N Scale Helper Units and DPU's

[tehachapifan]

Complicated issue, isn't it? Even adding weight to cars, which certainly makes sense on one hand, might make them want to push even harder on cars ahead when going down grade...potentially compounding the problem.

Appreciate all the suggestions thus far!

[Ed Kapuscinski]

In any case it's a good practice to avoid placing very light cars (or mixed long/short overhangs) in from of the DPUs.

Pffft.

[ Guests cannot view attachments ]

[jagged ben]

I've always controlled my helpers/DPUs separately (using a two-knob Digitrax throttle, fwiw).     But it doesn't help you breath easier though.  It's a challenging in-real-life video game no matter how you do it. 

Prototype helper/DPU operations do NOT involve simply setting the helpers/DPUs to the same throttle setting as the lead consist.  They have to respond to the often constantly changing difference in grades between the two locations.  (My understanding is that DPUs do with computers what helper engineers do with brains, but basically the same thing.)   Some trains are going to be more forgiving than others, but it's no different on the model.  What I usually do is pick a rational spot in the train in front of the helpers where I want them to be pushing the cars instead of getting pulled by the lead engines.  (This is on an uphill.  Downhill is kind of the opposite.)  Basically I set the lead engines to the speed I want the train to move, and then adjust the helpers faster or slower to get that keep that shift in coupler slack/compression around the right spot.  As you learn to do it, you learn to anticipate stuff that happens at certain spots.  For example, if the lead engines hit a grade while the helpers are still on the flat, you'll need to either throttle up the lead or throttle down the helpers, until the helpers hit the grade and then you do the opposite.

I was under the impression that consisted decoders turn off BEMF.  Maybe this is only true for certain decoders? (Digitrax?)   Interested to learn more about that.

[tehachapifan]

I've always controlled my helpers/DPUs separately (using a two-knob Digitrax throttle, fwiw).     But it doesn't help you breath easier though.  It's a challenging in-real-life video game no matter how you do it. 

Prototype helper/DPU operations do NOT involve simply setting the helpers/DPUs to the same throttle setting as the lead consist.  They have to respond to the often constantly changing difference in grades between the two locations.  (My understanding is that DPUs do with computers what helper engineers do with brains, but basically the same thing.)   Some trains are going to be more forgiving than others, but it's no different on the model.  What I usually do is pick a rational spot in the train in front of the helpers where I want them to be pushing the cars instead of getting pulled by the lead engines.  (This is on an uphill.  Downhill is kind of the opposite.)  Basically I set the lead engines to the speed I want the train to move, and then adjust the helpers faster or slower to get that keep that shift in coupler slack/compression around the right spot.  As you learn to do it, you learn to anticipate stuff that happens at certain spots.  For example, if the lead engines hit a grade while the helpers are still on the flat, you'll need to either throttle up the lead or throttle down the helpers, until the helpers hit the grade and then you do the opposite.

I was under the impression that consisted decoders turn off BEMF.  Maybe this is only true for certain decoders? (Digitrax?)   Interested to learn more about that.

thanks! I certainly don't want to take on that kind of operational headache if that's what it takes. I'm assuming this is with locos that haven't been de-rated with fewer powered wheelsets as described earlier?

[jagged ben]

Yes, they are all fully powered units.  I've mostly only done it with really reliable motors such as Kato Dash-9s.   

Personally I love the challenge, and it allows me to run longer trains on the club layout than would be possible otherwise, which is fun.   But to each his own! 

[jereising]

Russ, all I can do is relate my experience.

I have a 114 car earthworm with three FVM ES44DCs in front and one FVM ES44DC as DP in back, only mods to the locos was to set up the pusher to show a dim headlight to the rear per prototype.  I have 2% in and out of staging.  The first 90 or so cars have MT True Scale Couplers and all have metal wheels.  The first ten or fifteen have extra weight.  I ran the locos solo as a test and arranged it so the faster one leads, the slowest is the pusher.  Didn't do any speed matching other than that, decoders are the Digitrax plug-ins.

This is possibly my overall best running train.  I have never had a stringline, the worst thing has been a coupler separation along the route, usually not repeatable.  The thing just runs.

Of course now that I have bragged it I'll be very nervous next time I run it...

[ednadolski]

Hmm. I've not seen this, and I regularly run consists of 3-4 locos, sometimes of differing makes but nonetheless speed-matched. Is your observation across different decoder brands (or firmware revs)? I can imagine this being a problem induced by different algorithms. I run entirely TCS, or at least everything I expect to m.u. has TCS.

I've always used Digitrax, but we're talking all cheap, consumer-grade commodity electronics and mechanisms anyways, so I expect results will vary widely.

Ed

[superchief]

On my Santa Fe-All the Way layout I have a helper district that is 2% up and 2% down over "the hill". I use a Walther's 2-8-8-2 mallet in one direction (which could not pull itself out of a wet paper bag) and on the other side of the hill I have an F7 A-A. One loco is a dummy all motor and gears removed and the other is a DCC friendly unit that is very light. Most of my thru Red ball trains are 25 - 30 cars long and have an F7 A-B-B-A up front (these are the old Kato, very heavy, milled out only for a DZ 123 decoder) only the first 2 unit are powered and last 2 have all gears and motor removed. With no helper the train will pull about halfway up the hill before it stalls, the helpers are "just" enough to "push it over the hill. I can't remember the last time we had a derailment on the helper district.

