Yes, I can hear the groans now, but hear me out. I did look up previous threads, and, while useful and informative, there's still a gap that I'd like to have filled.
By way of background info: As my branch line track plan evolves, I'm considering a version that would include a helix between levels of a multi-deck design (I don't have room for a no-lix, unfortunately, so easy-mode is out). I intend to run Ten-Wheelers, Moguls, Mikes, and/or Consolidations with 40' cars, in consists not longer than 8 cars. I'm pretty much starting from scratch, so most of my rolling stock and motive power will be purchased new.
As I was doing the research, I came across several calculators that gave me the actual grade of my helix. Then, digging deeper, I found the reference formula of 32/R, for calculating the "equivalent grade" around a curve, where the result of this calculation is added to the grade to get what the slope around the curve actually "feels like" to your loco trying to haul cars up it. All well and good, though the results make me cringe, topping off around 3.5%, and everything I see says any grade above 2.4% is a bad idea.
But then, I've seen several track plans on the forum here and elsewhere that use 24 or even 22-inch radius helices (and they're not N), with the resulting grades being in that range, which greatly clouds the issue. However, those are diesel-powered layouts, which I'm given to understand haul better than steam locos in HO (sans Bullfrog-Snot), which could explain how they get away with it. But then if you add to it those theads where the disucssion turns to "Hey, I ran a 4.5% grade with half a dozen cars with no issue" sort of statements, and it all becomes clear as mud.
So I'm curious: recommendations being what they are, can anyone speak from experience with what they've been able to actually pull off with steam locos of the variety I'm looking to run? Could I realistically hope to do what I want, or should I just assume that other plans I see use red matter or other witchcraft to pull off what they do?
I've seen many threads where the bottom line is "test with your equipment", and I plan to, but I'd appreciate some advice on what's even worth testing.
Thanks
- Adam
When all else fails, wing it!
LonehawkI've seen several track plans on the forum here and elsewhere that use 24 or even 22-inch radius helices (and they're not N)
Unbuilt speculative plans include lots of potentially unreliable features. This includes a suprising number of plans in commercial magazines, unfortunately.
Tight radii multi-turn helixes such as that have worked for very short trains of short cars, especially if clearances are reduced by limiting the types of equipment that may be run. But for most builders, best practices call for significantly broader radii and adequate clearance.
Testing with your own equipment at your preferred train length will provide the most solid answer.
Byron
Layout Design GalleryLayout Design Special Interest Group
Think of a Slinky. With it perched, and stable, on one of its open ends, collapsed, those grades look quite reasonable. Who wouldn't want a helix like a Slinky?!?
Trouble is, how to shoehorn some tracks between those layers, and then a SD70 with...oh....only four flat cars with seacans...not even doublestacked. If you grasp the Slinky and begin to lift the coils until that bottom ramp has all the clearance you need, what has happened to the layers atop it? They all have lots of overhead clearance, but they have angled their ramps from 1% to about 6% gradient. I wonder if that SD70 is going to make it towing even one of those flat cars let along four of them. Oh, and there's that overhead clearance problem in a Slinky.
It's the same with a helix. The smaller it is, the steeper the grade you need in order to get all those Slinky layers separated enough, top to bottom, to get your tracks AND some scale rolling stock to pass through them.
Conversely, as you widen the helix, bringing its diameter well beyond the 50" mark, you find that you are getting that reasonable compromise between the grade and separation that will allow your trains to roll up those coils. But at 50", just off the top of my head, your ramp grade will still be north of 3%.
Our hobby is a series of compromises about 'bother'. How much bother are you willing to put up with from operating session to operating session before you conclude that it's just not a whole heckuva lot of fun? Are you willing to run consisted engines up that helix, or will you settle for small trains comprising a locomotive and four cars? Or only three heavyweight passenger cars? Maybe only two? If you make your helix 6' in diamater, with close to 32" radius curves running up it, you might find that a single diesel and seven cars is about the best you can hope for...or maybe ten cars. The only way to determine this all before you commit yourself to cutting and screwing the ramp sections together is to create such a ramp and test your typical consist. If you are going to want curved grades, you need to factor that in as well. Instead of testing the 3% ramp, add another 3/4% to it in order to cover your engine for the resistance that all the trailing cars will create for it on the curves.
