Ideas for a DIY Skeleton Rail Sled

The photos below show how to build an inexpensive video dolly that moves on rails. Perhaps these photos will spur ideas on how to build a practice skeleton sled.

This dolly is 36" square and is smooth gliding, even when supporting a person. The wheel truck assemblies are inexpensive and easy to construct. Two pieces of angled metal are attached to a sheet of plywood, and roller blade wheels are attached at right angles on each corner, as shown on the photos. More wheels can easily be added if needed to support more weight.



The rails can be made from 3/4" Schedule 40 PVC. The PVC can be coupled using pieces of wood doweling on the side (the usual PVC couplings can not be used because the PVC rail must be smooth on the outside for the entire length).



Comments and additional ideas welcomed.



How smooth does it slide on the rails? Can the sled go pretty fast on it?

What needs to be done now is just putting some handles on it, and putting some cushion on the board so it can be comfortable to land your torso on (probably just throw some towels or a pillow on it). Other than that, it would still be a challenge to connect a downhill surface of PVC piping from a horizontal flat surface of PVC pipe. I’m sure its possible somehow.

Even if I was only able to practice on it horizontally, I could still see myself being able to train for speed using this sled and railing. At the push champs, we were pushing the sled about 15m before going another 5-10m downhill and hopping on it. Another idea you could implement as a training effect would make it possible to add resistance to the sled maybe? A quick solution to that for me would be just putting a 40lb weightvest on there. Or maybe just put some thinner standard weight plates on the dolly and then throw the pillow, towels, or other cushion surface on top of it.

Why not just work on running 6.6-7 60m first then spend a month practicing on pushing the sled? Seems a lot more useful than suboptimal speed training with a sled when you need large advances in speed to make the practice worth while.

Right. That is Kyle’s objective. A month of practice on the sled before the combine would be ideal. Now the question is, how does he do that?


You need to get fast enough for it to matter first. There are plenty of ways you could get a little bit of practice here and there throughout the year (weekend trips to a track) if you are super serious about it and have some cash to spend (book flights in advance and stay at cheap hotels). The bigger question that needs to be answered is how you’re going to drop 3-5tenths off of a 60m time to make it worthwhile to do any of it. I say that in great seriousness because I think that’ll be a much bigger challenge and more worthy of focus.

One key factor is the rigidity of the rails. I tested with PVC Shed 20 and it is so-so, and that’s why I suggested Schedule 40. It’s still relatively cheap and readily available. You might test a length of Schedule 40 and Schedule 80 and see if there is a noticeable difference.

Another key factor is the number of trucks. I purchased a large pack of new rollerblade wheels on eBay for cheap. Just drill a few more holes and add a few more wheels to distribute the weight and reduce deformation of the PVC. The rails will not have any turns, so you can evenly distribute the trucks down the length; no need to gang the trucks at the four corners.

What needs to be done now is just putting some handles on it, and putting some cushion on the board so it can be comfortable to land your torso on (probably just throw some towels or a pillow on it).
Yes, just get a piece of foam and staple some fabric around the edges.

Other than that, it would still be a challenge to connect a downhill surface of PVC piping from a horizontal flat surface of PVC pipe. I’m sure its possible somehow.
That’s the hard part, and why I would also consider a rail-less design. Perhaps just build a really fast sled with rollerblade wheels fixed in a straight line that is the right height and weight and practice in a parking lot. Would be even better if you could find a lot with a suitable slope. Of course, you would also need a simple braking mechanism.


Ridin’ on the City of New Orleans
Illinois Central, Monday mornin’ rail
15 cars & 15 restless riders
Three conductors, 25 sacks of mail

Christopher, I think that I could probably do without hopping on the sled actually. It would be hard to slow down by going downhill with the sled, but I think I could probably just stand up and try to decelerate as much as possible. Sure it wouldn’t be as legit, but if this was the best access to sled pushing that I had, then I’d be fine with it.

Sounds like you don’t know much about the sport of skeleton… Ya having a blazing 60m would help out your combine scores, but those numbers mean nothing in the sport of skeleton (Bobsled is a different story). Hell, the 60m would be a poor indicator of determining a skeleton athlete, and thats why they don’t use it as one of their tests! I mean there’s probably maybe 2 guys on the team (national and developmental) who could run 6.6-6.7 FAT. Being strong and fast over shorter distances would be a lot more important because you’re in a crouched position the entire time (also a lot more quad dominant).

I just got back from the push champs last week. There were several guys who I could crush anyday of the week at the track yet they were able to have a push time that was .2-.3 times faster than mine. If I could have more time pushing the sled, there’s no doubt that I could become a top 10-15 pusher next year. The hardest thing to replicate is being able to run downhill with the sled. Its like doing overspeed training in a crouched position without being able to see what lies in front of you.

