What's new
What's new

VFD ?Single phase input= Three phase output

will a VFD that is not specifically rated for single phase input/three phase output function properly with single phase input.?
Unless the manual specifically says that it IS rated for single-phase input, it should not be used with single-phase input.

That's why they print manuals.

- Leigh
 
I have a lot of respect for Leigh but I have to differ with him on this one.

There's many electrical considerations separating Quasi's simple question from a simple answer. The answer is: yes you can run a three phase rated VFD from single phase - BUT!

Single phase input rated VFD's can be run to their full ratings. Period.

A three phase motor of a given HP demands 58% the Amps per conductor of an equivalant single phase motor. This difference is reflected in the lighter input circuitry for a three phase rated VFD. Three phase input rated VFD's when connected to single phase power may have to be de-rated (de-rating refers to using a 5 HP VFD to run a 3 HP three phase motor).

If the motor is installed in a manual machine tool, it's seldom loaded to full ratings and a lower de-rating factor can be used. If the motor runs a pump to full load amps then the three phase rated VFD serving it should be rated DOUBLE the motor's HP.

That said you still have to monkey with the VFD's parameter settings (the instructions walk you through it) to optimize the VFD to the motor it drives.

So, I guess that's a qualified "yes."
 
Last edited:
My 2.5 cents...

Some drives will fault out if only 2 phases applied..

Phase loss is important sometimes :)

Just like running controls on the wild leg on a RPC...

It does depend on the drive..
 
Question: can you fool phase loss detection by connecting one of the single-phase legs to two VFD input terminals, instead of just one? I imagine this might work if the phase loss detector looks at voltage phase to ground, rather than phase to phase.
 
Question: can you fool phase loss detection by connecting one of the single-phase legs to two VFD input terminals, instead of just one? I imagine this might work if the phase loss detector looks at voltage phase to ground, rather than phase to phase.
Nothing in a VFD is going to measure phase-to-ground. It will look at essentially phase angle differences between the 3 phases. So with 1 phase, no matter what you do, two of them will not have any difference.

Some VFDs monitor phase loss because the Diodes on the front-end are not sized to take the increased current that you will see if you feed it with 3 phase (and the capacitors are insufficient to filter the extra ripple). Some mfrs recognize that if you know what you are doing and double the size of the VFD with respect to the motor, you solve that problem, so they allow you to disable the phase loss tripping in programming. But some mfrs, knowing that the VF is not intended to be used on 1 phase sources, tap off of only two of the 3 phases for control power for the VFD electronics and if you are missing one phase, it doesn't work. This is a lot more common on 480V drives made for the North American market mainly because the designers failed to anticipate that 480V 1 phase exists.

But for the most part, 230V 3 phase drives can usually be used on 1 phase sources if you double the size. 3HP and below, there are a vast number that do not even need the over-sizing.
 
Oops! Jraef makes an excellent point I sould have addressed in my earlier post.

If you're going to run a 3 phase rated VFD on single phase you need to determine how the control power is derived. They have a little power supply connected to two of the three input terminals. This power supply runs all the internal electronics that controls the behavior of the VFD.

Make sure you connect your single phase power to the correct two input terminals. Which one? Check your manual. No manual? Scrounge the manufacturer's website or any of several manual location/download services.

You HAVE to have the manual for your VFD to make the right the parameter settings and for all the other info you need. A VFD is fairly simple to set up and run but without the manual you're lost.

If you know what to look for and know about electronic trouble shooting you can find the two terminals that connect to the control power supply.

My favorite trick is to try single phase power on all three terminals a pair at a time. When the panel lights up and the controls function you got the right terminals. Inelegant but effective.
 
A lot of drive designs use a DC-DC power supply and tap off for control power from the DC bus, which means it doesn't matter at all which line terminals you connect to. I'm not sure why they don't all do it that way, but they don't.
 
A lot of drive designs use a DC-DC power supply and tap off for control power from the DC bus, which means it doesn't matter at all which line terminals you connect to. I'm not sure why they don't all do it that way, but they don't.

easy answer: COST. think about it - an ac input is much easier to make logic from than DC - dont need to chop the dc back up to make ac again to work with. cheap is it.

I know all the Hitachis Ive messed with use 2 specific legs for logic - IIRC it is L2 & L3.
 
I'm not sure why they don't all do it that way, but they don't.
Because you're adding an entire additional inverter and drive system for a simple function that could be accomplished with a small transformer.

The AC source for the control system is not variable frequency, so the inverter drive would need to be independent of the drive signal for the VFD main outputs.

Regulating the HV DC bus down to 24 volts or so for the control system, given a drain of a couple of amps, would increase the power dissipation within the VFD by a huge amount.

Probably best to avoid the concept entirely and put in a little transformer and a standard switching supply.

- Leigh
 
Because you're adding an entire additional inverter and drive system for a simple function that could be accomplished with a small transformer.

The AC source for the control system is not variable frequency, so the inverter drive would need to be independent of the drive signal for the VFD main outputs.

Regulating the HV DC bus down to 24 volts or so for the control system, given a drain of a couple of amps, would increase the power dissipation within the VFD by a huge amount.

Probably best to avoid the concept entirely and put in a little transformer and a standard switching supply.

- Leigh

I think you agree with me, and with JRAEF, but what you say is partly opposite.....

The bulk of VFDs that I am familiar with DO derive the control power from the DC bus, because it is cheaper, and because the VFD needs no control unless tehre IS DC.

There is almost NO dissipation associated with the supply, simply because it is an SMPS, often a single-switch forward converter. For a 480V drive, you need a very high voltage IGBT, but you need very little current capability, because the SMPS is inherently a "power" converter, and takes in the HV at low current while outputting LV at higher current.

