Dave,
I'm not aware of any locomotives drawing AC power that do not use line frequency transformers. My comments are derived from reports of using SiC devices to replace standard distribution transformers with high frequency transformers. Think switching power supply versus transformer/rectifier power supply. The resulting unit would probably be less efficient at full load than a traditional transfer, but possibly more efficient at light load.
I do remember you mentioning your work at Mystic Transformer.
Most new catenary in USA is 25,000, 60 Hz.
Despite the electronic appriach being lighter, are there any locomotives operating, under construction or on-order that do not use transformers to draw power from that source?
When I was doing my diesel-locomotive load-regulator research at MIT on the Boston and Maine, a part-time job as transformer designer at Mystic Transformers Winchester, MA helped pay the bills.
bogie_engineer EMD's AC locomotives since the first with Siemens inverters use up to 2,800V on the DC Link depending on power level commanded. This allows for relative small cable size compared to DC locos, a major cost savings.
EMD's AC locomotives since the first with Siemens inverters use up to 2,800V on the DC Link depending on power level commanded. This allows for relative small cable size compared to DC locos, a major cost savings.
The original Siemens inverters used GTO Thyristors which are slow compared to the IGBT's and FET's used on more recent inverters. My understanding was that the Siemens inverters did pulse width modulation at low speeds which allowed for constant DC link voltage while generating variable voltage variable frequency (VVVF)output. With PWM, the GTO's were switching several times faster than the motor drive frequency. At higher speeds, the GTO's switched on and then off only once per cycle of the motor drive frequency so the variable voltage of the VVVF had to be done by varying the DC link voltage.
Increasing the terminal voltage on the motors does allow for smaller motor leads, but terminal voltages require that some of the copper in the winding be replaced by insulation.
I believe GE uses about 1,500V on their DC link.
The company that I used to work for was part of GE from late 2004 to 2009 - shortly after the GE acquistion we we given a presentation of work being done by the JWTC in India on traction motors. My recollection was that the DC link voltage used by GE's inverters was 800VDC, but my memory might be playing tricks on me. I do know that the commercially available large SiC FET modules are mostly made with 1200Vbr FET's which imply ~800VDC DC link voltages. Special customers may be getting FET's with much higher breakdown voltages. Film capacitor construction gets a bit weird above 800V.
Estimated practical limits for SiC technology is something like 15 to 20kV for SiC FET's, 25kV for SiC IGBT's and 30kV for SiC Thyritsors. This suggests that a 15kVDC catenary should be no problem for the electronics, which would be smaller and lighter than the transformer needed for an AC catenary.
bogie_engineer EMD's AC locomotives since the first with Siemens inverters use up to 2,800V on the DC Link depending on power level commanded. This allows for relative small cable size compared to DC locos, a major cost savings. I believe GE uses about 1,500V on their DC link. The dual mode DM30AC's had 4 chopper modules with inductors that raised the 750 VDC third rail up to the max 2,800V; they were set to pull up to 3,900A from the third rail. IIRC, the LIRR third rail breakers trip at about 14,000A. Dave
EMD's AC locomotives since the first with Siemens inverters use up to 2,800V on the DC Link depending on power level commanded. This allows for relative small cable size compared to DC locos, a major cost savings. I believe GE uses about 1,500V on their DC link.
The dual mode DM30AC's had 4 chopper modules with inductors that raised the 750 VDC third rail up to the max 2,800V; they were set to pull up to 3,900A from the third rail. IIRC, the LIRR third rail breakers trip at about 14,000A.
Dave
In fact I raised the question of voltage in my thread on the 2040 article, which appears not to have been read since I posted it after reading my digital copy, which was before most print issues had been received.
However, since 3000 volts DC is the standard Catenary voltage in Italy and (I think) in Belgium, any European manufacturer would need to accommodate this voltage in their AC drives. For many years, Belgium had dual 1500/3000 volts DC locomotives to run through trains into the Netherlands, but this is presumably simpler now.
So 3000 volts DC would be a good voltage for "intermittent catenary" to charge locomotive batteries. I think that battery technology isn't at the stage yet that would make this practical..
Peter
NYC subways use 600 volts, 400 amps through their third rail. This runs through all 855 miles of track and all trains draw power from the third rail. During rush hours trains depart every 90 seconds from their orignination point. I grew up in NY and asked the supervisor of the 42nd street tower.
Caldreamer
The 600V is an artifact; older diesels with generators instead of alternators commonly used it.
I believe modern DC-Link in diesel-electrics with AC synthesis drive, which is what would be used in a modern dual-mode-light locomotive, is 1200-1500V. (I had thought it was lower, but was corrected here.) I suspect higher is better, up to a point, but you'd want to be very conservative not to exceed the breakdown voltage.
I believe even the most current supercap designs are still very sensitive to overvoltage -- very low voltage as they are massive electron stores with very limited dielectric thickness -- so they would be arranged in massive serial strings with fast-acting spike and noise protection, the strings paralleled to give the current-carrying ability. Integrating this with other hybrid systems on the power -- probably in a MATE-like cabbed road slug or 'full battery-electric'
At this point the likeliest catenary voltage for incremental implementation would be fixed by vertical clearance and corona -- in many places where tunnels had to be 'notched' for doublestack clearance this may dramatically limit voltage. But note that different 'island' segments need not use a common or standard power, just one that the locomotives can support.
We have discussed using voltage-to-voltage converters to get '600 to 750V' third rail compatibility with DC-Link; think of this as an option of the kind seen on four-power electrics in Europe.
As I recall, one of the advantages of dual-mode lite is that the catenary boost can be used to supplement engine power, not just replace it. That can reduce the cost of the catenary and supply while giving high continuous power up and regenerative braking down.
My take is the DC Link (bus) voltage for the inverters will be somewhere around 800V, which is a practical max set by the capacitors. I would assume that the overhead voltage would be much higher for the reasons you stated. Power electronics are at a point where the overhead could be 10kVDC or up to 50kV for AC.
I have to guestion the 600 volts figure in the article.
The formula for Watts (power) is Volts (force) times Amperes (rate of flow)
A given sized train should take X amount of Watts no matter what. If we need 1,000 watts, then 50 volts at 20 amps or 500 volts at 2 amps will do the job.
Thus early electrifications like the NYC and PRR-Penn Station, which chose 600 or so volts needed a large amperage to carry sufficient power. This required a large and heavy conductor, which is why thet used third rail instead of catenary.
Westinghouse chose 11,000 volts for its AC system (NH, GN, PRR, RDG. VGN and N&W), which meant a lighter conductor, which could use catenary.
The laws of phuysics haven't changed in the past century, so I just don't see 600 volts distributed by catenary. So where am I going wrong.....
Our community is FREE to join. To participate you must either login or register for an account.