We're talking HO, right?

Scarpia,

We're talking HO, right? What length of cars will be typical? 24" R is okay for 40 footers and short locos, but it's kinda tight for longer stuff.

And a 24" radius helix is really asking for it - with the entire train on a curve, a 24" radius curve will add almost 1.5% extra drag to any helix grade (2.5% helix grade being typical in HO) - meaning any train on a 24" radius helix will act like it's on a 4% grade - not good unless you intend to only run very short trains with short locos and rolling stock.

If you handlay the helix track and put the rails at maximum gauge throughout you might get a bit better performance with less drag. The better option is to go to at least 4x your typical rolling stock length for a helix, to put it in terms of the radius article in issue 1.

HO it is

HO it is, Joe.

I understand the limitations of 24", and the added effect to the grade in the helix, and I appreciate you reminding me.

But this layout is again nothing more than a test - a bit more advanced than my last one, but not the production version. I'm not planning on keeping this thing around forever, and am therefore not overly concerned about running specialized equipment, or really long trains on it down the road.  My current roster is mostly 40', but everthing I own currently runs on my 22" radius and 4% grade initial test layout, (including my modern newsprint cars), so running that same equipment on the Beta version with 24" and less than 2% grades shouldn't be a problem - especially if I manage to not duplicate the kinks in the alpha.

You mention that 24"=1.5% additional drag. I'm presuming that this means a 24" helix at 2% grade equals a 3.5% grade on a straigth length of track. May I ask how this is calculated? Did I miss that in the radius article? Is that a universal ratio, or does it vary based on the length of the equipment? 

As you know, folks learn differently, and I'm a doer/reader (need to give it a whirl to cement it). I may not even do a to a full top deck on this layout, but just a helix up to see how it works and make sure I understand how to construct it and how the transitions in and out of it should work.  A 24" helix seems a good comprimise at this stage, as this will be nothing more than a 1:1 mockup. I have room for a larger helix though, so I'll take it into consideration before I construct it. It may be of value to make one real size.

I also hadn't planned on handlaying the helix, I figured I'd use Atlas code 100 in there (mostly as I have a few pieces still new laying about). That will be later, I need to develop a plan for the benchwork first and foremost. I'll probably be picking up a few of your videos as I go (thanks for selling individual pieces, that makes the aquisition a lot more convenient).  I'll do another post on that to keep it seperate from the rough track plan.

John Allen did a study

John Allen did a grade and curve radius drag study and found that curve radius drag can be calculated to be approximately 32/r, where the result is the extra grade equivalent drag.

So for a 24" radius curve, the extra drag is 32/24, or 1.33%.

My previous Siskiyou Line 24" radius helix had a grade of 2.65% and the extra drag computed with John Allen's formula showed the grade with the entire train on a curve would act more like a 3.98% grade (2.65 + 1.33). In actual practice, we found this predicted behavior to be right on the mark.

I discuss this is volume 2 of my how-to video series on layout Design and Construction.

I replaced my old bad "helix from hell" with a new 1.75% helix that uses a 40" radius. The extra drag on that helix is 32/40 or + 0.8 %, yielding a new equivalent grade on the helix of 2.55%. My visible track grade out of Coos Bay up the branch is 2.65% so I'm now guaranteed that if a train doesn't stall on the visible layout, it won't stall in the hidden track of my helix.

Having the steepest grade on the railroad being on hidden trackage is just asking for train stalls and related problems - as my operating crew quickly found out. They pleaded with me to replace my 24" radius "helix from hell" and once we figured out the 40" radius alternative would fit, they ripped out the old helix in a couple hours!

They could not WAIT to get rid of that major operational headache!

Induced grade on a curve

The extra 1.5% is emprically determined. No less a person than John Allen years ago obseved this effect and attempted to measure it. John Armstrong confirmed his measurements.

It occurs because:

• rubbing friction of the flanges rubbing on the side of the rail as the wheels are dragged around a curve.
• (and this is much more significant) the inner and outer wheels have different distances to travel on a curve. This is exacerbated on our models by the taper of the wheels. When proceeding upgrade the inner wheels will tend to be pulled up against the rail where the wheel diameter is largest while the outer wheel rides on the skinniest part of the wheel tread (due to the taper). So the wheel with the largest diameter tred patch needs to cover the shorter distance while the wheel with the smaller diameter tred patch needs to cover the longer distance (outer rail). Something has to give - one (or both) of the wheels will spend some time slipping. Which means lots of extra friction.

The tighter the radius of the curve the greater the difference in the length of the two rails (percentage wise) so the bigger this effect. The other thing effecting this is the distance between the rails. An N scale train on a 24" radius will have half (or less) of this problem than an HO train. An HOn3 train will have a lot less induced drag than a standard gauge HO counter part on the same radius curve.

Joe is also speaking from experience having replaced a 30" radius helix on his Siskiyou Line layout with a 40" radius helix because trains struggled to get up the hill with the smaller radius.

A 24" radius helix will appear to be a much tougher grade than the 30".

There;'s another reason to go with a bigger helix radius. And that's the grade required to get railhead to railhead clearance for each loop.

Using the following variables:

Hope this was useful.

Charlie

It was

It was, thanks to both you and Joe. I'm going to file those formulae away for later use.

Charlie, about your second point, (mostly because I cannot leave well enough alone). It would seem that if the two wheels spun seperately, like a limited slip differental, than the drag issue would be negated. I wonder why someone doesn't make this for a replacement wheel set - it couldn't be that hard to engineer. One half could even just ride over the other half (like a sleeve), so it could free wheel. Proper guage could be maintained by the truck itself.

Bernie, are you listening?  If anyone makes this, I want the credit! 

One nit in Charlie's post ...

... that was a 24" radius helix I replaced, not a 30" radius one. With a 24" radius helix the drag effect in HO is simply awful!

Differentials

Probably because it would double or triple the cost of a wheel set and (I suspect) become a point of frequent failure not to mention it wouldn't be very prototypical...

Charlie

Differential wheelsets

The engineering of such a model wheelset would require some sort of inner flange within the narrower axle end against a bearing surface within the thicker axle, to maintain wheel-to-wheel distance.  It would probably cost many dollars in machining, plus adding internal friction (which is what we're trying to reduce) and weight (in a very good place, though), and would probably fail regularly because it needs constant lubrication.  The real problem is that in this case, our models don't work the same way as the prototype, where the wheels MUST be tapered to allow a solid steel wheel on a solid steel axle to cover a greater distance (on the outside of a curve) than the other wheel on the axle, and be able to change almost instantly to the other extreme where the curve is in the other direction.  The screeching we hear as a train goes around a curve is this sliding of the wheels across the rail while carrying many tons of weight.  Talk about your friction!  Of course, the prototypes have 10 or 20 thousand horsepower to overcome this friction, and inertia keeps it moving along.

Scarpia, I'd say you should build your 24" radius helix, to see for yourself.  If (when) it doesn't work out for you, you've learned something valuable, and gained experience.

     Paul

Tag

Just tagging for reference.