Model Railroading > More helix questions: guard rails and helper locos

Dave's helix thread below reminds me of some pressing helix concerns of mine:

What, if anything do you helix owners do about ensuring trains don't fall into the precipice when cars string-line or pop out, or is this even a problem?
I'm thinking about low guard rails attached to the inside and outside edges of the helix, but...

I ask this because my helix is going to be the site of a helper operation, and there is a lot of potential for knocking the cars off the track with this arrangement. Not only do I worry about my zamac-boilered 2-10-10-2 kit-bash exploding after a dive to the floor, I'm also concerned that operators would have to wear hard hats to protect their personal safety, since the helix will extend through the ceiling into a second story. That 2-10-10-2 could really hurt!

The helix will be HO scale 48" radius, and due to the length of the run, I would like to place a siding in it - which creates even more potential for derailments. I do understand that meticulous track-work is key.

I'm going to mount my helix using threaded rods so that I can conduct experiments with varying grades to see what happens with particular locomotives and consists. Trial and error, just like the prototype. I'd like to make grades as steep as possible to reduce the length of the run through the helix, but conventional wisdom seems to be that many HO locomotives don't like steep downgrades. Any thoughts here?

So, am I paranoid? Or I need guard rails alongside the track? Or should I make the roadbed extra wide to contain wayward equipment? Or add weight to the cars to make them hold the rail better?

I used to run prototype helpers, and although a lot of the real aspects of helper operations should be applicable to models, as you I realize that there probably are as many differences as similarities. I feel like a rookie fireman needing guidance from seasoned engineer.

EO



Edited 1 time(s). Last edit at 05/26/16 15:51 by hogheaded.

My helix is 48 inch diameter.  I have trouble with trains straight lining when the lenght is more then 17 54 foot covered hoppers plus caboose.  I have never tryed mid-train helpers.  I would think adding weight would increase the chance of straight lining.  I have not had trouble with my engines derailing, but the head cars tend to straight line if I have too many cars.  That is how I found that 17 cars was the limit. 



Edited 1 time(s). Last edit at 05/26/16 11:22 by Conductor_Pappy.

Conductor_Pappy Wrote:
-------------------------------------------------------
> My helix is 48 inch diameter.  I have trouble
> with trains straight lining when the lenght is
> more then 17 54 foot covered hoppers plus
> caboose.  I have never tryed mid-train helpers. 
> I would think adding weight would increase the
> chance of straight lining.  I have not had
> trouble with my engines derailing, but the head
> cars tend to straight line if I have too many
> cars.  That is how I found that 17 cars was the
> limit. 

That's good to know. I'm surprised that so few cars cause the string-lining on such a wide radius. Interesting! What kind of grade does your helix have and are your cars weighted to full NMRA specs? Optimum car weight certainly is a major issue. You are certainly correct about car weight being a function of string-lining. In heavy grade territory, railroads rule specify placing heavy loads to be placed near the head end, with empties on the rear. Violation of this principle is one of the reasons for the SP train string-lining the curve at Cantara Loop (my old territory) and killing all of the fish in the Sacramento River a quarter century ago.

String-lining is one of the primary reasons for using helper engines. Uphill trains naturally want to straighten out curves. Downhill trains naturally want to pop-out on curves (Here's another question: How long does a downhill train have to be before cars start to pop out?). This makes modeling helper service a fun prospect, but a very challenging one. A fundamental part of the deal is that if either engine momentarily hesitates due to electrical continuity problems, the train is likely to head for the ditch. Likewise, if the two engineers don't act in fair unison. Fun fact: Any self-respecting helper engineer never radioed the road engineer about what he wanted the helper to do. A holdover from radio-less steam days, I guess.

For once, I'm going to take a scientific approach to designing the "helper helix", including using a home-built dynomometer car. I've found a small digital scale to measure drawbar pull, but am still looking for a scale to measure "buff", i.e. compression. I wish that I could find Bluetooth-equipped scales.

