Multi Motor shop on one VFD

Hi

I know this has been discussed a number of times but I ran across the following on another forum in a thread on using a VFD. Seems simple. I almost tried this earlier this year with a 15HP TECO MA 7200 that is not used very often, but decided it would be my luck to blow it out!

Anyone else try this simple approach, ie no speed control just whole shop simple switches phase conversion?

Is there something special about the Mitsubishi he is using? Is it really a VFD or just a static converter? I looked at the manual and it reads like a VFD / Inverter.

For those interested in the whole thread:

The Home Machinist! • View topic - Remotor or Rotary Phase Converter

Bob

_____

A post in a thread:

“friend of mine runs his whole shop on one VFD and most VFD's are capable of doing that. The VFD is big enough to run at least two motors at a time, and the switches to turn on the machines are at the machines. So he turns on the VFD when he walks into the shop in the morning, and turns it off when he leaves at night. The list of machines are as follows: 1, 3-axis CNC Bridgeport, a large bandsaw, 2 lathes, 1 horizontal band saw and several tool grinders. He is not careful about the two motor rule either. So far, has has lost one VFD in the past 10 years and it was to lightning.”


My question:

“I have always wanted to try the one VFD approach and am about to do some changes in the shop. Do you know which VFD he used and it Rating. I was thinking a 15HP unit would cover me, just wondered what model he used.”

His response:

“I will have to take a run up to his shop to see what brand it is”

Next Post:

“It's a Mitsubishi FR-E500, 5hp, 240vac, single phase input. I was wrong about the lightning. He had one mounted behind the BP, not enclosed. He was using an air hose blowing chips off the table and they went into the drive with the following pffft report. This one is mounted right next to the electrical panel.
He did say that if he is running the CNC Bridgeport, he will pause the program if he is starting another machine. The spindle will slow down for a couple of seconds, and then pick right back up, but the program doesn't slow down at all. Raises hell with 1/16" end mills. He does a lot of engraving of wood pictures.”

My next response:

I thanked him for the trip and said if I tried it I would post a thread.

piedmontg said:

Hi

I know this has been discussed a number of times but I ran across the following on another forum ...

Click to expand...

*sigh*.. "another forum" eh?

In yet-another venue, a guy who worked in food processing had been getting away with putting his d**k in the pickle-slicer with no adverse effects.

Eventually, however, he was caught at it and fired.

To be fair and unbiased, the company fired the pickle-slicer as well.

Eventually, you'll probably find shared VFD kludges have a similar outcome.

But it's your shop, your anatomy, and your pocket, so nothing stopping you.

Bill

When I did my experiments with my 400w vfd (internal igbts rated 15 amps), using a 5% reactor and capacitors to generate a sinewave, I had no problems hot plugging stuff on it, including capacitors. However it is simple physics to see that the vfd must be able to handle the inrush current for a long enough time, usually 6 times the motors full load current.

Heck, this old tale is as old as the hills.

MODERN VFDs do not have an issue with anything but the start current involved. If they can deal with that, the rest is easy.

NO reason it will not work if suitably sized.

My favorite example is the hundreds of thousands of solar inverters running motors of all kinds from batteries. I have one myself. 20 yeas now, on a 2500W inverter with no special circuits, turning on and off anything, including the air compressor. The thing is built just like a VFD, but is single phase.

Any of the VFDs I have designed I would not expect any problems from turn-on or turn-off of loads.

People should be forbidden to spread this tale unless they can explain in detail what the failure mechanism is, and how it differs from normal operation,

JST said:

Heck, this old tale is as old as the hills.

Click to expand...

Have a care. PM's own archives will show you have been on BOTH SIDES of the argument, too! "Explanation" could be as simple as citing your own earlier posts.

In short - selecting the 'right sort' of VFD then configuring it advantageously is not assured for use with several different possible loads.

New-arrival comes in the door with minimalist research and the queries of a naif, I do not presume he has the years of experience to 'make it right' first-go as you, Jraef, Dave Kamp, meself, or ten thousand OTHER experienced PM grey-hairs or their students could do.

We've each already blown up one or more..



Better bet for less-experienced folk is one dedicated VFD per machine that actually NEEDS the "Variable", and/or 'soft start' feature, and a dumb-but-rugged RPC for those machines that do not.

Which might be ALL of them, actually.

Harder to screw-up. Less monkey-motion to use. No need of bypassing existing controls.

Brand-new VFD keep getting cheaper, add one or more LATER.

Bill

I have, indeed.

OLD VFDs are susceptible to issues. Particularly ones so old as to use regular bipolar transistors. "SOME" others are potentially susceptible to problems, but for perhaps 10 years, they have been fine. My inverter was an old design when I bought it, early MOSFET design. It careth not what you do so long as you do not exceed the 5 kW limit, current-wise. And it is from the middle 90's.

OT- Inverters

JST said:

I have, indeed.

OLD VFDs are susceptible to issues. Particularly ones so old as to use regular bipolar transistors. "SOME" others are potentially susceptible to problems, but for perhaps 10 years, they have been fine.

Click to expand...

