Questions about lack of wheel conductivity and metals

[narrowminded]

I imagine that was a huge relief.  

I did some pretty extensive testing to come up with those numbers on my project, knowing that by virtue of the diminutive size of my loco that the weight would be one of my largest problems, right after getting the gearing low enough and getting it to move at all.    For reliable performance I found those numbers to be the range.  Per wheel, less than 5 grams became sensitive and over 7.5 grams didn't seem to have any additional benefit.  That shows again in your test.  Eight wheels is 40 grams to 60 grams by those numbers. So your base of 31 grams plus your test range of 15 to 30 grams additional equals 46 grams and 61 grams.  How very interesting. It's nice to see an independent confirmation of my own observations.   The lower number probably has a little unavoidable extra stress from the wires to overcome even though they are now much improved and in an acceptable range.  Thanks for your report and I'm glad you got it.   That's a sweet project.

[nscaleSPF2]

I'm glad that you all have found a collective solution.  This also agrees with my experience.  Extra tender weight seems to help low speed performance.  I try to add as much weight as possible, but never did the kind of scientific study as you have done.

One other point.  I too am using Atlas code 55 track, painted.  After I cleaned the top of the rails, I discovered that there was a tiny, almost microscopic, amount paint on the upper inside corners.  Not everywhere, just in some places.  This was causing the locomotives to stall.  Not all of my locomotives, just some.  I eventually made a metal scraper that followed the contour of the top inside corner, and this cured the stalling issue, with all of the locomotives, all of the time. 

I mention this because, if you were to check those places on your layout where this locomotive stalled (this time with a magnifying glass), you may find an anomaly.  There may be a very tiny amount of paint that you missed.

Hope this helps.

[mmagliaro]

I must say after reviewing all this, I do like the idea of a contoured scraper.  I already have used a flat piece of stainless steel to scrape/burnish my rail tops and after I did that some months ago, the general cleanliness and runnability of the layout did improve.  Just a little bit of "relief" in two notched
areas of such a plate would allow just barely scraping off the shoulders of the railhead without disturbing the paint, neolube, or whatever weathering I have in the sides of the rails.

[ednadolski]

Just a little bit of "relief" in two notched
areas of such a plate would allow just barely scraping off the shoulders of the railhead without disturbing the paint, neolube, or whatever weathering I have in the sides of the rails.

Another option might be to wrap some +800 grit sandpaper around a short length of cylinder (e.g. dowel, pipe, etc.) of about say 2" diameter and run that up & down the track a few times.  You just need enough diameter so that the sandpaper contacts the inside corners of the railheads but does not rub against the centers of the ties.

Ed

[mmagliaro]

Good news!
With 22.5g of added weight, the engine passes all the tests. 

Hooray!

Thank you... THANK YOU  to everyone in here.  You bothered to indulge me, to think about my problem, and to suggest
solutions.   You gave your time, your consideration and your brains.  Thank you.
It would have taken me so much longer to figure this out without your help.

Now... on to finish the tender truck detailing so I can show the next progress installment!

[nkalanaga]

Glad you solved it.  If you hadn't, my next suggestion would have been to to add track wipers.  I did that, many many years ago, to an Atlas 0-8-0 tender, and they worked.  Mine were crudely made from shim brass, and stuck under the truck screw, so I'm sure you could do a much neater job, with better materials.  Mine sat between the wheels, behind the sideframe, and weren't too obvious, although I didn't try to keep them as small as possible.  Running on Atlas code 80 track, they had to be able to survive large flangeways and rough joints.

[Kisatchie]

Thank you... THANK YOU  to everyone in here.

Hmm... I guess I'll have to
give up writing a best
selling science fiction
novel about this mystery
now...

[mmagliaro]

You know Max, I'm really puzzled and surprised with just how fussy and unreliable that all-wheel-pickup tender is.  It doesn't seem to make sense.  Do you think that a stock Kato Mikado tender (with its original weight) is as unreliable as your tender?  I don't recall if you tried using it for testing or not.