Though I'm a transition era modeler, I anticipate that some of my data will have a broad application, what with distributed power being use on so many modern trains. Though my era came before railroads began to comprehend the benefits of using dynamic brake equipped helpers on downhill trains, I'm nevertheless going to due some testing of this, as well.

Hopefully I'll be ready to conduct tests by the end of the summer, plenty of time to lay in a sufficient beer supply.

EO

hogheaded Wrote:
-------------------------------------------------------
> Dave's helix thread below reminds me of some
> pressing helix concerns of mine:
>
> What, if anything do you helix owners do about
> ensuring trains don't fall into the precipice when
> cars string-line or pop out, or is this even a
> problem?
> I'm thinking about low guard rails attached to the
> inside and outside edges of the helix, but...
>
> I ask this because my helix is going to be the
> site of a helper operation, and there is a lot of
> potential for knocking the cars off the track with
> this arrangement. Not only do I worry about my
> zamac-boilered 2-10-10-2 kit-bash exploding after
> a dive to the floor, I'm also concerned that
> operators would have to wear hard hats to protect
> their personal safety, since the helix will extend
> through the ceiling into a second story. That
> 2-10-10-2 could really hurt!

Running helpers thru the helix (presuming with 2 crews can be a challenge.   Best way to address this is to provide lots of visibility to the crew so they can keep an eagle eye on how their train is progressing.  Keep the helix exposed if possible, especially on the lower level when helper crews will need to accurately adjust their speed to account for the grade.  Going up to a 2nd story, as in to an overhead room?  Wow!  That's a lot of elevation gain, and probably the highest helix I've heard of.

> The helix will be HO scale 48" radius, and due to
> the length of the run, I would like to place a
> siding in it - which creates even more potential
> for derailments. I do understand that meticulous
> track-work is key.

Using a 48" radius is great.  That means you can gain the necessary overhead clearance between helix levels without being forced into stiffer grades. 

I highly recommend against a siding in the middle of the helix for a couple of reasons:

#1 is that mid-helix switches introduce the potential for derailments.   Derailments in a helix are a disaster.   Besides the awkward clean-up, there is also the potential for cars to uncouple, with the rear portion of the train backing down the grade at and ever-increasing rate of speed until centrigual forces sends the entire string of cars flying to the floor.  Or at least derailing wiithin the helix, but with enough force to physically damage the rolling stock.

#2 is a steeper helix increases the drawbar stresses on the trains going uphill.

I suspect you may be concerned about the length of time a train requires to transit this large helix so an opposing movement can use the helix in the opposite direction.   It's good you are aware of this.   Some quick calculations (2 x helix radius X 3.14 X number of helix levels) will indicate how many linear feet of track a helix train will traverse.  A helix like the one you are contemplating will take many minutes for a train to negotiate.   And you certainly don't want crews trying to address this by speeding thru the helix potentially causing a derailment. 

In place of of a mid-helix passing siding, I'd recommend slightly expanding the helix and double-track it.  The benchwork is exactly the same construction challenge as building a single track helix.  Adding second outer track at 50 1/2" radius (I'm assuing HO here) only would add another 5" to the total helix diameter.  Then by running all the uphill trains on one track and the downhill trains on the other, the two tracks would accommodate multpile trains in the helix on each track, greatly increasing helix traffic capacity and reducing running time to the minimum.  Routing the uphill trains to the outside track will provide them with a slightly easier grade due to the approximately 16" longer run to gain the same elevation.  Adding the second helix track does mean more track to purchase and install, but the benefits are well worth it.  Your comment further below indicates you have the space to make the helix a bit wider if needed for guardrails so hopefully the modest size increase to double-track the helix is do-able.

Our club layout has a 7-layer double track helix right next to the staging yard and it has proved to be a handy layout feature.  It was oroginally designed as a single-track helix, but second thoughts had us revise d the design to double track it.   This turned out to be one of the wisest design modificaitons we've made.  Staged westward trains are run up to the top of the helix by the staging yard crew for later pick up by the road crews.   We usually stage 4 or more trains sequentially in the helix, thus clearing track space in the staging yard for handling other trains.

