power
supplies, drivers and BoBs. Opt-isolators won't stop it.

Craig
My wife left with my best friend...
     and I miss him!
Re: We live to learn from small steenky explosions. Now what?
« Reply #4 on: March 30, 2019, 02:54:56 AM »
Hi boydage,
this is the sort of thing I mean:

https://www.ebay.com/itm/Allen-Bradley-Line-Reactor-1321-3R12-B-2-5mH-12A-3Ph-600V-50-60Hz-Used/273408795376?epid=16012063544&hash=item3fa86f22f0:g:YRQAAOSwF31bsoTJ

It is installed between the incoming mains supply and the VFD. It vastly reduces the current surges which in turn mean other devices
like your PC and power supplies don't get hammered by wild fluctuations in voltage.

I know this one I second hand but pretty good value at $50.

Craig
My wife left with my best friend...
     and I miss him!

Offline ger21

*
  • *
  •  6,288 6,288
    • View Profile
    • The CNC Woodworker
Re: We live to learn from small steenky explosions. Now what?
« Reply #5 on: March 30, 2019, 08:45:45 AM »
Quote
Ok so the problem with that, is the huge surge of current, and I may be wrong here so correct me if I am, but once the motor reaches a speed, it no longer needs all that current being delivered so it surges for a small amount of time back into the supply phase?

This should not happen. I have everything in my setup powered from one 240V outlet, and the spindle set to accelerate in 2-3 seconds. If you look at the current rating on the VFD, it only draws about 1 amp while accelerating.
Also, the VFD stores power in capacitors, which the spindle draws from. It shouldn't be sending voltage back through the electrical system.

Is this one of those 110V VFD and spindles? Those can be very problematic.
Gerry

2010 Screenset
http://www.thecncwoodworker.com/2010.html

JointCAM Dovetail and Box Joint software
http://www.g-forcecnc.com/jointcam.html
Re: We live to learn from small steenky explosions. Now what?
« Reply #6 on: March 30, 2019, 06:42:27 PM »
Hi Gerry,

Quote
Also, the VFD stores power in capacitors, which the spindle draws from. It shouldn't be sending voltage back through the electrical system.

You are correct, the only way the charge stored in the capacitors to get back into the power supply would be a catastrophic
failure of the input rectifier bridge.

What that does not accommodate is the natural inductance of the supply.

If the load (VFD or whatever) suddenly increases its current draw the voltage will dip. Part of the dip will be the resistance
of the supply, we hope its small but we know its there. The other part of the voltage dip is caused by the inductance
of the power supply. Even a straight wire has an inductance and the secondary of the distribution transformer will
have both magnetizing reactance and leakage inductance. The net result of the wiring and the distribution transformer
is a residual inductance commonly of the order of 1mH.

Lets now imagine the load which had been drawing 20A from the supply sudden stops drawing current, lets say you threw
open the switch, then the inductance will mean that the current cannot stop immediately an a marked increase in voltage
in the supply will result.

In high power electronic circuits where large currents are switched on and off extremely quickly the residual inductance
of the power supply can cause very high voltage spikes.

Take a single phase VFD for instance. The incoming line voltage is 230VAC(rms), its rectified and capacitively
smoothed to 320VDC. At a given instant one of the MOSFET's is conducting 20A and it switches off. The current
will stop within, and ideally, less that 1 microsecond. The inductance of the input supply will cause a voltage spike
(without control or precaution) of another 300V or so. Your nice rugged industrial grade 500V MOSFET has now
a voltage spike of 620VDC to contend with. Whats more the same MOSFET will turn on and off rapidly, say 50,000
times a second, so it gets repeated hits and promptly blows up.

The voltage spikes due to residual inductance are very troublesome to the design engineer who has to design and
implement snubber circuits. Indeed the reliability of the equipment depends on it. Often the quality of the design
can be estimated by the effectiveness and ruggedness of the components used in the snubber circuits.

The upshot is that it is not impossible for the current drawn by a VFD to cause a very high voltage spike which
could easily damage other equipment like PC's connected to the same circuit. I like you have my VFD connected
to the same outlet as my PC and stepper supplies and have not encountered any difficulty. It is only a 750W VFD
however and I treat pretty lightly as a rule.

More recently I made another spindle based on a 1.8kW AC servo with a single phase 10A output/phase servo drive.
I give it hell......and found that I was getting conducted noise issues through the common supply. Nothing blew up but
I think I would have if I'd carried on. What I did was fit a line reactor and all those issues went away and seemed incidentally
to improve the servo drives overload capacity.

That OP managed not just have a voltage spike but a surge enough to take out all other stuff connected to the same circuit,
PC, stepper supplies etc suggests a rather bigger fault than a regular voltage spike. I don't recall ever having seen
a 110V VFD. What is it about them that causes them to be troublesome?

Craig

My wife left with my best friend...
     and I miss him!
Re: We live to learn from small steenky explosions. Now what?
« Reply #7 on: March 31, 2019, 11:25:01 PM »
Hi Guys,

Ok ahem. This is where I need to let everyone know where I went wrong, and perhaps might save someone, or to less worry some equipment. When I knew nothing about CNCs, the very first one I built I put the 4 lead wire to the motor inside an earthing sleeve. You know, one of those expensive stainless steel woven sleeves that may have ended up in my toolbox when I worked for an airline. But I put it there thinking there may be noise generated by the inverter/spindle, best shield it, earth it.

Ok. So. My mistake was not putting a megga on my new connection to the new spindle plug (4pin instead of 3pin). A tiny unseen little hair piece of the sleeve was touching one of the connections inside the plug. Well thats what I assume, I couldnt see it, or find it, but I had a few ohms between a phase and the sleeve and now since I pulled it off and resoldered it, this time test with a megga, its all good.

I wanted to repost with something else, but best you know.