One solution (some can call "cheating") would be to use a DCC decoder with a simple keep alive circuit which would make this loco run for few seconds without reliable power pickup. They do work.    I know, it is not to be.

I wanted to get back to this question, but forgot about it.

My tender, before I added the extra weight, was 31g.  I weighed a Mikado tender (stock weight) and waddya know?  It weighed
almost exactly 31g.   A Kato Mikado tender with its stock weight it very reliable.  In fact, I had used this very Mikado tender to tug behind my engine for testing before I had my own tender built, and I don't recall any stalling problems.  But then, I wasn't excruciatingly crawling the engine at minimum speed over the whole layout, which is what I've been doing for these tests.

Also, remember that a Kato tender has trucks that are set further apart than my little tender.  That can be a huge factor.  When you go over a turnout, or a slightly uneven section of track, the odds of a Mikado tender having one truck away from the "trouble" spot are much better.

Even so, I think we all know that a Kato Mikado with bad driver pickups has "very good, but not perfect" pickup qualities.  They do stall sometimes and have some spotty problems at low speed.  You can't really get top "Kato" pickup quality out of a Mikado until you fix the drivers.  Well, once again, a Mikado has a big advantage over me.  It has more drivers, and more weight, in the engine.

And I always add weight to Mikado tenders because they do need it to get that last edge of stall-free pickup, which is what I'm trying to squeeze out of this little switcher.

Steam switchers have an almost superhuman task to do in N Scale. They have the shortest wheelbases and the fewest pickup points, and yet they are the engines that, above all others, are expected to crawl at slow speeds through yards full of turnouts and other gnarly trackwork.

[peteski]

Thanks for addressing my question Max.  Now that we know that it reall wasn't your tender that caused the problem it all makes more sense.

[mmagliaro]

Thanks for addressing my question Max.  Now that we know that it reall wasn't your tender that caused the problem it all makes more sense.

I agree.  I was much more worried that it was something like my wheelsets, the accuracy of my axle cones, or something like that.
But all along, since I could repeat the problem with Kato trucks, I did take a lot of solace in that (as in, "whew, at least I know I didn't screw the truck frames or wheelsets up after all that work!)
The wire lifting, and not having enough weight, make sense.

I am up to 25.4g added weight now, for a total of 56.4, just a smidge over narrowminded's experimentally determined value of 7g per wheel.  And I have an ace in the hole.  An oil bunker tank is going to sit on top of this tender.  When I make that, I can put a solid sheet of tungsten inside it and I think I will gain an extra 10g of weight there, so I'm going to do it.  This engine has torque to burn with that 16:1 gearhead, so pulling the effective weight of 2 freight cars behind it in the tender is not going to hurt.

[peteski]

This engine has torque to burn with that 16:1 gearhead, so pulling the effective weight of 2 freight cars behind it in the tender is not going to hurt.

That is something that I just don't understand.  I have never encountered any N scale engine which didn't have enough torque to stall the motor before the wheels start slipping.  Your engine can have gobs of torque but its pulling ability will be determined by its weights (which relates to the adhesion of the wheels to the track).  Even with a coreless motor and a gearhead (with gobs of torque) if the engine is very light, it will not be able to pull squat  because it will simply spin its wheels on the track .

[mmagliaro]

Peteski,
It's a tough call.  Although most of the time, I agree, engines slip out long before they are limited by motor torque,
I do have some different experiences.  I have seen some engines that will grind to a complete halt on a hill or a curve, running at low speed, just from the load of pulling the cars.  The Bachmann Berk I had here was really obvoius case, running on DC (so it doesn't have a decoder with back emf to keep juicing the motor when it goes up a hill).  With more than 14 cars, at I'm guessing about 20 mph, it would grind to a complete halt on my 1.7% grade unless I pushed up the throttle. 

Quite a few brass steam locos I've fixed up have this problem.  The motors are so weak that the engines badly slow down on curves because of the friction of the mechanism and the curve.   And it's even worse on hills.

I know, some slow-down on a hill is "prototypical", but I wouldn't want it to be so extreme.