> I'm going to mount my helix using threaded rods so
> that I can conduct experiments with varying grades
> to see what happens with particular locomotives
> and consists. Trial and error, just like the
> prototype. I'd like to make grades as steep as
> possible to reduce the length of the run through
> the helix, but conventional wisdom seems to be
> that many HO locomotives don't like steep
> downgrades. Any thoughts here?

Using threaded rods to support the helix layers provides for precise adjustments for each set of supports so you can dial in exacly the grade you've planned. 

Here is a helix construction trick to consider:  Take as much time as you need to be SUPER picky about installing the roadbed on the lowest level of the helix.  Ensure the grade is constant and the roadbed is perfectly level from side to side.   Once you've got the first helix level set, then you can gear every subsqeuent helix layer off the lowest one.  A time-saving trick is to calculate the distance between the top of the lower helix layer and the underside of the roadbed on the layer above.  Then cut a bunch of consistent, carefully measured spacers out of PVC pipe to slip over the threaded rods.  This will ensure the first layer grade you carefully measured and built will be consistently reflected on every helix layer.  Repeat for every layer up to the top of the helix.

(You likely know this, but you will definitely want to lay the track, wire it and test it on each level before installing the next level benchwork.  That way you won't have to try to work in between levels where you can't see and there's not enough room to use tools properly.)

Steep grades in a helix are asking for trouble.  I'd avoid it.  With a 48" radius you only need a 2% grade to get a 6" elevation gain per layer.  Depending on the thickness of your sub-roadbed + roadbed sandwich, this will provide at least 4 1/2" of clearance for trains.  You could use a lesser grade by reducing the trackwork clearance.  You can even "build in" the computed grade by simply setting the roadbed spacers to be, say, 4" long and leave it at that.   This is easier to figure and measure vs. calculating a spacer length of 4.38865" just so the grade works out to some nice round number.

I'd advise against deliberately making the helix grade any steeper than about 2.5%.  2.5% is already in the "caution" zone in my opinion.  If you double track the helix, it removes the incentive to address the opposing traffic congestion issue by making the helix run as short as possible  

> So, am I paranoid? Or I need guard rails alongside
> the track? Or should I make the roadbed extra wide
> to contain wayward equipment? Or add weight to the
> cars to make them hold the rail better?
>

Helix guard rails are a cheap insurance policy vs. the potential derailment damage.  You could fashion some benchwork edge guards to keep derailed rolling stock from leaving the helix benchwork.  Our club uses thin clear acrylic that projects aobut 3/4" above the benchwork. This keeps any wayward cars on the helix and still provides room for your hands to reach in and fix things.

Normal, properly-weighted cars with good rolling wheels should be all you need.  Also be a stickler for accurate coupler height, which is more important for layouts with helixes due to the risks described above with break-in-twos while in the helix.  You'll want to do all of this rolling stock standards enforcement anyway since this will translate into better rolling stock performance everywhere on the railroad, not just in the helix.

Heavier than normal car weighting will actually make things worse.  Doing so will reduce locomotive pulling capacity on every grade and degrade car rolling performance over time.   Adding more locos to compensate means a heavier drawbar pull which in turn increases the risk of helix stringlining with longer trains. 

> I used to run prototype helpers, and although a
> lot of the real aspects of helper operations
> should be applicable to models, as you I realize
> that there probably are as many differences as
> similarities. I feel like a rookie fireman needing
> guidance from seasoned engineer.
>
> EO

Anyway, that's my 2 cents.   Well, OK, a nickel.

Good luck

- Norm

NCA1022 Wrote:
-------------------------------------------------------
A whole lot of useful stuff!


Norm, your sage advice is exactly what I've been looking for!