Perhaps I didn't see it in rereading it twice, but I did NOT see where the OP posted the anticipated radius he was intending to use, ....nor his total rise??
Brian
My Layout Plan
Interesting new Plan Consideration
railandsailPerhaps I didn't see it in rereading it twice, but I did NOT see where the OP posted the anticipated radius he was intending to use, ....nor his total rise??
Thats what he's asking.
Mike.
My You Tube
For steamers, figure up front that you will need traction tires or Bullfrog Snot (which works very well). Steamers are not known for their pulling power, though they have gotten much better in the last ten years or so.
Here's what I identified as the major considerations in helix design and operation in a clinic I've given at several NMRA National, Regional and Division meets:
Space Available - What’s The Footprint?
Grade Desired - How Steeply Do You Want To Climb?
Minimum Radius - How Sharp A Curve Do You Want To Use?
Potential For “Stringlining” - How Long Are Your Trains And Cars?
Deck Separation - How Far Apart?
Length Of Hidden Run - How Much Track Is Out Of Sight?
Time “In The Hole” (Perceived And Actual) - How LONG Will The Train Be Out Of Sight?
“Slot Car” Tendency With Hidden Trains - How Fast Will The Train Be Moving When It Finally Reappears
Here's a sample of calculated grades for various radii, based on 4" railhead-to-railhead separation between decks:
Radius Grade Compensated Grade* Run Length
18 3.5 5.05 565" (47', 3/4 mile HO)
24 2.7 3.87 754" (63', 1 mile HO)
30 2.1 3.03 942" (79', 1.3 miles HO)
36 1.8 2.58 1131" (94', 1.6 miles HO)
*LDSIG-Modified John Allen Formula – CG = G+28/Radius
So who am I? I'm the guy who built this:
Ten scale miles of track in a four-track helix. I called it "The Behemoth Helix."
Here's a short (and somewhat blurry - sorry) video of action on the helix:
Mark P.
Website: http://www.thecbandqinwyoming.comVideos: https://www.youtube.com/user/mabrunton
Wow, Mark.
To quote Darth Vader... "Impressive. Most impressive."
With the calculator I had used in exploring the feasibility of a helix, I came out with about a 64" unit using a 30" radius of track with 3" spacing from one railhead to the bottom of the deck above, which yields a 1.99% grade on an 18" total rise. Factoring in the original John Allen formula, I get 3.22%. (Is the one you mention closer to a "realistic" measurement?) I could flex a bit on the space between decks, but the helix can't get much bigger. I could go as far perhaps as a 32" radius, but that would be maximum, and I really want to get below a 3% grade.
I'm not AS worried about stringlining as I'll be using shorter equipment due to the era I'm modeling, and as for hidden track, I planned to just enclose the helix altogether, with access up the center. Mine would only need to be single track. I'll be operating at a scale 25 mph max, and I don't really care how long the train is out of sight. I'd consider the helix part of my "selective compression."
And to Byron's point... I'm guessing my assumption that the plans posted on MR are all built and functional may be incorrect?
Two of the smaller radius helices I was looking at were the Loup Creek Branch (22" R) and the Oakhurst RR (18" - Wow). Those would fit my footprint well, but they seem prohibitively small based on the calculations I've run.
Looks like the helix at the Nazareth Society of Model Engineers in Stockertown.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
LonehawkAnd to Byron's point... I'm guessing my assumption that the plans posted on MR are all built and functional may be incorrect?
Many have never been built. And, unfortunately for readers, many cannot be built as drawn.
LonehawkTwo of the smaller radius helices I was looking at were the Loup Creek Branch (22" R) and the Oakhurst RR (18" - Wow).
Loup Creek appears not to ever have been built, it was a contest entry. And the grade seems higher than quoted -- but there's not enough information to know for sure. The few provided elevations suggest a 4.25" railhead to railhead elevation change. The nominal grade at 22" would be 3.075%, with an additional effective grade (32/R) of 1.45% for an effective grade of over 4.5%. Pretty stout. The article text describes the helix grade as 2%, but I don’t think that can be correct for HO scale based on the elevations (at least not with adequate clearance).
Oakhurst is a different story. Note that the builder indicates grades of up to 6% on the layout. It’s not something I would suggest a client try, but the owner is modeling an earlier era with smaller equipment and was willing to make the compromises necessary.