If I still had a need of 'em, I'd be as concerned about PROGRAMMING the critters for a variety of loads. Bloody nuisance enough as it is for a new user OR a new VFD without having to second-guess where to depart the needs of a single motor it can actually run tests against.

. My inverter was an old design when I bought it, early MOSFET design. It careth not what you do so long as you do not exceed the 5 kW limit, current-wise. And it is from the middle 90's.

Click to expand...

Off-topic, but I am actually in the market for 1 to 3 such NON 'grid-tie' critters.

Solar dasn't yet match grid economy in my 'plant' size, but I want basic stairwell & bath LED lighting, WiFi/routers, handheld recharge strip, and most of all the fridge/freezers on battery bank --> inverters.

That way, the 10 KW Diesel dasn't have to be run 24 X 7 during an outage, nor go 'wet stacked' from miniscule loading.

Their sleeping-hours, 'tween meal prep, door SHUT maintenance loads are rather modest. Under 10A @ 120 VAC, compressor-start and all. Lighting is on motion-detectors, so that's modest as well.

Bill

Xantrex, Outback, either of them make a good product. There are others also. And the usual stack of imports from wherever that probably will work for a while.

Move as much as possible to the 12V DC side. Routers etc work on low volts. Lighting can, my outdoor motion lights are all on 12V, converted from 120V.

Make sure the inverters "load seek", that they shut off when lo load, and come on if there is a load. Great reason to move small loads to the 12V (or 24V) side.

JST said:

Xantrex, Outback, either of them make a good product. There are others also. And the usual stack of imports from wherever that probably will work for a while.

Move as much as possible to the 12V DC side. Routers etc work on low volts. Lighting can, my outdoor motion lights are all on 12V, converted from 120V.

Make sure the inverters "load seek", that they shut off when lo load, and come on if there is a load. Great reason to move small loads to the 12V (or 24V) side.

Click to expand...

I'm fighting 'too many' years of neg 48 VDC, nominal, knee-jerk habits.

Will probably set up the bank as +/- 12 VDC, Earth referred 'centre tap' equivalent so I can use common batteries & chargers, then either 12 V or 24 V goods downline.

And yes .. directly off the DC where possible.

Project has been 20+ years in the unfolding since I DID have 48 VDC recycled Lorraine ex-telco gear.

During that time the 'whole house' need has dropped from 30 KW to 10 KW, HVAC inclusive, and yet - I now have to worry about 'wet stacking' even that.

Guess I'm not so bad at 'going green' after all. Waste less rather than 'capture' more.



Bill

Yes, it is now LESS LIKELY (not impossible) to damage the output transistors from OPENING a contactor/switch on the output of a running VFD, but it is VERY likely to be harmed by CLOSING a load into one. The motor starting current MIGHT be able to be handled by software, but not the magnetizing current, the "inrush" that takes place when there in nothing but the winding wire resistance keeping it from being a short circuit. The problem is, that all takes place in the first cycle (1/60th of a second), so too fast for most meters to see it and too fast for most protection circuits to react to it. But from a "molecular" standpoint, the dI/dt (delta [change] in current over delta ]change] in time, basically the rate of rise of the current) is so steep that the transistors can suffer incremental, and usually cumulative, damage. We have this argument in here from time to time and I get that you and some other people do NOT see the damage, but I'm on the other end of the phone with people that DO see it all the time. Those that don't see it... lucky, and for me, luck is not a strategy ...

if a protection circuit cannot see that surge in time, it's just a lousy, nay, ROTTEN design. CHEAP, not good.

If you are going to protect IGBTs, you need to detect the surge and shut off the device in something under 5 to 10 microseconds. 3 microseconds detection and shutdown for a "square load line" (dead short on output) is a good goal with most common IGBTs. Some are more robust, the short circuit time is up to 10 uS for newer ones, but you also need time to shut down the gate drive, and you must keep the current below any "latch-up" current limit.

If you do not do that, then the protection is just not capable of protecting the IGBT, it must be there for controlling average or RMS output, etc. Software stuff, not a hardware detector. You need a hardware detector as an ultimate limit, and it is cheap to do, usually, with the very same detectors used for the software limits.

This is perfectly possible, done all the time. Our units had to survive repeated operation into shorted or (usually, not always) grounded output lines, and they do. I know this from the tests, and also from field data. One series does not protect against grounds, and that is a known accepted thing.

Yes, there are units out there with short-cuts.... Not usually for units large enough to power a shop.

And, of course, the "Phase perfect" allows any type operation that you would do with a regular 3 phase line......... And it is based on what is really the same circuitry as a VFD.

Just as a point of interest, the most common issue with new designs in my experience is deciding how much DESENSITIZING to use with the protection. You MUST allow a short time of "blanking" to allow various noise and parasitic currents to pass before limiting. Otherwise your ultimate limit is too low. That's a far cry from not being fast enough to "see" magnetic charging currents, which are reasonably slow.

One issue for protection is winding capacitance..... it has to be charged, and a capacitor is a dead short for pulse "edges". Generally there is enough inductance to slow that down enough so as not to trigger the protection. But sometimes it needs a "shim" inductor. But this again is a matter of protection being TOO FAST, not too slow.