Re double tracking and helix transit time, I originally contemplated double tracking, but then elected to go with single track for operational variety. The trackage leading into the bottom end of the helper terminal is going to be double track, so I was thinking that single track ABS or CTC territory would keep operations from being "too" easy on the trackage above the yard.  I'll take your advice to heart about no switches in a helix to heart. I certainly personally understand the potential ramifications of "rough" track under the extra stress of uphill trains. One option in this regard might be to construct switches in the old continuous frog  True Scale method, which eliminates frog gaps. This might be worth a try for test purposes.

But ultimately you are right about double tracking, since the helix begins immediately at a junction wye at the top end of helper terminal, potentially the busiest spot on the railroad. Double tracking really makes sense here, and would not necessarily require two tracks on the entire helix, since the helix actually will be spit in two with a 50 foot long shelf in between.

The unavoidable shelf portion rise of 18 inches in 50 feet now seems to be the issue under your 2% rule, since this calls for approximately a 3% grade. Luckily, it is roughly 50% tangent track, so there may be hope, assuming that the locos don't "cog" going downhill. More testing.

Now, using your formulas for the upper helix, I could get the needed 22" rise in four loops, at slightly less than 2% grade. That works out to be 100 feet of track. Hmm… That's roughly 1⅔ miles, which for a coal train slogging uphill at 15 mph means roughly a seven minute minimum wait for an opposing train IF things timed out perfectly, which they rarely do. This is quite a long time when converted to a fast clock, eh? In real life, it seems like T&E crews spend most of their careers waiting for something to happen, but this isn't something that one wants to replicate on a model railroad, unless one could situate a mini fridge full of beer near the waiting point.

I guess that I should sit down and  make some rough computations about anticipated schedules and such to see if a second track on the upper helix is necessary, and regardless make provisions for a future second track when it later becomes obvious that I should have installed one in the first place.

I'll take your advice about installing guard rails, and thinking about this further, I've come up with an easy solution, construction and access-wise. The helix is going to be built out of straight lengths of plywood, beveled at either end to form the correct radius. So, if I go with the clear acrylic guard rails that you suggested, I could cut these to match the length of the helix edges and slide them into some sort of clips to hold them and make them easy to remove for "big hook" purposes.

Your words about rolling stock characteristics are well-received. My test record-keeping is going to keep track of every car in the test, beyond the locos, and their position in the train. That way, I'll have a better idea of what causes a derailment. I'm mindful of an incident about 40 years ago, when one particular Holly Sugar covered hopper derailed twice at virtually the same spot within a span of a few days on SP's Cuesta grade out of San Luis Obispo. The second time was in a tunnel! SP really should have checked to see how tight the car's bolster screws were after the first derailment (-:

I kinda liked the idea of using "overweight" cars for various reasons, since all the trains will be made-up with way more power than necessary. I hadn't thought about the added strain on axle bearings causing premature wear as you imply. I guess that this is an area for long term testing to determine potential additional wear.

Thanks for your nickel's worth, Norm, but I would say that its more like two-bits-worth, at least!

Regards,
EO

 

hogheaded Wrote:

> That's good to know. I'm surprised that so few
> cars cause the string-lining on such a wide
> radius. Interesting!

I believe you may have misread the post in question. He said it was a 48" DIAMETER helix. That would translate to a 24" radius curve ... which is steep and tight for a helix. That would likely explain the string lining with only 17 cars.

Posted from Android

bioyans Wrote:
-------------------------------------------------------
> hogheaded Wrote:
>
> > That's good to know. I'm surprised that so few
> > cars cause the string-lining on such a wide
> > radius. Interesting!
>
> I believe you may have misread the post in
> question. He said it was a 48" DIAMETER helix.
> That would translate to a 24" radius curve ...
> which is steep and tight for a helix. That would
> likely explain the string lining with only 17
> cars.

Ahem.
Thanks for pointing out my error. I suspect that you are enjoying this, knowing how we loco engineers hate to be told that we are wrong. Of course, this IS the first time that I've personally erred, so...