Back to my engine...
The thing is, going from the 4:1 to 16:1 got me the super-uniform motion at really low speeds that I was looking for.
My point about torque was really meant to explain that I now have so much that a really heavy tender isn't going to hurt the gains in performance I got by going to the 16:1 gearhead.  It's not so much about stalling or slipping drivers.  I'm thinking about
the motion.

Yes, I am ultimately limited in total pull by slipping drivers.  But even before the drivers slip, torque matters to how smoothly and evenly the engine runs, especially at low speed.

[peteski]

Peteski,
It's a tough call.  Although most of the time, I agree, engines slip out long before they are limited by motor torque,
I do have some different experiences.  I have seen some engines that will grind to a complete halt on a hill or a curve, running at low speed, just from the load of pulling the cars.  The Bachmann Berk I had here was really obvoius case, running on DC (so it doesn't have a decoder with back emf to keep juicing the motor when it goes up a hill).  With more than 14 cars, at I'm guessing about 20 mph, it would grind to a complete halt on my 1.7% grade unless I pushed up the throttle. 

Ah, now I understand what didn't make sense to me before.  Most N scale models don't have high-ratio gearbox between the motor and wheels.  They also utilize fairly inefficient motors with iron armature rotors. At low speeds the motor is supplied  with low voltage (and as Ohm's law applies, it also uses very little current). The motor only produces fraction of the mechanical power it is designed for.  So yes, if the load increases on a slow moving train (when it climbs a hill or enters a curve) then the motor under low power will slow down or even stall.  I didn't consider this scenario.  The remedy is to simply turn up the throttle to increase the voltage to the motor which will make it produce more power and pick up the pace.  You are right, with DCC and BEMF this can easily be done automatically. 

Your solution multiplies the torque of the motor by having it run faster, even at very slow speeds of the model.   You also use very efficient coreless motor.  Because of that your model does have plenty of low end torque without any additional electronic-based compensation.

But is that truly realistic?  If you are running a real train pulled by a locomotive (either steam, diesel or electric) at a very slow speed (partial throttle) and it enters a curve or starts climbing a hill, that loco will slow down (just like a model locomotive). The engineer has to open up the throttle to give it more power in order for it to maintain constant speed.  So is your setup an accurate model of a real locomotive? 

Of course I'm just busting your cookies.  It is your model and you are doing whatever you see fit.  And now I also understand the lack of torque you were referring to.   Thanks! 

[casoken]

I'm new at posting to this board and so please bear with me. Thanks. This is a fascinating discussion re conductivity of electricity from  track to wheel on whatever locomotive power we are working with, whether artfully scratch built  or purchased ready to go. My thoughts re the fine conversationexamiing the mechanical and electrical construction involved lean me toward the  problem possibly residing mostly in the quality of the electrical aspect of rail /wheel contact that weight added does help.  My guess is that discussions about  track cleaning alone could produce a book with considerable heft to it. Like many, I have tried a variety of cleaners and non- destructive  friction cleaning devices/materials (OK, sometimes that rule was ignored!) and have now settled for liquid cleaning followed by the light use of  4B  graphite applied with a stick or pencil. This idea was passed on to me by folks operating a large HO scale club in my southwestern Ontario region and so it was given a try on my medium sized layout built in 1:160 using codes 55 and 40 rail. Graphite is quite conductive as a 4B stick set across the rails will quickly prove. My graphite items were purchased at Michaels. The 4B has enough clay in the mix  to act as a binder for the graphite (6B  may have a little too much 'clay gum' and 'smear' I suspect). This 4B mix will not fall off both contact surfaces, wheel and rail, and this graphite application has significantly  improved my experience with electrical operating reliability while reducing  the need to clean rail as often of dust, tarnish etc.

[nscale1964]

I have had the same problem in the past with making my own frame in the tenders. My solution, with the tender, was it was not perfectly straight, slight warpage in hot and humid weather was to add a couple of lead shot weights, glued them to the center of the frame and it cured my problem.