AC/AC Matrix Converters 1

[murpia]

I've seen a lot of research literature on AC/AC matrix converters but actual products seem hard to find. Do they exist yet?

Our application is a test rig with two permanent magnet 3-phase motors. Currently we use 2 inverter drives with a DC link between them. We'd like to build a new rig and if we can use a direct AC/AC conversion system between the two motors that would be preferred.

Thanks, Ian

 

[Skogsgurra]

I have seen lab models for many years. Excellent to write a thesis on. But never seen a practical inverter yet. And definitely no commercial units that I know about.

What power range are you in? And frequency?

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[ScottyUK]

Yaskawa have one:

http://www.yaskawa.com/site/Products.nsf/products/Industrial AC Drives~matrixconverter.html

----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[Skogsgurra]

Thanks for that piece of news! Very useful.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[LionelHutz]

I don't see how that would work. There has to be external input power.

From what I can tell, A matrix converter just uses active switches instead of the diodes on the 3-phase input to keep the highest and lowest voltage lines connected to the DC buss points. It also allows the DC buss to ripple with the AC input voltage instead of trying to smooth the ripple with capacitors.

In your test situation, the power to feed the losses comes from the AC line and then the DC buss connection transfers the "load" power. I don't see how an AC Matrix Converter would help your application besides using new technology. Maybe you're looking to eliminate the DC buss interconnection?

 

[murpia]

LionelHutz:

The test rig does have external input power, from devices under test. The two permanent magnet 3-phase motors are acting as a kind of electric transmission.

The reason for suggesting matrix converters is they are supposed to be smaller and hopefully cheaper than 2 inverter drives and the DC interconnection. If they are not, or don't exist, then we won't try and use one...

The motors run at between 20kW and 80kW, with a DC bus voltage about 360V

Regards, Ian

 

[ozmosis]

I am pretty sure the matrix drives (I only know of the Yaskawa as a commercially available unit) will not be lower cost.
They are incredibly complex and whilst have no need for DC link caps, this is offset in the cost of a fully active front end and inverter section.

 

[LionelHutz]

No, I mean that you have 2 VFD's right now... One of them will be used in motoring mode to accelerate the motors and the other will be used in regen mode providing a load and feeding DC back to the motoring VFD. The losses of the system come from the AC line connected to the motoring VFD.

If you connect a matrix converter between the 2 motors where will the external power to get the motors turning and to feed the losses come from?

The "traditional" VFD works as a sort of "combiner" taking energy either from the line or from the load motor to drive the driven motor.

 

[murpia]

"If you connect a matrix converter between the 2 motors where will the external power to get the motors turning and to feed the losses come from?"

From the devices under test. They have their own separate power source.

Regards, Ian

 

[LionelHutz]

The motors have some other mechanical device driving them?

 

[murpia]

Yes. Think of the two motors as a kind of electric CVT between two mechanical systems.

Regards, Ian

 

[LionelHutz]

OK, that makes sense. I just didn't get what you were saying before.

I doubt you'll find one suitable for your application though. You have a varying 3-phase AC power source you have to use as the source to produce another 3-phase AC power source. I'd bet any in production or even prototyping would be like the Yaskawa - one side is a constant voltage and frequency.