I some what agree with your comment - similar to CF speed reserve. Take someone who can run 6.5-6.6 60m they could transfer over pretty quickly or someone like twhite whose speed/power numbers are fairly high. Thoughts KU?

6.5 is blazing, twhite is probably good for 6.7-6.8, maybe faster if he trained for it. That would probably be at least a 3.4 30m done bobsled style. There aren’t too many guys who come in running that fast though, probably because they’d rather do track and avoid the -20 degree weather lol. The top pusher ran 3.68 for 30m and beat a guy who runs 3.56 last week. So I’d say that the 6.5 guy could definitely come in at least a top 3-5 finish.

But for most guys out there like me, it probably isn’t possible that our genetics will allow us to even break 7 in the 60m. I understand that there is a necessity to get faster, but I don’t see how training for the 60m/upright running will help guys like us transfer a whole lot to the push track. There needs to be a greater emphasis on strength/power and more push track reps.

Hi Kyle!

I’m confident you could build an easy gliding run-along sled for very modest funds. To stop the sled after you let go, one possible solution is a 10 pound weight and a length of rope. Place the weight in the middle of the rails half way down the slope. As you pass over the weight, the sled will run out of rope and begin dragging the weight as it slows.

If you use PVC rails, you could set them up for a modest number of days, and then break them down for storage.

Anyway, just some ideas to kick around. I’m sure others with suggest improvements or alternatives.


Ku, you can make up whatever reason or excuse you want but the fact is if you aren’t competitive over 30-60m you have no chance of pushing that sled fast. Who cares if people slower than you push the sled faster. Of the top 3 skeleton athletes, how many would you say are in the top 5 for speed among skeleton athletes? My point is to be productive and spend your time becoming a faster athlete who has a chance to make the team. Pushing the sled marginally faster isn’t going to help if someone else runs two tenths or more faster over short distances.

And toscoff off the 60m distance is beyond silly. Talk to your guy Beat Hefti to see whether or not it has a significant relationship.

  1. Regarding your last comment: Comparing bobsled athletes to skeleton athletes is like comparing apples to oranges. Another huge factor that I haven’t yet mentioned is that you do have to drive your own sled. This is the most difficult part of the sport. In bobsled, you have your own pilot that will take care of that for you. It becomes an even greater necessity to become the biggest, fastest, and strongest for bobsled. Regarding the push aspect of bobsled and skeleton, take a look at the posture of pushing the sled downhill on skeleton compared to bobsled. Going downhill overspeed in a crouched position without seeing whats in front of you isn’t the same as running in a lot more upright position downhill and seeing how far you have before you get to hop inside of the sled. When I was up there, I heard from many top bobsledders that the two sports are very different and that they wouldn’t want to bother learning how to drive a skeleton sled

  2. I’m not making an excuse, nor have I EVER made an excuse not to get faster for this sport. I already have been busting my ass for a long time in order to do well for this sport and I will never stop and you are correct in saying that I still need to keep busting my ass in getting faster. What I’m saying that its a reality that I have to face that I will always be at a disadvantage to those who have access to the push track up in Parks city and Lake Placid. Coming into the push champs, I was thinking that I would have a pretty good chance at doing well because of my speed and how well I ranked in the combine (as well as my bodytype). It did not turn out that way for me even and the absolute fastest guy (3.55) who had just a little more pushing experience than I did. I could only dream of skeleton being a sport like bobsled where the fastest and the strongest could win.

The biodynamic structure of the start mechanics is one in which the amplitude of movement is significantly different than dry land sprinting from a low start.

In the skeleton start there is no opportunity for complete extension due to the athlete maintaining hold of the sled. This creates an even more quadriceps dominant regime of muscle work than in the dry land sprint start and early acceleration.

As for increasing the speed of the skeleton start, the same rules still apply in principle:

  • one must train near top speed for that distance in order to improve top speed for that distance

This is where the similarities end, however; due to the fact the athlete is restricted to keeping their hands close to the level of the sprint surface and the degree of resistance one must overcome is far in excess of bodyweight.

It is my contention that dry land sprint work be used in the GPP and begin to yield towards dry land or ice (if able) pushes in the SPP

FIBT indicates that the male sled weighs 43kg with a combined allowance with bodyweight of 115kg.

Aside from making a dry land sled set up as Glaeser has shown, another option are sled or low handle Prowler pushes. Sprint sleds come in various dimensions/weights, the econo Prowler weighs 65lbs, and the regular Prowler weighs 75lbs. Adjusting load is easy enough by using weight plates for an overspeed, comparable, or overloaded effect.

While I haven’t pushed a skeleton sled, I am able to attest to the fact that asphalt, not concrete, makes for a quick surface on which to push a sled or prowler that may very well prove useful for a skeleton athlete.

Add a Skeleton Simulator to your holiday wish list.