Lonehawk Wow, Mark. To quote Darth Vader... "Impressive. Most impressive." ... Factoring in the original John Allen formula, I get 3.22%. (Is the one you mention closer to a "realistic" measurement?)
...
Factoring in the original John Allen formula, I get 3.22%. (Is the one you mention closer to a "realistic" measurement?)
That beast took almost a year to build, and I was thoroughly sick of track laying when it was finished!
The orginal John Allen formula of Compensated Grade = Grade + 32/R was apparently based on testing John did.
The LDSIG (Layout Design Special Interest Group) did their own series of tests and adjusted the second term of the formula to 28/R from John's 32/R. That's about a 14% reduction.
So using John's formula might be a bit conservative, but that's not a bad thing. I'd rather find out that I could handle a bit more load in the helix than I expected, than to find out I could handle less.
cuyama Many have never been built. And, unfortunately for readers, many cannot be built as drawn.
cuyama Oakhurst is a different story. Note that the builder indicates grades of up to 6% on the layout. It’s not something I would suggest a client try, but the owner is modeling an earlier era with smaller equipment and was willing to make the compromises necessary.
Yeah, went back and saw that. Wish I’d seen that initially before going full hur-durr.
I took another look at my space available, and I’m going to look into an oval helix as well. I understand the straight stretches help ease the haul up-grade.
BruntonThe LDSIG (Layout Design Special Interest Group) did their own series of tests and adjusted the second term of the formula to 28/R from John's 32/R. That's about a 14% reduction.
I've been in the LDSIG since 1995 and edit their magazine, and I don't remember this change, but perhaps I missed it in the early years. The LDSIG is in the process of testing the John Allen rule-of-thumb and for very tight curves it looks as if the 32/R is a little low. At very broad curves it might be a little high, but I don't recall anyone saying that it should be 28/R in general.
LonehawkI wonder how many people have built these sort of layouts only to realize that they don’t work as advertised. No bueno.
Agreed.
LonehawkI took another look at my space available, and I’m going to look into an oval helix as well. I understand the straight stretches help ease the haul up-grade.
In a large enough space, that can help, but it can be tricky in tight spaces because the effective grade changes in and out of the end curves unless one is very careful. Empirically, oval helixes seem slightly more prone to stringlining, but it's hard to compare apples-to-apples.
cuyamaI've been in the LDSIG since 1995 and edit their magazine, and I don't remember this change, but perhaps I missed it in the early years. The LDSIG is in the process of testing the John Allen rule-of-thumb and for very tight curves it looks as if the 32/R is a little low. At very broad curves it might be a little high, but I don't recall anyone saying that it should be 28/R in general. Byron
I wasn't involved at all in the actual testing, but I read about it in the late 1990's or thereabouts. Maybe what I read was inaccurate. I seem to remember it coming from someone of some prominence in the hobby these days, but since I'm not certain I won't be more specific. But the source was pretty well regarded...
The formula worked well enough for me - I could haul 15-20 car trains up my helix without an issue, which was how I planned it. Whether the term is more correct at 28/R or 32/R or even something a little bit different, it still provides a good ballpark estimate of compensated grade your trains will have to deal with.
Brunton Great minds think alike! I did exactly that - 24" straight sections on each side of the helix to reduce the grade. Worked great! It also gave me a spot to add rerailers in each tier, just in case a wheel came off the track in the preceeding 180 degrees.
Great minds think alike! I did exactly that - 24" straight sections on each side of the helix to reduce the grade. Worked great! It also gave me a spot to add rerailers in each tier, just in case a wheel came off the track in the preceeding 180 degrees.
That's a great idea. I stuck that in the notebook in case I do end up with the ovals.
cuyama In a large enough space, that can help, but it can be tricky in tight spaces because the effective grade changes in and out of the end curves unless one is very careful. Empirically, oval helixes seem slightly more prone to stringlining, but it's hard to compare apples-to-apples.
Yeah, good point. And if I look at it by the metric of average grade, the benefits of making the oval are less significant, although, by the numbers, I could be shaving off 1% or more on the straight grades, and on a 3% average grade, that could mean something. But then, the transitions may still bite me. Only one thing to do: Mo' mafs, mo' testin'.