EO

 

As I will be building two (!) double-track 30"-radius helices next year, I've really appreciated the advice and comments in both of these threads. Thanks!
-Alex

Here's another pressing question: What do I do with the nearly eight feet of circular dead space within a helix (w/ a 36" tall doorway)? Build a safe room to hide from the in-laws?

EO

The drawback to a helix is the huge amount of space one can take up in a train room.  I've heard of some people being creative with the space inside the helix by turning it into a dispatching desk area, storage, or other use. 

milwrdfan Wrote:
-------------------------------------------------------
> The drawback to a helix is the huge amount of
> space one can take up in a train room. 

Yes, such as the 48 inch helix mentioned earlier.  That over 8 feet from one side to the other; your going to have to have a very sizable train room to fit something that big in to change levels.  Many of us simply don't have that kind of space avaiable.  I had considered a smaller helix to change elevation but in a 10x18' train room, it would have not been workable so I settle for 2.9% grades to go from staging to an upper level.

Cheers, Jim Fitch

I am sorry that I mislead you.  I was trying to answer your straight lining question with a few words.  My helix is 24 inch radius.  It has two loops and rises nine inches.  I can pull 17 54 foot covered hoppers up the helix and bring 24 down the helix.  I can bring up 48 loaded 26 foot ore Jennys from Tow Harbors to the main line.  I can bring up 30 40 foot side dump gons.  The point I am trying to make is the longer the wheel base and the heavier the train and the more curves or loops you have to travel; the more problems you are going to have.  My trains are  only on one and a half loops before they start on straight track.  You are going to have big problems if you are halling 86 foot auto racks.  My main lin is 31 feet by 23 feet with 48 inch radius curves, and is flat.  I have had a 120 car mix freight on the main.  The 86 foot auto rack derails every time unless it is on the rea end.  The empy 60 foot flats and bulk head flats striaght line if they are in trained from the head end to the middle.  My point is the wheel base lenght and weight is what causes straight line derailments not the grade.



Edited 1 time(s). Last edit at 05/27/16 11:35 by Conductor_Pappy.

Conductor_Pappy Wrote:
-------------------------------------------------------
> I am sorry that I mislead you.  I was trying to
> answer your straight lining question with a few
> words.  My helix is 24 inch radius.  It has two
> loops and rises nine inches.  I can pull 17 54
> foot covered hoppers up the helix and bring 24
> down the helix.  I can bring up 48 loaded 26 foot
> ore Jennys from Tow Harbors to the main line.  I
> can bring up 30 40 foot side dump gons.  The
> point I am trying to make is the longer the wheel
> base and the heavier the train and the more curves
> or loops you have to travel; the more problems you
> are going to have.  My trains are  only on one
> and a half loops before they start on straight
> track.  You are going to have big problems if you
> are halling 86 foot auto racks.  My main lin is
> 31 feet by 23 feet with 48 inch radius curves, and
> is flat.  I have had a 120 car mix freight on the
> main.  The 86 foot auto rack derails every time
> unless it is on the rea end.  The empy 60 foot
> flats and bulk head flats striaght line if they
> are in trained from the head end to the middle. 
> My point is the wheel base lenght and weight is
> what causes straight line derailments not the
> grade.



Granted that your grades may be too light to be an issue, grades are actually an important factor in stringlining, because increasing the grade increases the train's resistance to being pulled, and resistance is the name of the game. If you doubled the grade on your helix you would see this play out.

Train dynamics is an incredibly complex subject, but how this basically sorts out is that there are a lot of commonalities between how prototype and model equipment behave. Likewise, a lot of what we know about prototype and model behavior is based upon experience, not scientific analysis.

Whether a train stringlines, or not, is a function of a host of things that create increased resistance to movement, including  (but not limited to) grade, curvature/lateral forces, train weight & composition and amount of "pull" from the engines. Higher grades and curvature equates to more resistance, as does increasing train weight and/or the number of cars (more cars = more wheels=more friction). The amount of resistance is particularly critical at mid train in curves, because this is where consequential lateral forces - and associated tendency for flanges to want to climb over the rail - are highest.

