N Scale Helper Units and DPU's

[tehachapifan]

I think some of this may have been discussed to some degree before, but darn if I could find anything doing a search this AM.

Anyway, I have a modern-era, mountainous layout with up to 2 to 2.5(ish) percent grades and 20" min. mainlines. Like to run fairly long trains. Something I've always wanted was (working) helpers or DPU's but, at least prior to converting to DCC, could never get them to reliably without causing an eventual derailment (plus I wouldn't breath while running the helpers ). One stutter with a loco somewhere or a tad too much pushing or pulling and the train is on the ground. I only just recently tried again since converting to DCC, with as close to a perfectly speed-matched loco as I could get placed near the rear of the train, and again had a mid-train derailment going down the helix. This was the too-much-pushing type of derailment resulting in a jackknifed set of trucks with truck-mounted couplers.

All this said, for years I've pondered the idea of the dedicated helper engine that was modified to have less powered wheelsets. Maybe down to even just one powered wheelset. I figure this could be accomplished by simply removing some idler gears in the trucks, but never actually tried it. The idea being that maybe the engine(s) might not have enough power to push the train off the tracks, or stringline it, but would still (hopefully) provide at least a little added tractive effort.

Thoughts? Anyone try this or have success with something else? I've also wondered if making the BEMF settings in the helper units super sensitive might help. I figure body-mount couplers might also help with the truck jackknife-type derailment but I doubt I'll ever convert everything.

[greenwizard88]

If you have issues and BEMF is on, what if you disabled BEMF completely?

[tehachapifan]

If you have issues and BEMF is on, what if you disabled BEMF completely?

It's bee a while since I played with BEMF, but I believe I do have it turned off in almost everything. But, it may be worth checking.

….actually, on further thought, it might be on in some locos but probably not set very strong. I really can't picture clearly what BEMF may or may not do to help or hurt in this scenario. On one hand, having it on seems like it might help by sensing if the loco is pushing too hard or not enough but, on the other hand, may cause too much speed fluctuation resulting in too much slack or tension. I hate just messing/experimenting with BEMF setting as speed tables get messed up and have to be re-done each time. Still, perhaps this is where the solution lies.

[C855B]

BEMF is your friend in multi-consist operations like this. The whole point of BEMF is using commutator feedback to maintain constant motor RPM versus a variable load. Speed-matched locomotives all with the same BEMF settings should be much more dependable for split ops.

[tehachapifan]

BEMF is your friend in multi-consist operations like this. The whole point of BEMF is using commutator feedback to maintain constant motor RPM versus a variable load. Speed-matched locomotives all with the same BEMF settings should be much more dependable for split ops.

Thanks. I was modifying my previous reply above when you replied. While I've never seen helpers discussed specifically, I do seem to recall discussions on various forums saying its best to turn BEMF off when running locos in consist. Otherwise, they all keep pushing and pulling against each other or something along those lines. Perhaps some brand of decoders handle this better than others? I have mostly ESU decoders and, the more I think about it, I think they're either still set at the factory BEMF setting or whatever results from their "automatic" adjustment feature and they run great in consist.

[Bill H]

Hi Russ:
Any chance it could be the "slinky effect" from the MT couplers starting an oscillation that just magnifies itself and not the engines?

Kind regards,
Bill

[tehachapifan]

Hi Russ:
Any chance it could be the "slinky effect" from the MT couplers starting an oscillation that just magnifies itself and not the engines?

Kind regards,
Bill

I don't think so, as trains run derailment-free without helpers.

[Kuchler10]

I have done some basic testing with the "derated" helpers by only powering one truck. It was successful in terms of not derailing even when the lead engines cut out, but it was on flat track. I would definitely recommend testing this method, as it is very reversible if it doesn't work out for you. I read in it the N Scale Magazine article titled "Helpers!" by Mike Danneman about his fantastic layout. It is the July-August 2012 issue. His layout is perhaps my favorite that I have seen in print.

