I understand that streetcars have infinite throttles , they also have low current draw, I wonder how long a grid resistor would last at 6500 amps? I know that electric frieght locomotives have a large amount of grid resistors , I also know that using them too much spells disaster, they always have assorted blowers to try to cool them along with open cicuit protection but if you blow one the destruction is going to be major !
Imagine how many wires would be in an MU jumper if there were more than 8 notches ! Aren’t they heavy enough ? And in cold weather they are not easy to work with with 27 wires !!
Yes the old GE’s had half notches but they also had throttle relays that would stay picked up through the half notches to give a true 1-8 notch reference via the MU, AVR= A valve relay , BVR= B valve relay , etc.
I think the original designer of the 8 throttle notch system was missing two fingers
While there may have been a few diesels with “straight electric” control systems built, most of the diesel-electrics in the US used the control system developed by Lemp for GE. By using a combination of a battery winding, shunt winding and differential winding, he was able to get the generator to operate in a more-or-less constant power mode. The transitions from full series to series parallel to full parallel was to overcome the voltage and current limits for the generator. In the high horsepower extreme for generators (e.g. GP-35), field shunting became necessary to keep the voltage and current within limits.
The big problem with using resistors is that they waste energy, and they were not needed with the Lemp control.
That’s the problem with the Woodward governor. You can only set the speed for 4 notches independently. The rest just are a result. Sometimes, you can wind up with a speed schedule that’s “lumpy”.
An electronic governor has the advantage of being able to set each notch’s load and speed independently. This is particularly attractive on a roots blown EMD where you could get somewhat better fuel efficiency in the lower and middle notches.
Mention was made of the electric locomotive heritage of diesel electrics, and I should point out that streetcars, up to the PCC, had controls with eight or ten notches, the old K-type controller. Also, the earliest EMS diesels had manual transition from series to parallel and for field shunting, but starting with the F-3, transition was made speed dependent (approximately) on the basis of the voltage across the armature commutator, and automatic. Resistors are still employed for field shunting of dc motors, but not in the main power leads, and field shunting does not waste much power. All this has been replaced with electronics and very precise control of voltage and current to both field and comutator in ac-generator current production locomotives, even with dc motors. (AC motors do not need comutators.) Continuously variable control would be a lot simpler now, but still is not needed.
Even in early thyristor controlled DC motor rail cars, speed controlled relays would kick in automatically, ie 0-10 mph series 1, 10-25 mph series 2, 25-35 mph series 3, 35-50mph shunt 1, 50+ mph shunt 2. In effect this 5 notch system gave good acceleration at 2mphps…Even later IGBT controlled AC motor rail cars using VFVV would step through different modes of waveform generation as speed increases, similar to a three notch system, ie six step, quasi six step, pwm, etc…
I find it interesting how the mentality towards “self-driving” trains has shifted in the 12 years since ths thread has started. We’ve gone from a “wouldn’t that be great? The crew can just put their feet up on the windowsill and snooze” philosophy, to where… now that automating trains is possible, it’s the crews themselves who are the strategy’s greatest critics.
Building self-driving aircraft is far easier than building self-driving cars, which in turn are more difficult to do than self-driving trains. In the tested range of conditions or in the lab, with only predicted failures of maintenance, for the predicted operations. All with “perfect safety” and without requiring the engine an to stay ready to leap to fix a computer mistake … perhaps an insidious or complex one … within a few seconds with little warning or indeed no warning.
Crews often appreciate these things better than systems architects. They also don’t trust the motives of those paying for the systems they get.
As I was told, GE wanted the cachet of a more-precise 15-notch throttle (perhaps mostly for marketing appeal) while maintaining full AAR standard MU compatibility with 8-notch locomotives. As Erik noted, this was possible with additional stepping in the load regulation. At one point I saw schematics that showed how the logic worked for any combination of unit types.