*on a note when the FP45 came out with sound I bought thinking it would be great with sound and dynamic brakes!!!!! I shoved and entire 30 car reefer train off the tracks!!!! That's when I came up with this idea!!
                                                                                                                               Gordon Bliss- Santa Fe-All the Way

[SD452XR]

I've always controlled my helpers/DPUs separately (using a two-knob Digitrax throttle, fwiw).     But it doesn't help you breath easier though.  It's a challenging in-real-life video game no matter how you do it. 

Prototype helper/DPU operations do NOT involve simply setting the helpers/DPUs to the same throttle setting as the lead consist.  They have to respond to the often constantly changing difference in grades between the two locations.  (My understanding is that DPUs do with computers what helper engineers do with brains, but basically the same thing.)   Some trains are going to be more forgiving than others, but it's no different on the model.  What I usually do is pick a rational spot in the train in front of the helpers where I want them to be pushing the cars instead of getting pulled by the lead engines.  (This is on an uphill.  Downhill is kind of the opposite.)  Basically I set the lead engines to the speed I want the train to move, and then adjust the helpers faster or slower to get that keep that shift in coupler slack/compression around the right spot.  As you learn to do it, you learn to anticipate stuff that happens at certain spots.  For example, if the lead engines hit a grade while the helpers are still on the flat, you'll need to either throttle up the lead or throttle down the helpers, until the helpers hit the grade and then you do the opposite.

I was under the impression that consisted decoders turn off BEMF.  Maybe this is only true for certain decoders? (Digitrax?)   Interested to learn more about that.

The whole thing with DP is to control the slack in the train. Sometimes you want it running the same as the headend (fence down), while other times you want to control it independently from the headend (fence up). Depending on the train length and tonnage, and the grade, and finally the speed of the track and train, you have to be able to think on the fly to keep your train in one piece. An engineer is always thinking about the slack in his train as well as planning ahead. A lot of running a train is by feel. You can feel what the slack in your train is doing. When it comes to N-scale, it would be very hard to control each end differently. Its hard to see what the slack is doing exactly. Would be curious to know with the guys that run DP or helpers in N scale where exactly in the train do the cars go from being stretched out, to being bunched up in front of the said DP or helper.   

[C855B]

... When it comes to N-scale, it would be very hard to control each end differently. Its hard to see what the slack is doing exactly. ...

Easy. When everything is still on the rails, your slack management is OK. When half the train is on the ground, not so much.   

[tehachapifan]

Russ, all I can do is relate my experience.

I have a 114 car earthworm with three FVM ES44DCs in front and one FVM ES44DC as DP in back, only mods to the locos was to set up the pusher to show a dim headlight to the rear per prototype.  I have 2% in and out of staging.  The first 90 or so cars have MT True Scale Couplers and all have metal wheels.  The first ten or fifteen have extra weight.  I ran the locos solo as a test and arranged it so the faster one leads, the slowest is the pusher.  Didn't do any speed matching other than that, decoders are the Digitrax plug-ins.

This is possibly my overall best running train.  I have never had a stringline, the worst thing has been a coupler separation along the route, usually not repeatable.  The thing just runs.

Of course now that I have bragged it I'll be very nervous next time I run it...

Thanks! That is good news that you have it working more or less on its own. This is what I want as well, with no need for me to sweat over making multiple manual adjustments depending on grade, slack, etc. Still, my recent test shows that having a fully-powered loco is not going to work. Probably going to have to "de-rate" one and run some more tests. Causing even one derailment, even if it's 1 out of 100 laps, is unacceptable for me to adopt adding helpers (things run 100% derailment free now without). Appreciate all the feedback!

[jagged ben]

The whole thing with DP is to control the slack in the train. Sometimes you want it running the same as the headend (fence down), while other times you want to control it independently from the headend (fence up). Depending on the train length and tonnage, and the grade, and finally the speed of the track and train, you have to be able to think on the fly to keep your train in one piece. An engineer is always thinking about the slack in his train as well as planning ahead. A lot of running a train is by feel. You can feel what the slack in your train is doing. When it comes to N-scale, it would be very hard to control each end differently. Its hard to see what the slack is doing exactly. Would be curious to know with the guys that run DP or helpers in N scale where exactly in the train do the cars go from being stretched out, to being bunched up in front of the said DP or helper.

Great comment.  Thanks for reminding me about the 'fence'.

Your first sentence applies to the model as much as the prototype.    As far as seeing the slack, I don't find it too difficult to see with typical Micro-trains N scale couplers.  It does require walking along the aisle following the train and looking often.   But I can't imagine it's harder than going by the feel or by interpolating drawbar pull indicators on the prototype!   I think I do a combination of spying couplers themselves for whether they are compressed or stretched, and of watching the car bodies for the slack action.  It also requires a certain level of trust going through tunnels; experience helps.

