Repurposing a VFD for a fixed frequency specialty single phase motor?

[GregBuchholz]

I'm working for a company that is making Stirling engine cryocoolers. We have a single phase, permanent magnet linear motor that reciprocates at 60Hz. You vary the amount of cooling by modulating the voltage applied the the windings. More voltage equals more linear motion, which displaces a higher volume of the working fluid, which in turn extracts more heat from the heat exchangers. Se we need a 10kW, 0-270VAC variable source of 60Hz power. It seems like a sufficiently over-sized off-the-shelf VFD has all the necessary hardware to perform the power conversion that is required to go from 3-phase 480V on the input side, to single phase variable AC voltage on the motor side, at a constant 60Hz. The problem as I see it is the embedded firmware in the VFD. Does anyone know of a VFD that has enough flexibility which would allow us to operate in a fixed frequency, variable voltage manner? Or one that allows extensive reprogramming of the internal operation of the drive? I've contacted a few VFD manufacturers, and so far the message is that they don't support something like this.

Thoughts?

 

[iceworm]

I can't help with the VFDs.

Just curious, have you considered a lamp dimmer? 10kw modules are available.

iceworm

 

[GregBuchholz]

As for light dimmers, we're looking into things like that (especially "sinewave" dimmers, rather than just the standard phase-angle fired triac approach), but we're also looking for a 3-phase 480V input to help slot into existing infrastructure. Do you have a manufacturer that you like for 10kW light dimmers?

 

[iceworm]

Input: Even loading on 3 phase 480V input

Output:
Single phase
Sine wave
Fixed 60hz
variable voltage 0V - 270V

Nope - No help. I don't have any good ideas at all.
Motorized 10kva variacs are available. But that won't meet your input even loading spec. And one would have to gin up a 4-20 (or what ever control signal you are using) to control the variac motor.

As for theratrical dimmers, the few I've seen are triacs (or parallel SCRs) The control inputs are available, but you would need an output filter, input transformer, and it still would not meet the 3ph input even loading

I'm pretty sure you have already considered all these things.

ice

 

[ScottyUK]

You might consider using a back-to-back 3-phase rectifier and a single phase inverter from (e.g.) Schaefer in Germany or maybe Sorensen and/or Xantrex?

From experience Schaefer's C-Series is fairly versatile and very good reliability. They have a 'specials' group who can take on custom work based on their standard blocks. This link is something which I haven't seen before from them - looks like a good match to your requirements.

 

[cswilson]

A smart servo drive should be able to do this quite easily, but it might be overkill for the application.

Curt Wilson
Delta Tau Data Systems

 

[waross]

Three transformers with 120 Volt secondaries. The primaries may be 277 or 480 Volts.
Connect the secondaries in zig-zag, such as is used to convert three phase generators to single phase generators.
At unity power factor the KVA loading will be equal on all three phases.
The power factor on one phase will be unity, on one phase the power factor will be leading and on the other phase the power factor will be lagging.
The sum of the transformer KVAs will be 150% of the load KVA.
Because of the power factor issues, one leg will supply 50% of the KW loading and the other two phases will each supply 25% of the KW loading.
Now you have 240 Volts AC with the KVA load balanced across three phases.
Add a 240 Volt to 280 Volt Variac and you are done.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[ScottyUK]

I like heavyweight old-school solutions.

 

[waross]

Thanks Scotty.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[iceworm]

The primaries may be 277 or 480 Volts.

How were you thinking on connecting a 277 primary? Wye? If so, what were you going to do with the neutral? Just curious.

ice

 

[GregBuchholz]

I'll definitely check out the Schaefer AC/AC converters. Thanks.

 

[GregBuchholz]

Thanks for the ideas waross, keep them coming. I should probably have provided more details in my original post. There are potentially a lot of ways to skin this cat. For prototyping, we using a lab grade Pacific Power AC source that works well, but costs and arm and a leg, and weighs more than the cooler it is driving. We're looking for a relatively low cost, small/lightweight, efficient power supply with good power factor, low harmonics, phase balance, etc.. Before we go out and get a custom supply built, I thought we should check to see if an off-the-shelf VFD could be contorted to fit the bill. The variac approach was considered, but voltage transients on the input side show up on the output side. With our linear alternator, there is a hard mechanical limit, which we can't exceed (or things break). And the linear mechanical amplitude is proportional to the voltage, so we can't tolerate utility fluctuations. Then you could think about adding a Constant Voltage Transformer somewhere in the mix. And all that iron starts to really add up.

 

[Compositepro]

A phase angle SCR power control seems to me ideal for your needs. These are readily available as heater controllers. Your cryo-cooler is, I believe, a resonant device that needs constant 60 cycle power. The power lines are pretty good at providing that. If the frequency is slightly off, that can probably be compensated by a voltage adjustment. This would certainly be your least expensive option.

 

[Skogsgurra]

Is the linear motor fed directly with the incoming 10kW, 0-270VAC variable 60Hz voltage? Or is there some electronics involved? In the former case, I would not expect a good result with a dimmer style voltage "variator" because there are lots of harmonics that makes life difficult for any phase angle dependent device. Which I assume that your linear motor is. In the latter case, the front end of an electronic device doesn't like the phase controlled (steep edges) voltage delivered by a standard dimmer. I use a variac in cases like that. I have a 5 kW unit, but I think there must be larger ones.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[waross]

If transformers with 277 Volt primaries are used, connect them in wye. Connect the neutral to the system neutral. If that is a problem then go for 480 Volt primaries.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[iceworm]

If transformers with 277 Volt primaries are used, connect them in wye. Connect the neutral to the system neutral. If that is a problem then go for 480 Volt primaries.

Not a problem for me. I wouldn't use a Wye primary. I was curious to what you would do with the neutral if you did.

Truthfully, I would not have a clue how to calculate ground fault currents with a grounded Wye primary and a grounded zig-zag secondary. Of course , maybe it doesn't matter - it's only 10kva.

In any event, the idea of using a 3ph primary and a 120/240 zig-zag secondary to partialy balance a single phase load across 3 phase is good.

ice

 

[waross]

Hi Iceworm;
The zig-zag connection that I was referring to is the connection used to convert a three phase generator to true single phase output. No relation to a zig-zag grounding transformer.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[iceworm]

The zig-zag connection that I was referring to is the connection used to convert a three phase generator to true single phase output

Okay ---- I think I knew that. Essentially half of a Double D connection for a 12 lead gen. No, not just one D, rather one cup and one bar.

No relation to a zig-zag grounding transformer

Possibily I knew that too. About the same as I figured you knew that when you said

Connect the secondaries in zig-zag, such as is used to convert three phase generators to single phase generators

I don't know what brought up z-z grounding transformers.

ice

 

[ozmosis]

This company

http://en.amk-antriebe.de/default.asp

(AMK Drives) offer quite specialised controllers that would provide a fairly straightforward solution to your requirements, as would Curt's company Delta Tau, but are quite expensive for simply controlling volts/frequency independently.

 

[GregBuchholz]

I'll definitely be checking out AMK Drives. Is this the Delta Tau you were referring to?

http://www.deltatau.com/DT_IndexPage/index.aspx