In general it is only a few watts which is needed anyway.

I actually design drives every so often (special purpose ones), and I would never consider for one moment putting a non-SMPS control power supply in any VFD I design.... Too bulky, for one thing, and actually may dissipate more power than an SMPS of proper design. The SMPS will output all my voltages, 5V, 15V, 24V, all nicely regulated and accurate, no linear regulator needed.

"cold" side power comes off another winding, insulated from the "hot" side windings...... but still regulated.

And, I know I have power so long as there is DC, even if I am given DC on a common bus.
 
Teco

I have a TECO by Westinghouse. I was also concerned about this because the factory manual did not clearly state that it could be fed with single phase. I called TECO and they actually sent me a signed document verifying and authorizing the VFD to be fed with single phase. This approach gives you factory and warranted assurance that its OK.
 
"And, I know I have power so long as there is DC, even if I am given DC on a common bus."

Which is a possibility, should you have several or many VFDs and one HUGE external converter section.

It was more than 20 years ago that my employer in a former lifetime was making mainframe computer PSUs in this way ... one HUGE converter and as many individual PSUs as were required. A few were +5 volts, a couple were +12 volts and -12 volts, but most were -5.2 and -2.0 volts for the ECL in the CPU. Almost everything was CMOS, except for the ECL central processor and I/O channel units.

Back on VFDs ...

A 240 or a 120-to-240 voltage-doubler VFD will have a 325 volt bus ... a 480 VFD will have a 650 volt bus.
 
[quote JST]... because the VFD needs no control unless there IS DC.[/quote]

One thing that just occurred to me after reading this was that there are some VFDs that use an SCR front-end converter so as to ramp initial voltage and do away with a pre-charge resistor. In that case, you WOULD need control power before there is DC, so you would HAVE TO have an AC power supply. A-B 1336 drives were like this.
 
And, I know I have power so long as there is DC, even if I am given DC on a common bus.
That's not sufficient for safe operation.

Modern controllers, i.e. those using any type of "intelligent" control system, be it CPU-based or similar, require time to initialize before they are operational.

Simply applying power to the control system at the same time you apply power to the output stages leaves a time interval during which there is no active control of the output. In fact you would have output stage power applied BEFORE power was available to the control system due to capacitor charge delays.

It also prevents the control system from doing a proper power-up self test to determine whether it and associated systems are functioning correctly before applying power to the output stages.

In other words...
Extremely bad product design from a safety standpoint. :toetap:

- Leigh
 
That's not sufficient for safe operation.

Modern controllers, i.e. those using any type of "intelligent" control system, be it CPU-based or similar, require time to initialize before they are operational.

Simply applying power to the control system at the same time you apply power to the output stages leaves a time interval during which there is no active control of the output. In fact you would have output stage power applied BEFORE power was available to the control system due to capacitor charge delays.

It also prevents the control system from doing a proper power-up self test to determine whether it and associated systems are functioning correctly before applying power to the output stages.

In other words...
Extremely bad product design from a safety standpoint.

- Leigh

Well, now that you mention that.... NO IT IS NOT bad design..... it is standard

You may be rather out of touch with commercial VFD design..... There are many, many ,many, in fact MOST VFDs which will work just as well on DC input as AC..... And they obviously use DC for the control power source.

ANY "common bus" type will HAVE to use DC.

1) Any gate drive system that does not flat out GUARANTEE pull-down of the gates with no power..... is unacceptable. And just about all gate drivers of any reasonable sort have , at the least, a positive pull-down system as part of the 'UVLO", so that if the voltage is not at least the minimum , there is no gate drive. One can add extra safety equipment if that isn't sufficient.

2) Power on testing is easily performed whether or not the main power is applied..... it depends on what you decide to test.....

3) At some point power is going to be applied, and unless you are in an entirely different class of ultra-safe products certified as disconnects, the only protection you have against a bad power IGBT is the input fusing.

4) The pre-charge issue is easily handled by positively bypassing the pre-charge with a relay closure. Obviously no power, no relay closure, and very limited power available. The best precharge is a capacitive type, no dissipation. But most are resistive, with the resistors acting as a fuse if the precharge fails to pull up the voltage.
 
2) Power on testing is easily performed whether or not the main power is applied..... it depends on what you decide to test.....
That's an interesting concept... Power-on testing with power off.

Given your earlier statement that ALL power had to come from the main DC bus, there's no other interpretation possible.

I guess the laws of physics don't apply to VFDs. :eek:

And BTW, I don't dispute the claim that your solution was common practice. I just said it was bad design.

- Leigh
 
That's an interesting concept... Power-on testing with power off.

Given your earlier statement that ALL power had to come from the main DC bus, there's no other interpretation possible.

I guess the laws of physics don't apply to VFDs. :eek:

And BTW, I don't dispute the claim that your solution was common practice. I just said it was bad design.

- Leigh

Power on to IGBTs or power OFF to IGBTs.... :rolleyes5:

With power from DC, it would be power to IGBTs, obviously

With power from a separate AC supply, it could be power to IGBTs OR power not connected to IGBTs yet......

Not all VFDS are supposed to be disconnects, few are, as far as I know.

And I'm not really sure what the hazard is considered to be here, if there is a fault in the DC bus or IGBTs, the bus will not charge up, the precharge resistors will act as a fuse (they normally are so rated) and power will be removed from the bus, as well as from the local PS. That will remove the problem before the precharge finishes, most likely.

In any case, this thread has veered off course far enough.......
 








 
Back
Top