A software failure is a possibility, though the likelihood of the failure is inversely proportional to the care that went into writing the software.
I would also wonder about the physical controls in that they may or may not give prompt and accurate feedback of what command the controls are responding to. Worst example was the Airbus crash over the Atlantic ocean a few years back, the captain didn't know the first office was applying full node up on the side stick - in a plane with control yokes, it would have been immediately abovious what was going on and the captain could tell the F.O. to let go of the controls. A more recent (and rail related) was when the woman drove into the pat of the Metro North MU train - there was some speculation that the confusing nature of the controls led her to think she was in reverse.
As for your comment on VFD and l-o-n-g cables - makes perfect sense that you would want to avoid resonances. The remote location of the VFD versus the pump also explains why 18 levels of DC link voltage was practical.
- Erik
My thought was that the engineer thought ACS 64 was in dynamic brake mode, but was really still in throttle mode. Instead of applying regenerative braking the locomotive accelerated.
A software failure could account for this.
Z manAnyway, back to trains - based on what I have read, I believe that the Amtrak 188 accident could have been the result of a malfunctioning regenerative braking system - has any one heard anything about that?
The only issue I've heard isn't exactly a 'malfunction' in the usual sense of the word. I believe the ACS64 is set up for blended braking, where a combination of air and dynamic/regenerative braking would be 'automatically' used if the train air were put in emergency.
The consist brakes were originally designed as ECP, but the fast electric actuation appears to have been removed (there are people on list who will know exactly when and why).
I can easily see the locomotive developing momentarily much higher braking force than the train, which might produce severe run-in force between the locomotive and the adjacent car, precisely where the dramatic 'buckling' of the consist seems to have occurred.
Thanks for the reply
You have a lot of insite into power elecronics.
The main reason harmonics are such a problem for us is cable resonance. The umbilicals we use can be anywhere from 5 to 20 miles long and the first resonance point is in the 800ish Hz range, so its real easy to set the woods on fire
Anyway, back to trains - based on what I have read, I believe that the Amtrak 188 accident could have been the result of a malfunctioning regenerative braking system - has any one heard anything about that?
Z man
Welcome to the forum!
I first heard of cycloconverters as part of a Power Electronics course some (sigh) 39 years ago. At that time, the dominant device for power electronics was the standard thyristor, which needed some means of external commutation.
Interesting that low harmonic content is so improtant in your application. My guess is that the pump motors are much larger than locomotive traction motors and that eddy currents induced by harmonic currents would create severe design headaches. I've seen references to inverter grade motors, presumably wiring changes (e.g. Litz wire) and extra thin laminations which would be impractical on a very large motor.
Anoher reason for using multiple DC busses for low harmonics instead of PWM is to allow for use of very large IGBT's. The Westcode subsidiary of IXYS makes one rated at >2000A and >4000V, but switching energy limits its use to less than 1kHz.
1) The short answer is (as stated in earlier posts) the DC bus stores energy, allows dynamic braking and provides a clean starting point for constructing a wave of any frequency.
2) To say that the power is converted to “DC” is an over simplification – in the converters I work on there is not one, but 18 separate DC busses which are used to make “cells” that are used to construct an AC waveform with a very low harmonic content. I am sure that locomotives are different, but the point is that the actual way the VFD is put together is probably more complicated than a simple rectifier. Take a look at http://en.wikipedia.org/wiki/Variable-frequency_drive and http://www.vfds.com/blog/what-is-a-vfd
3) There is an AC to AC technology that is starting to emerge in the market called a “Matrix Converter” which is a modern version of the Cycloconverter mentioned in an earlier post – take a look at http://en.wikipedia.org/wiki/AC/AC_converter - Based on what I know, I don’t think a Matrix converter is the best choice for a locomotive, but there may be a way to make it work.
daveklepper and absolutely correct. converting from one frequency to another is easier done by going to dc first. and when the second frequency must be variable, since the locomotive does not travel at constant speed, it would really bevery comples indeed. Add regenrative or dynamic braking, and you have another comp;lication.
and absolutely correct. converting from one frequency to another is easier done by going to dc first. and when the second frequency must be variable, since the locomotive does not travel at constant speed, it would really bevery comples indeed. Add regenrative or dynamic braking, and you have another comp;lication.
Another bonus: The software controlling the inverters has become very smart! Very specific torque control is possible, and thus is possible to actively control (minimize / prevent) wheel slip! All can be done without tachometers on the axels!
from the Far East of the Sunset Route
(In the shadow of the Huey P Long bridge)
Redore AC traction motors turn at an RPM proportional to the incoming frequency. Alternators produce at a frequency proportional to the RPM they are driven at. The two frequencies are rarely if ever the same. Converting to DC and then back to AC at a different frequency provides a "common denominator" to link the two. It's the easiest way electrically to do this.