Lonehawk....So I'm curious: recommendations being what they are, can anyone speak from experience with what they've been able to actually pull off with steam locos of the variety I'm looking to run? Could I realistically hope to do what I want...
Well, I don't have a helix, but I do have a fairly long grade from the main level of my layout up to a partial second level.
On the rough drawing shown below, the grade extends from South Cayuga, and starts about halfway between the written name and the Speed River, then rises continually at about 2.8% around the peninsula and ends pretty-much where the green colour meets the grey over the area shown as Elfrida. Minimum radius is 34", and the difference in elevation is 15".
I run Moguls, Ten-Wheelers, Consolidations, and Mikados, all somewhat modified to increase their pulling power. My rolling stock varies in weight from empty hoppers at about 2.25oz. to loaded hoppers at 8oz., gondolas of scrap at perhaps 10oz., and the rest somewhere around the NMRA recommended practice. Cars vary in rolling qualities, from a passenger car like this, which will roll only a couple of inches down the grade, even when push-started....
...to ones like this, which will roll, without a push, downhill from South Cayuga, through Elfrida, Lowbanks, and Port Maitland, almost to the lift-out at the room's entrance, a distance of about 70'...
Train length varies from one or two cars, to as many as 18 or 20, and I have run trains in excess of 70 cars up that long hill. I do perform tests for all of my locomotives, and keep a record of their "tonnage ratings". This allows me to assign as many locomotives to a train as would be required to get it to its destination. This may involve double or triple heading and/or pushers, too. I'm a DC operator, and have found that if a train actually requires more than one locomotive, most locomotives will run well with others regardless of type or manufacturer. There may be some exceptions to that, but mostly with older locomotives using low-quality motors or gear trains. I'd hazard a guess that any of your locomotives should be able to handle 8 relatively free-rolling cars of recommended weight if you can keep the radius to a minimum of 30" and the grade no greater than 2.5%. And if they can't, it's a good excuse to use more locomotives. You could then simply designate the helix as a helper district, which could be an enhancement to operations.
Wayne
BruntonI wasn't involved at all in the actual testing, but I read about it in the late 1990's or thereabouts.
I'll look through the archives; there may well be something I don’t remember. There was a lot of speculation around that time that the (then) new better-rolling trucks would change the rule-of-thumb. Current empirical testing previews suggest it's the friction of the flanges against the rail that's more critical, so free-rolling trucks may not help as much as one would hope.
Lonehawk With the calculator I had used in exploring the feasibility of a helix, I came out with about a 64" unit using a 30" radius of track with 3" spacing from one railhead to the bottom of the deck above, which yields a 1.99% grade on an 18" total rise. Factoring in the original John Allen formula, I get 3.22%. (Is the one you mention closer to a "realistic" measurement?) I could flex a bit on the space between decks, but the helix can't get much bigger. I could go as far perhaps as a 32" radius, but that would be maximum, and I really want to get below a 3% grade. I'm not AS worried about stringlining as I'll be using shorter equipment due to the era I'm modeling, and as for hidden track, I planned to just enclose the helix altogether, with access up the center. Mine would only need to be single track. I'll be operating at a scale 25 mph max, and I don't really care how long the train is out of sight. I'd consider the helix part of my "selective compression.
I'm not AS worried about stringlining as I'll be using shorter equipment due to the era I'm modeling, and as for hidden track, I planned to just enclose the helix altogether, with access up the center. Mine would only need to be single track. I'll be operating at a scale 25 mph max, and I don't really care how long the train is out of sight. I'd consider the helix part of my "selective compression.
I would say that a 30" radius as you describe will work well. The John Allen adjustment is an area around which there is still some debate. My experience is that for a 30" radius it overstates the grade. This is based from experience running trains on my helix (R 30/32.5) over the past 10 years.
At this point there is enough data from helix construction and performance to make some reasonable assumptions about helix construction practices.
Where the wheels come off the bus is when people want to try and cram a helix in an area that it won't fit and thus they usually try to go below reccomended radii. There are many well documented cases of this not working out well.