Train composition is key to mitigating the tendency to stringline, as you have discovered on your own. Short cars stay on the track better than long cars, and like the prototype, empty pig flats are a pain. Their long length and long drawbars generate a lot of lateral force (particularly downhill) at the couplers, and real railroads have rules against entraining them next to short cars, because of "angular incompatibility", I guess that you would call it. You've done the same thing in using your experiences to place them at the rear of the train, and presumably not next to one of your 26 foot ore jennies, which would get a conductor fired on the prototype.

So, as with the prototype, the rules of thumb for models in heavy grade/curvature are about the same. Place empty pig flats on the rear of the train, making sure that they are separated from short (less than 40 feet on the prototype, but probably longer on models). Loaded pig flats should not be placed in the middle of a long train without helpers, or behind the helpers. Heavy loads should be placed on the head end, if the train is otherwise mostly empties, to eliminate the lead-weight-on-the-end-of-a-string effect. Fundamentally, it is pretty intuitive that a train can only have so many cars, so much weight before the cars will stringline under certain conditions.

How much pull being generated on the head end is an important part of all of this. Too much pull + too much train resistance = trouble, especially in the high tractive effort diesel era. This is where helpers come in. From what you say, a 20 car train of your 54 foot covered hoppers will stringline going up your helix. If you were to place a helper mid-train or on the rear, the problem would likely disappear, because you have distributed the train's resistance between the two sets of power. The lateral stress in the middle of the train is reduced.

Real railroads generally allow less horsepower to be employed on the head end in heavy grade/curvature territory than on the flat for this general reason. Trains might come into my old helper terminal with six big units in power on the head end, but at the time (30 years ago), only four were allowed to be on-line (in power) upon leaving due to the increases in grade and curvature. Either the two excess units would be taken off-line ("isolated"), or removed off of the head end and sent back in the train somewhere to serve as helpers.

For the fun of it, you might try running a helper (or DPU, depending upon your era) just for the fun of it to see what happens. This is possible even without DCC if you have two units that are well-matched in speed. Actually, you want just a slight variation between the two. On uphill trains, the slightly faster loco should be placed at the rear to mitigate stringlining by bunching up the slack a little towards the rear. On downhill, just the opposite. The slower unit on the rear will tend to stretch the train a bit, reducing the tendency to pop out on the outside of curves.

Give it a try, if you can and let me know about your results. I'm interested.

EO




 

One of our club geniuses solved a lot of potential helix problems by building a track soldering jig to guarantee smooth joints between flex track sections. The "plywood pie wedge" has 1/4" ply pieces overlaid on a half-inch base to create a curved channel for each of the three tracks that make up the helix. The jig covered about 40 degrees of the helix's circle.

Starting at the bottom of the helix, Atlas 83 flex track was laid into the channel, trimmed, and soldered to the next piece. The jig is then rotated around to take care of the next set of joints, and in this fashion, spiraled its way all the distance to the top. Once free of the jig, the track could be fastened to the roadbed. Because all of the track is curved, expansion is not a significant factor.

Train handling is pretty easy because each radius is constant from top to bottom. The grade was set to allow a standard maximum-length train to run without helpers. We are running two cars per powered axle and having no slippage problems. Cars are weighted to NMRA RP plus one ounce, and the only derailments so far are related to equipment failures or operator error.

The helix levels are supported by threaded rods and steel crossplates. There is a slight sag in the grade in one quadrant because of the need for clearance beneath a soil pipe. If I recall correctly, the inner set of rods wasn't installed until after tracklaying was completed, but I wasn't present for all of the work.

grahamline Wrote:
-------------------------------------------------------
Some intriguing stuff!