When I said that I pick a 'rational' spot in the train, here's an example.  Say I have 60 cars and 5 equally strong locomotives divided 3 front and 2 in rear, then I expect the transition from pulling to pushing to be about 3/5ths towards the back of the train, but usually erring a little farther toward the back of the train because pushing too much is more likely to lead to derailment in front of the rear engines than pulling too much in front.  So I'll pick a car about 20 cars from the back, and aim to keep the transition around that car.  If I notice the slack transition drifting farther back, I speed up the helpers (or slow down the lead), and vice versa.   Going downhill I'll usually shift to the opposite ratio, looking for the front 20 cars to be compressed and the rear 40 stretched.  When running helpers I also am more meticulous to follow other guidelines for avoiding string-lining trains (e.g. no underweighted or 89ft cars too close to front of train), and err on the side of being overpowered, not underpowered, for the total train weight. 

[ednadolski]

I have a 114 car earthworm with three FVM ES44DCs in front and one FVM ES44DC as DP in back, only mods to the locos was to set up the pusher to show a dim headlight to the rear per prototype.  I have 2% in and out of staging.  The first 90 or so cars have MT True Scale Couplers and all have metal wheels.  The first ten or fifteen have extra weight.  I ran the locos solo as a test and arranged it so the faster one leads, the slowest is the pusher.  Didn't do any speed matching other than that, decoders are the Digitrax plug-ins.

@jereising  what is the minimum radius curvature that you run this on?

Thx,
Ed

[jereising]

@jereising  what is the minimum radius curvature that you run this on?
Thx,
Ed

I'd say 24" but think it's likely closer to 23", and to make it worse it's at the top of the 2%.   Can't measure exactly because there's a mountain in the way.  If I had my druthers I'd redo that section and try to ease it a bit, but at this point it would be 'way too much work for me.
Every layout has it's problem area, that's mine.  If a train makes it around the curves routinely I'm happy.
Also note generally speaking slack does not apply with this train since it's mostly True Scale Couplers...

[robert3985]

This topic is spot-on for much of the operation I'm planning when my layout is finished enough to accommodate it, BUT with significant differences since I won't be running multiple diesel units, or exceptionally long trains.  Since I don't have any long grades in my sectional/modular portable layout, actual helpers aren't needed except to represent prototype operations with over-tonnage trains climbing the Wasatch Grade from Ogden to the top of the grade at Wahsatch, where the helpers were cut off from the rear of the train, the caboose on the rear of the helper was "kicked" back to the end of the waiting train, and the helper (note it is a HELPER...not HELPERS) works its way through spring-loaded switches to the wye, turns itself...comes back to the station at Wahsatch, awaits its orders, and then proceeds downhill running light as an extra for 70 miles, with red markers on the rear of the tender to designate the lone engine as a "train", to Ogden where it's turned and readied to assist another over-tonnage east-bound train pulled by a single Big Boy (or Standard Turbine, or Veranda Turbine) 70 miles uphill to Wahsatch again.

Since I don't have any Standard Turbines (yet) or Verandas that look prototypical enough to run (I have a single Con-Cor turbine bought brand new and only run a couple of times after adjusting its gauge to make sure it will clear my tight-NMRA turnouts), so my 30 to 35 car trains will be pulled by Athearn Big Boys, and pushed by Athearn 700 Class Challengers...not because the train is so long it needs a helper, but because of the helper operation, including cutting the helper off, and kicking the caboose, turning the helper and running light back to Ogden.

In actual operating sessions, this would require an engineer for both the lead power, and the helper power...but, until then, I need to be able to do it alone with a single throttle.

After reading some of the solutions here, a couple sound promising, such as using my Digitrax DT500D Super Throttles to run trains with an un-consisted helper on the rear, using the two knobs to control the speeds of both lead and helper engines separately.  Observing "slack" in the trains as they move along the mainline will be impossible since I'm equipping all of my cars with MTL True Scale couplers, which have virtually no slack...and no slinky-ing either.  The TS conversion requires that I body mount the couplers, so I won't have any truck mounted coupler problems to deal with. 

However, de-rating the Challenger helpers is a problem.  Thinking about it, I could de-couple the gearing for the front six drivers, which would probably de-rate the 4-6-6-4 somewhat, but perhaps speed matching my 7000 Class Challengers to run just a tad slower or a tad faster after removing the rubber traction tires would allow just enough slipping to neither push or pull cars off the track in front of the helper.

I still haven't got my layout up in its new layout room after moving 2 years ago, so I can't test any of my theories right now.  Maybe this will give me the kick in the butt I need to take a day and set the layout up.

Lastly, I didn't read in any of the postings that removing traction tires from helpers might be necessary, or at least helpful.  Since I am not familiar with modern diesel power, nor with N-scale models of such, maybe traction tires on diesels are rare, so this would mainly apply to steam engines???

Thoughts???

Cheerio!
Bob Gilmore