[djconway]

My thoughts on this:
1~ get all the car's in the train up to NMRA recommended weight (1/2 oz + 0.15 oz per inch)
2 - body mount all couplers, this allows the trucks to turn and not climb the outer rail
3 - make sure you cut the helpers in beyond what the lead engine can pull, and make sure the helper can not push all of the cars in front of it.  The helpers have to be realy needed.

[tehachapifan]

I have done some basic testing with the "derated" helpers by only powering one truck. It was successful in terms of not derailing even when the lead engines cut out, but it was on flat track. I would definitely recommend testing this method, as it is very reversible if it doesn't work out for you. I read in it the N Scale Magazine article titled "Helpers!" by Mike Danneman about his fantastic layout. It is the July-August 2012 issue. His layout is perhaps my favorite that I have seen in print.

Thanks! Glad to hear this idea may indeed work. Thanks for the article info. Will need to look for that.

[ednadolski]

I read in it the N Scale Magazine article titled "Helpers!" by Mike Danneman about his fantastic layout.

@pbrennecke does this too on his amazing Grand Road layout.  It even works with regular DC and MT truck-mounted couplers.

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

Ed

[ednadolski]

One stutter with a loco somewhere ...

Groupings of 2-3 DPUs will help in this case, so if one stutters the others can push/pull it without putting as much stress on the adjacent cars.

I've found BEMF not to be a great solution, as the locos in a consist can end up fighting each other.  Then if you put them into a DPU, they can take their aggression out on the adjacent cars (and guess who will lose).

Likewise, speed-matching has its limits.  The tuning tends to drift over time and eventually has to be re-adjusted.

Ed

[C855B]

... I've found BEMF not to be a great solution, as the locos in a consist can end up fighting each other. ...

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.

[peteski]

Groupings of 2-3 DPUs will help in this case, so if one stutters the others can push/pull it without putting as much stress on the adjacent cars.

I've found BEMF not to be a great solution, as the locos in a consist can end up fighting each other.  Then if you put them into a DPU, they can take their aggression out on the adjacent cars (and guess who will lose).

Likewise, speed-matching has its limits.  The tuning tends to drift over time and eventually has to be re-adjusted.

Ed

I don't have extensive experience with this, but I recall lengthy discussion on the ESU Yahoo forum about BEMF and consisting not playing well. Solution was to either reduce the effect, or disable it totally.  I also agree that there is a limit as to how well you can speed match multiple locomotives.

[Maletrain]

My thoughts on this:
1~ get all the car's in the train up to NMRA recommended weight (1/2 oz + 0.15 oz per inch)
2 - body mount all couplers, this allows the trucks to turn and not climb the outer rail
3 - make sure you cut the helpers in beyond what the lead engine can pull, and make sure the helper can not push all of the cars in front of it.  The helpers have to be realy needed.

I will second the advice about making sure that the helpers are really needed.  But, our club successfully runs 100 car N scale trains up and down grades (no helix) with a lot of truck-mounted couplers and a variety of car weights that do not all meet NMRA specs.  The guy who set it up simply joined together several trains that were running at the same speed, coupling the loco in one to the rear of another to make a mid-train helper. 

I think things might be tougher for actual pushers, located on the rear of the trains.  Those rear locomotives are not pulling anything, so they are free to bunch cars towards the middle of the train.  That is not a problem so long as they don't bunch the cars all the way to the lead locomotive(s).  On level track, it is pretty simple to put cars in front of the to-be-pusher loco and see how many it can push, then make the train longer than that by a decent amount.  But, going down a steep grade, where the cars will descend on their own without any locomotives at all, it is probably important to make sure that the rear locomotive runs just a tad slower than the head-end locomotive at the same speed steps.

I am not sure how back-EMF performs in the pusher situation.  I think we want the pusher to run a tad slower than the lead loco going down hill, but not up hill.  So, if the lead loco does slow down a little under load, then the pusher helps, and if the lead loco can pull the whole train, then it makes the pusher run just a bit faster because it has some help.  That sounds like we do not want perfect back-EMF.  But, if back-EMF were successful in making the two locos run at exactly the same speed under every condition, then maybe that would be sufficient to keep the locos spaced properly apart to neither completely bunch or completely remove all slack from the consist. 

Anybody else with some practical experience on helixes?