That's the great thing about clubs - there's always some "genius" around to make things easier on nincompoops like me. A channel within which to lay tracks, as well as a soldering jig, is certainly a mark of advanced thinking, and I shall be happy to steal these inventions. With all of the intelligence that I have received in this thread, I no longer assess that I need a helix genius, but instead a DCC/computer programming savant. Oh, there's things that I want to try!

At any rate, how come your club doesn't try some monster trains with a swing helper cut into the middle and a kicker behind the caboose? Or, at least, a helper behind the caboose or a DPU, for starters? Your club must have some operating nuts, er... geniuses, who are up to a challenge.

EO

With that much elevation maybe a 1 or 2 loop helix on a elevator. Pull in stop , helix it's self it's raised to the next floor train pulls out. By using a helix as the elevator you could have the length for trains to it on the elevator. A simple motorized pin that could lift up and down could act as a hand brake on the lower end so the train don't roll out. Pull in to clear, pin raises up, slack the train back till the rear cae touches then lift away

Posted from Android

NSDTK Wrote:
-------------------------------------------------------
> With that much elevation maybe a 1 or 2 loop helix
> on a elevator. Pull in stop , helix it's self it's
> raised to the next floor train pulls out. By using
> a helix as the elevator you could have the length
> for trains to it on the elevator. A simple
> motorized pin that could lift up and down could
> act as a hand brake on the lower end so the train
> don't roll out. Pull in to clear, pin raises up,
> slack the train back till the rear cae touches
> then lift away

That's quite a concept! Are you a mad scientist by trade?
Actually, this general idea might work for a reversing loop (also 48" radius) that needs to be stowable, so-to-speak. Hmm... it just might work.
Thanks!

EO
> Posted from Android

The Blissfield Model Railroad Club (Blissfield, MI) operates an "exposed" helix, which will ultimately be a mountain scene.  The grade is a constant 2%, except for flats where wyes connect with 4 different levels of the layout.  The elevation change from bottom to top is about 4 feet.  So far, we have been able to handle 40 cars up the grade with 2 average diesel locomotive units and no helpers.  However, there are several tricks to operating this beast.  One, cars should have free-rolling metal wheelsets (such as Intermountain).  Two, couplers should be metal, not plastic and all heights should match.  Three, cars should be weighted according to NMRA standards (total weight 1 ounce plus 1/2 ounce per inch of car length).  The only time we've had straight-lining flopovers was when someone had cars that were too light in the front and some lead sleds toward the rear.  Also, any time you have a string of free-rolling cars........if a coupler fails for any reason, those cars will exceed flatland passenger train speeds by the time they reach the bottom.  Also, descending trains (in compression) can run into major problems with jamming when the engineer dumps the air for some reason and those lead sleds on the rear turn into battering rams.  Careful operation is required, even after all cars meet the standards.  We have experimented with helpers and this also requires careful operation.  I think it's called "slack management" by the proto boys.  The transition from tension to compression has to be kept somewhere in the middle of the train.  If it gets near the power at either end, you're close to a major fiasco.  I witnessed (on a previous club layout 20 years ago) a beautiful Santa Fe auto rack train ascending a 6-ft helix (3.5-ft radius) with two units on the point and two more helping on the rear.  Due to some unknown electrical problem, the front units hit a dead spot on the track and, of course, the helpers just kept right on helping.  The engineer was on the other side of the layout shooting the bull with another member when he heard his auto racks crashing down in the middle of the helix.  By the time he got it stopped, there were at least a dozen cars on the concrete floor in various wrecked configurations.  It's best to think of these things before they happen.  Low guard rails are a good idea to try to save derailed cars or locos.  But they also make it more difficult to retrieve said derailed cars.  If you are fortunate enough to have a young son (or daughter) who can easily clamber under and up the middle of a helix retrieving derailed cars, that is a distinct advantage.  However, if you're anywere near my age, you can probably no longer crawl under and into a helix and ever hope to get back out.  And your kids are full sized and gone by now.........hopefully.    So......................using a helix is sometimes a necessary evil.  But it's a good idea to consider all the possible scenarios that can occur when operating in one.  Good luck.