Of course opinions vary,
Guy
see stuff at: the Willoughby Line Site
trainnut1250 Where the wheels come off the bus is when people want to try and cram a helix in an area that it won't fit and thus they usually try to go below reccomended radii. There are many well documented cases of this not working out well. Of course opinions vary, Guy
That’s what I definitely want to avoid. I like the wheels on my bus to go ’round and ‘round, not fall off. I mean, I’d love to pull off a 24 inch radius, but the more I read and the more numbers I crunch, the worse that idea looks. I’m going to work up the plans using a 30-inch diameter and see how that works in the room.
Brunton cuyama I've been in the LDSIG since 1995 and edit their magazine, and I don't remember this change, but perhaps I missed it in the early years. The LDSIG is in the process of testing the John Allen rule-of-thumb and for very tight curves it looks as if the 32/R is a little low. At very broad curves it might be a little high, but I don't recall anyone saying that it should be 28/R in general. Byron Very interesting! I wasn't involved at all in the actual testing, but I read about it in the late 1990's or thereabouts. Maybe what I read was inaccurate. I seem to remember it coming from someone of some prominence in the hobby these days, but since I'm not certain I won't be more specific. But the source was pretty well regarded... The formula worked well enough for me - I could haul 15-20 car trains up my helix without an issue, which was how I planned it. Whether the term is more correct at 28/R or 32/R or even something a little bit different, it still provides a good ballpark estimate of compensated grade your trains will have to deal with.
cuyama I've been in the LDSIG since 1995 and edit their magazine, and I don't remember this change, but perhaps I missed it in the early years. The LDSIG is in the process of testing the John Allen rule-of-thumb and for very tight curves it looks as if the 32/R is a little low. At very broad curves it might be a little high, but I don't recall anyone saying that it should be 28/R in general. Byron
Very interesting!
So let me ask for clarity. For a 30" R helix with 4" between railheads I should be looking at a 2.1% grade, compensated by apprx +30R,...or 3.1% compensated grade.....correct?
This helix would gain 4" in each complete circle, ....equals 1" per each quarter of a complete circle.
4" between railheads should allow for dble stack containers cars?....and most other tall cars??. Particularly if my roadbed is fairly thin and stiff, .....double thickness tempered hardboard supported at 15 inch spans by 12 upright columns/all-thread/whatever.
railandsail So let me ask for clarity. For a 30" R helix with 4" between railheads I should be looking at a 2.1% grade, compensated by apprx +30R,...or 3.1% compensated grade.... 4" between railheads should allow for dble stack containers cars?....and most other tall cars??. Particularly if my roadbed is fairly thin and stiff, .....double thickness tempered hardboard supported at 15 inch spans by 12 upright columns/all-thread/whatever.
So let me ask for clarity. For a 30" R helix with 4" between railheads I should be looking at a 2.1% grade, compensated by apprx +30R,...or 3.1% compensated grade....
Based on what I can see, yes your compensated grade estimate is about right. As far as the containers... you can answer that yourself. Just measure them.
Rail and Sail is correct; do not guess. Measure. I learned that lesson the hard way on three sepa...well, okay, I only learned it the last time, I hope, but the incidents were portal, tunnel, and bridge related. First was the unusually high stack on a CPR caboose. Second was on the cable-routing frame atop a crane, and the third time was with the extended pantograph on my GG1. The GG1 frighened me because I thought I had broken off the pantograph. Happily, not so. I took a measurement from the surface of my sub-roadbed up to the clearance level of all bridges, portals, and overpasses. I added 1/4" for cheap insurance, and that figure became my minimum overhead clearance throughout my track system on my current build. So far, so good. You HAVE TO know that figure.
selector ....I added 1/4" for cheap insurance, and that figure became my minimum overhead clearance throughout my track system on my current build. So far, so good. You HAVE TO know that figure.
....I added 1/4" for cheap insurance, and that figure became my minimum overhead clearance throughout my track system on my current build. So far, so good. You HAVE TO know that figure.
So what is that figure you currently utilize?
If you are putting your helix outside the shed, is there anything actually forcing a 30" radius? Why not go bigger - 36", 40", even 48"? At 48" radius, each turn could go up 5" and be less than a 1.7% grade, before adding the compensation. Plenty of clearance for tall cars and/or maybe no need to use any special techniques to keep the support structure as thin as possible.
railandsailSo what is that figure you currently utilize?