AC traction motors turn at an RPM proportional to the incoming frequency. Alternators produce at a frequency proportional to the RPM they are driven at. The two frequencies are rarely if ever the same.
Converting to DC and then back to AC at a different frequency provides a "common denominator" to link the two. It's the easiest way electrically to do this.
+1 Nice and simple!
-Don (Random stuff, mostly about trains - what else? http://blerfblog.blogspot.com/)
timz caldreamer why is the AC current converted to DC then back to AC before being directed to the traction motors? Nobody has answered the guy's question. When the locomotive is trying to produce full power, the diesel itself is running at full speed-- i.e. constant speed. Which means it's producing constant frequency AC. Like any other induction motors, AC traction motors on today's diesels run on AC with a frequency roughly proportional to the traction motor's RPM. If the locomotive accelerates at full power from 20 to 80 mph, the AC frequency to the motor likewise has to multiply by four.
caldreamer why is the AC current converted to DC then back to AC before being directed to the traction motors?
Nobody has answered the guy's question.
When the locomotive is trying to produce full power, the diesel itself is running at full speed-- i.e. constant speed. Which means it's producing constant frequency AC.
Like any other induction motors, AC traction motors on today's diesels run on AC with a frequency roughly proportional to the traction motor's RPM. If the locomotive accelerates at full power from 20 to 80 mph, the AC frequency to the motor likewise has to multiply by four.
And this is still begging the question that the locomotive is supposed to be producing full power, but the AC from the traction alternator is anything but constant power; in fact, half the time it's effectively negative in producing net torque unless commutated or phase-switched.
Transversion to DC involves more than just 'rectification' -- there is energy storage in both capacitance and inductance to smooth out any ripple. Then when it is time to synthesize the precise phases of AC that are necessary to produce maximum effective torque in the motors at a given moment, there is full source power (on the DC link) for synthesis of each phase at full power.
All you really have to do is look at the original three-phase electric locomotives that were built toward the beginning of the 20th Century to see the result of designing something that is dependent on the output of a traction alternator (corresponding to the fixed-frequency AC supply), for whatever the output frequency corresponding to a 'notch' on the diesel-engine governor represents. It's quite possible to design a locomotive that achieves highest effective power from AC motors at whatever 'slip' difference from synchronous speed at each notch is involved. Up hill and down dale, there's the speed corresponding to each throttle position. But that isn't the best way to use a constant-horsepower locomotive like a diesel-electric... much, MUCH better to have the sync proportional to desired armature rotational speed... which is desired track speed, right down to locked-rotor at zero mph. And that is what synthesized-AC drive from a DC intermediate bus does.
caldreamerwhy is the AC current converted to DC then back to AC before being directed to the traction motors?
It is possible to convert directly from fixed frequency AC to variable frequency AC by using cycloconverters. The problem is that approach uses about 3 times as many active devices as used in a rectifier/inverter approach. One advantage of the cycloconverter is that it can be made with standard thyristors.
The rectifier/inverter approach isn't as bad as one might think, especially if there isn't a transformer involved. The main loss mechanism with line frequency rectifiers is the forward voltage drop of the diodes, which may be on the order of 2 volts out of 800 (implying an efficiency of 99.75%). Switching devices for inverters are getting more efficient, with a main driver focusing on reducing heat dissipation.
The next advance in inverters will likely be silicon carbide (SiC) MOSFET's. One advantage is that SiC can handle higher temperatures than silicon, allowing more heat to be dissipated from a given package size. Anoher is that the MOSFET's can switch at a higher frequency than IGBT's, allowing for filtering that would then allow for ordinary wire to be used in the motors.
Read this thread from a while back for a discussion on this subject:
http://cs.trains.com/trn/f/741/t/221349.aspx
_____________
"A stranger's just a friend you ain't met yet." --- Dave Gardner
it looks like traction motors require a frequency of 16 2/3 Hz. But don't generators produce AC current at their rotation rate and doesn't their rpm vary depending on load and the power needed. So the varying and probably higher frequency AC coming from the generator must be converted to a fixed frequency the AC traction motors run most efficiently at.
greg - Philadelphia & Reading / Reading
It stabilizes the wave form of the power due to engine speed fluctuations. It basically cleans it up and makes it suitable for the traction motors.
I suspect that will be eliminated one of these days as advances are made. You lose efficency each time the power is rectified or inverted, not to mention the space and cost it involves to have rectifiers and inverters.
It's not desirable, but it's a necessary evil for now.
On AC powered locomotives, why is the AC current converted to DC then back to AC before being directed to the traction motors? Couln't the AC from the alternator be modulated then sent directly to the traction motors?
Our community is FREE to join. To participate you must either login or register for an account.