Bolt on adhesion: to overmod & other electrical people

[Randy Stahl]

I've been asked to come up with some ideas for a bolt on adhesion system for a fleet of ALCO C-420s. It'll be fun to use the forums for a sounding board, there should be alot to learn and have some fun doing it . I already mentioned using the EM 2000 and that plan was rejected because of cost. Lets hear some of your ideas.
Randy

[Overmod]

I just got done recounting that early-'60s idea about the extended axles, cranks, and siderods. Even has Alco origins! And in a sense you couldn't get semantically closer to a "bolt-on" system than one that uses actual bolts (although admittedly they're likely to be Superbolts between the cranks and axles ;-})


Seriously: Are you planning to implement microslipping, or are you interested in improving static adhesion without breakaway? Are you willing to consider systems that have vertical eddy-current 'plates' on the wheels a la AEM-7s? How well can you sense the back EMF in each of the traction motors, sensitively and rapidly, without having stray-field detection problems or difficulties sealing the motor cases, etc.?

I don't need to worry about the competence of the maintenance staff, so that's a big relief.

What kind of budget are we talking about, per locomotive? And what kind of numbers, both absolute and percent improvement?

[Anonymous]

I know just about nothing about wheel slip systems, but I feel it helpfull to suggest using something that works independently of the traction motors and their power supply. A system that compares ground speed to axle speed, or compares speeds between axles. This would help avoid any problems of compatability of electrical systems. Didn't EMD have such a system using a ground facing radar?

[Overmod]

AFAIK, all the current wheelslip systems use external sensors to determine slip. The point of EMD's system is that it's completely independent of the rotating machinery, which gives you better 'baseline' precision and far less maintenance tinkering than a rotary encoder on or connected to a journal box. You could also use ultrasound or lasers, but microwaves have the right characteristics for short-range measurement, and attenuation in moist atmosphere doesn't matter at the distances involved.

Note that the EMD system is also useful for microslipping, which basically involves spinning the wheels 'faster' than adhesion speed, in conjunction with very precise rotary encoders on the wheels (Hall-effect 'teeth' near the wheelrim allows you to provide a substantial number of 'steps' and fairly cheap pairs of detector give you good offset precision within steps -- ever look at the positioning sensor on those old HP 755x-series plotters?

The point of looking at the back EMF of the traction motors is something completely different: the rate-of-change of the field strength can tell you quite a bit about what is happening in the motors, and by using fast-acting solid-state switches you can effectively modulate the field strength and armature current to save the brushes (etc.) when actually implementing some of the fancy high-frequency power modulation that most methods of 'adhesion control' of DC motors will use. PWM in motors this size will give you some fearsome inductive switching currents -- much of which will be passing through those carbon brushes...

[Anonymous]

O, is there anything I can read to learn more about wheel slip control?

[Randy Stahl]

Micro slipping ? absolutely ! I decided that using the emd K band radar is the way to go. I am pricing out a mod 3 processor...basically an SD-60. I think I can save money by not needing all the digital & analog I/Os. I agree , the only way to go is with a true ground speed sensor. I need to learn more about the ALCO system.. It's been a while, I think the ALCO has an exiter that directly exites the traction generator, this may be a disadvantage i.e. exite the exiter to exite the generator I'm thinking that I may not be able to make field current corrections quickly.
Randy

[oltmannd]

Randy-

How about transplanting from a late Dash 7? The CHEC/Sentry system would get you what you're looking for. Might be easier to swap the whole wheelslip and excitation system out than to try to patch a wheelslip system in. It would also leave you with a more modern and better supported design - no having to scrounge old "blue face" cards up when things go bad.

Wonder if you could get the stuff cheap as there's not much of a market for used GEs. You would have to add the traction motor speed sensors, though. Not too hard to do since you're already dealing with 752 motors.

Just curious, but what do you need to increase adhesion on a C420? At 24% adhesion, you'd probably be "in the red". Maybe you'd need AF motors, too?

[Randy Stahl]

The G.E. system is an affordable option, I have most of the parts to do the mod, however I'm not impressed with the loading characteristics of the dash 7 and I don't really want to recable the ALCOs. The dash 7s have the traction motor current modules
underneath the locomotive, about half way back below the running boards. I think the traction motors will be OK, I'm not changing the horsepower of the locomotive, the currents and voltages will remain in the ALCO design specs , only the wheels will slip less.. I hope.
Randy

[oltmannd]

QUOTE: Originally posted by Randy Stahl

The G.E. system is an affordable option, I have most of the parts to do the mod, however I'm not impressed with the loading characteristics of the dash 7 and I don't really want to recable the ALCOs. The dash 7s have the traction motor current modules
underneath the locomotive, about half way back below the running boards. I think the traction motors will be OK, I'm not changing the horsepower of the locomotive, the currents and voltages will remain in the ALCO design specs , only the wheels will slip less.. I hope.
Randy

The early Dash 7s had the CMR system with that worked off motor current comparison - it was just as "lousy" as IDAC/WS10. The newer ones with Sentry just compare motor speeds. Also, know you can adjust HP on the card face on the CHEC excitation system, but am pretty sure you can adjust the the 3 slope loading curve, as well. I'd think with some playing around, you could actually make the unit run better and smoke less.

[Overmod]

Randy, have you looked into the GE BrightStar retrofit system? Seems as if they implement CHEC excitation via what I believe is xNIX-based software running on a 486 microprocessor (which also handles the wheelslip calculations). My impression was that they were doing fast field modulation (probably with IGBT variable resistance in the leads from the exciter?) That would certainly give you good, and variable-from-the-cab, control over the loading curves. Then all you'd need would be spot meters for the TMs to be sure nothing was being given 'more than it can handle'.

I still think that some method of reducing instantaneous current at the time the brushes make contact with the comm may be a good idea, whether that gets done in a main microprocessor or down at each traction motor... am I just being old mother hen?

Randy, are you more interested in TE enhancement or slip elimination? The methods and programs are a bit different. How much of the existing Alco plant do you want, or need, to keep?

[Randy Stahl]

Hmmmm, The bright star system is one I havn't though of. The RR wants an improvement in adhesion.. Above 20 percent. The price of the upgrades shold not exceed the value of the locomotive, which unfortunatly is quite low. I'm looking at a price tag of around $7000.00, not including labor.
Randy