The Budds didn’t really need notches as they only had two ‘speeds’ – converter unlocked and direct-drive. Presumably one of the ‘notches’ corresponded to idle with the converter empty or slipped, and the others giving simple two-relay ‘binary’ engine-governor command using the same principle of seven-power-notch ‘8-notch’ system with one fewer solenoid and no fancy triangle plate setup…
The basic Baldwin system was a stepless air throttle for engine governing, and to my knowledge this was analog between units (which again IIRC imposed limits on how many could be in one consist, etc. as noted… the response time and positive authority being reasons as noted.) I do not remember how critical speeds were handled. When they had to offer “electric MU compatibility” as an option, I think most examples, certainly those I have read about, were standard 8-notch compatible. It is possible they would have a 22-notch ‘compatible’ with two steps between each eight-notch ‘required’ position, but the
Amtrak’s F40PH’s had a notchless throttle when the diesel was running at high speed to generate the head end power. When not generating head end power they used the usual 8 notch throttle.
Lol, I was reading the “cruise control” posts on this thread and saying to myself “don’t you guys know about EMS?” Then I noticed the posts were from 2008.
I know we’ve had this discussion many times, but I’ll say it again: I can’t imagine a time when freight trains run autonomously without any human involvement. Maybe we will reach a point some day where EMS can manage in-train forces as well as an experienced engineer, but there are so many other factors involved in getting a train safely over the road…
As far as I remember, they still had 8 traction power steps with the same notched handle, just that the engine ran at a fixed 900 rpm to hold HEP frequency steady.
Older push-pull control trainlined through dedicated consists to dedicated power with 567/645/710 might not use stepless control, either. The last time I rode MN service on the Hudson Line, they were still using the rebuilt FL9s and the ‘bumps’ corresponding to throttle positions were very noticeable at the head end.
It would be theoretically possible to get somewhat more ‘stepless’ control of an F40 refitted with EFI by using one of the extended RF protocols for the engine control. However the microprocessor would still be programmed primarily to avoid the critical speeds in the EMD engine, and might still have to be arranged to use speeds corresponding to HEP frequency if using main-engine-driven alternator supply without transversion, so there might be some extended version of a ‘notch’ system in proprietary control.
My memory is that the throttle could vary the amperage smoothly without any steps just like the dynamic brake. It has been around 40 years since I ran one of those though. The physical throttle notches on the control stand definitely went away when the HEP was engaged.
I don’t know how the control of trailing units was handled. Only one unit on a train was allowed to supply HEP .
That one is new to me, but perhaps VIA sets things up differently than Amtrak, due to the high power draw from the 30 car ‘Canadian’ consists that are run during summer tourist season.
Is one HEP unit really enough to power a long consist like the Auto Train?
So I got curious and looked up the differences between the VIA and Amtrak HEP systems. I also didn’t know that GO or any other commuter operators still used the 575v system that originated on CN’s Tempo equipment back in the 1960s.
On an F40 the alternator needs to run at a constant fixed high speed not to make lots of power, but to keep the frequency of the AC HEP correct. That is a function of the Woodward governor in a non-EFI engine, and the logical thing to do would be to have Run 8 adjusted for 900rpm so that at all times, including maximum acceleration at the same time peak HEP is demanded, the engine will have adequate performance.
As load, including amp draw of the HEP system, changes, the Woodward governor, virtually steplessly, adjusts the fuel rack to hold the engine stable at 900rpm. It will do that faithfully, and mechanically, even if there is no actual electrical load on the engine crankshaft. The device connected to the notch throttle only controls engine rotational speed.
The colossal fuel burn at 900rpm is an artifact of how diesels work: just as jet engines require a large share of their turbine power just to turn the compressor, a Diesel engine uses a great deal of its fuel to keep itself turning through all the required compression. It is not a measure of increased “power” from the alternator that causes the added fuel burn. (In all probability if the fixed frequency were not mandatory, the engine could be run at a lower speed notch, for example when the ‘output for traction’ from the alternator would just match the HEP load (whatever that happens to be at the moment) and the engine could be thought to be producing comparable horsepower to a separate smaller HEP genset in one of the F40s that was so equipped at that point. That won’t work, though, if stable AC frequency is a design requirement (and you can’t use modern AC-synthesis transversion, as you can for example in some of the modern Siemens passenger locomotives…)
The load regulator, or alternator field (the thing ‘generator field switch’ drops out in three-step) is what is varied to change the electrical output of the alternator, and this could be give