Brian, re-read what you quoted from Selector's post. It IS his current figure, on his current layout.
I have been following this thread, I don't know if you've posted this before, but what part of the country to you live in? An area with heavy winters? North of the good old Mason-Dixon line?
Your idea of the helix being outside of your building has me interested.
rrinker If you are putting your helix outside the shed, is there anything actually forcing a 30" radius? Why not go bigger -
If you are putting your helix outside the shed, is there anything actually forcing a 30" radius? Why not go bigger -
mbinsewi railandsail So what is that figure you currently utilize? Brian, re-read what you quoted from Selector's post. It IS his current figure, on his current layout.
railandsail
mbinsewi I have been following this thread, I don't know if you've posted this before, but what part of the country to you live in? An area with heavy winters? North of the good old Mason-Dixon line? Your idea of the helix being outside of your building has me interested. Mike.
I posted this early on before I decided on a dbl track helix
Helix I have debated this question to some great extent, and done quite a bit of reading about it. I am not excited about the gradual rise of an around the wall grade. It would take up that much additional width that I do not care to give up in my relatively narrow shed. In fact I even considered not insulating the shed and thus being able to build some portions of the layout into the 2x4 studded areas, but the better part of valor said I will need AC at various times here in warm humid FL. I am not married to the idea that my trains need to make constant use of this exchange between levels. With that in mind I am only considering a single-track helix, unless someone can convinced me otherwise. Trains will have to wait their turn to use the single trackage...not unusual? And someone commented about 'how about when it is raining' on my 'external helix'. First off when its raining perhaps I will have NO traffic on the helix,...just run my multiple trains on the lower level, and a single freight/passenger train on the upper level, and perhaps concurrently my logging train(s) on that mountainous area in the upper level (isolated from the mainline) there. I'm imagining the donut shaped helix housed inside a short flat box like structure built of square tube aluminum tubes that can be bolted up to rear external face of my metal shed (to the studs of the shed), and with two 'legs' at its outer edges. I will be able to access the inner hollow of the helix from up underneath. Naturally the box structure that houses the helix will have metal sheeting covering it just like the shed itself. I have all of this alum metal already, and chose to utilize light weight alum rather than heavier and rot prone wood-frame construction. I am imagining that the lower entrance to the helix will be from a track that is already rising in grade from the blob/head portion of peninsula (in order to pass over other tracks at the root of the peninsula)....so one less level required of the helix itself. It will then rise up to the upper level and enter back into the layout room in a straight shot down the long edge of the shed/layout.
railandsailSo let me ask for clarity. For a 30" R helix with 4" between railheads
That seems a little tight to me for modern-era equipment, but I am conservative in this regard. Here's how I would calculate:
Modern-era clearance from NMRA S-7: 3 5/32”Helix roadbed (2X 1/8” Masonite): 8/32”Track + adhesive: 6/32”Metal bar or strap below roadbed: ?Subtotal: 3 19/32"Amount left for construction variance, finger access, etc.: 13/32” at best
The wild card here is an additional loss of clearance if your helix roadbed support is a strap or bar below the subroadbed.
Given this, I’d suggest at least 4¼” railhead-to-railhead elevations. But again, I am conservative, since I design for others to build and, for most modelers, some construction variance is inevitable.
railandsail selector ....I added 1/4" for cheap insurance, and that figure became my minimum overhead clearance throughout my track system on my current build. So far, so good. You HAVE TO know that figure. So what is that figure you currently utilize?
Sorry, a little late to rejoin the discussion here on the Left Coast. I ended up deciding that, in the absence of ever trying to run doublestacks in a futuristic way (I'm in the transition era), my maximum clearance was determined by the extended pantograph, that that came to almost exactly 4", a hair under. So, I opted for an absolute minimum of 4.25" (4 1/4").
I ended up doing the same thing on my last layout, the one with a nearly-six-feet diameter helix. I decided, later, that while I didn't regret my decision, or the fun I had building it, the guys who described them as space hogs and almost always a LOT of hidden running were right. So, no more helices for me.
Last comment...I knew I'd have to reach between the layers a few times during the layout's life to retrieve items or to attempt to rerail them, or if a loco died or dropped electrical contact, to reach in and nudge them along. In that respect 4" was about right, but also not terribly generous. Not for fat fingers, anyway.