New AC traction tech may be game changer ?

[blue streak 1]

There keeps being improvements in power modules.  If this works EMUs may become all AC . 

http://www.railjournal.com/index.php/high-speed/jr-central-tests-sic-power-modules-on-shinkansen-train.html?channel=523

 

 

[MidlandMike]

Is the 3.3kV mentioned in the article at he catenary?

[Wizlish]

MidlandMike

Is the 3.3kV mentioned in the article at the catenary?

http://www.mitsubishielectric.com/news/2013/1225.html

The nominal catenary voltage is given as 1500 VDC.  I believe the 3.3kV is the maximal peak voltage the SiC inverter will handle without damage.

Note that a lower-voltage version of this (the 600 to 750VDC range) had been produced in 2011.  That suggests that transit equipment could be going 'all-AC' too...

 Here are some papers on the advantages of SiC (these are links to the actual PDF papers and not abstracts)

http://www.ornl.gov/~webworks/cppr/y2001/pres/111896.pdf

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.457.1060&rep=rep1&type=pdf

http://www.researchgate.net/profile/J_Rabkowski/publication/254060210_Silicon_Carbide_Power_Transistors_A_New_Era_in_Power_Electronics_Is_Initiated/links/544759570cf2d62c30505db8.pdf

https://www.pes.ee.ethz.ch/uploads/tx_ethpublications/16_SiC_versus_Si_evaluation.pdf

 

[NorthWest]

The CTA 5000s are AC. With the improvements in technology, I would not be surprised to see a transition to all new AC powered rail vehicles in the next decade.

[erikem]

I'm guessing that the 2.5kV AC power input is fed to a rectifier before being fed into the SiC modules. The SiC devices that I am aware of only hold off voltages being applied in one direction since they are is pretty much the same "vertical" layour as existing silicon FET's and IGBT's (vertical means the current flows through the thickness of the device wafer). The "lateral" geometry used by the GaN power FET's appears to be more compatible with the "all AC" converter design, but those FET's don't have the voltage or current ratins of SiC FET's.

There are two advantages of SiC FET's over silicon IGBT's. One is that the SiC FET's can operate at much higher switching speed than IGBT's, which should allow for relatively lightweight filters to provide almost pure sine wave power. The other is that SiC devices can operate at higher junction temperatures than silicon, making cooling a bit easier.

 - Erik

[Wizlish]

erikem

There are two advantages of SiC FET's over silicon IGBT's. One is that the SiC FET's can operate at much higher switching speed than IGBT's, which should allow for relatively lightweight filters to provide almost pure sine wave power. The other is that SiC devices can operate at higher junction temperatures than silicon, making cooling a bit easier.

Another PDF reference that shows this in a different context:

http://engr.case.edu/bhunia_swarup/papers/J/J8.pdf

[Buslist]

Don't see this as a game changer, just another step in the evolution of AC traction. Most all new EMUs are AC already.

[blue streak 1]

Erikem:  Thanks for the lesson.  I find the idea of lighter weight inverters has much potential.  But the lighter weight traction motors means less unsprung weight which may be important.  

Probably several generations of this tech before it may live up to promise.

All this change makes me want to go stuff my first EE instrucctor's words down his throat who said that there was not much that would be changed in the future as the tech was very mature and had reached practical limits. 

[erikem]

Streak,

You're quite welcome. Silicon power technology is relatively mature, with evolutionary versus revolutionary progress since the SCR's of the 60's, HexFET's from the late 70's and IGBT's from the early 90's. High voltage silicon FET's are pretty much at theoretical limits for performance.

The interesting part of the origonal article was reduced motor weight, higher frequency operation of the switches would allow for higher RPM motors. The weight of an electric motor sort of scales with the raw (i.e. before gear reduction) output torrque. The  tradeoff is likely a bit higher gear loss due to the higher reduction ratio, which might be traded off by using a modified nose suspension versus truck mounted motors.

It would be interesting to what new MU cars for the Lackawanna lines would have been like if the 3kVDC electrification had been kept. Lightweight AC motors and no transformer would have made for some light cars.

Buslist,

The SiC mechanical logic circuity doesn't look like it would be readily applicable to power applications, still interesting to see it operate at 500C. I do remember seeing a Belgian company specializing in packaging SiC components for operation above 200C ambient, don't remember how much above 200C.

Then again, silicon is a high temperature semicinductor when compared to germanium. There was an oil services company that started making silicon transistors for well logging, but they made the mistake of showing the devices to Convair when Convair was working on the Atlas missile project - the Air Force told the company to set up a production line for the Atlas - amazingly enough, Texas Instruments is still in business...

 - Erik

P.S. 25 years ago, I would have bet that diamond would be the hot power semicondutcor material, but it looks like GaN for under 600V and SiC for over 600V.

[CSSHEGEWISCH]

While lighter weight of traction motors is a plus, it may not reduce unsprung weight that much on rapid transit equipment.  CTA and other transit operators do not use nose-hung traction motors, preferring a right-angle drive similar to an automotive drive shaft.

[rcdrye]

CTA did a lot of testing on motor and truck layout in the 1960s and 1970s before settling on the current one, which is essentially the CTA-1, a modernized PCC layout.  At least one nose-hung design was built for testing, known as the CTA-2, and was not considered worth developing.

[erikem]

Truck mounted motors for transit go back a long ways, Westinghouse was advertising the W-N drive in the 1920's.

Not sure if nose suspenion would make sense for a LRV, can make sense for an EMU and usually makes sense for a locomotive. A small high speed traction motor requiring a double reduction could be mounted with its center of mass much closer to the nose end than to the axle. If movement of the nose end was much more constrained than the axle, this would decrease the effective unsprung mass more than what the raw ratio of weight supported by the nose versus weight supported by the axle.

I'd also expect even a double reduction gear to have less gear losses than a right angle drive would have.

- Erik