We are building a WCD and have a design requirement to put the Chain in tube for the front set of wheels. We are using the Versa blocks so the drop is set. Since we have never ran Chain in the history of the current members we decided to error to caution and run #35. We did get 12t sprockets but at a loose glance they don’t look like they are going to play nicely. We did get a set of https://www.andymark.com/products/sprocket-17t-25-chain-double-sprocket-for-wcd for this just in case sinareo and 25H chain but still would like to run with #35.
There has been lots of discussions on this topic and I have been sorting through the post. It seems at times one run of #35 should fit in a 1x2 Versa Tube Chassis. If we were to switch to #25H then both runs would fit.
We had chain-in-tube in 2017 and ran 25. I was initially skeptical that it would be strong enough, but we pushed a lot that year and didn’t have any problems. (That chassis was still running strong until 2 weeks ago when we tore it apart.) Based on what I saw in that frame, I don’t know how 35 would ever fit.
One thing that it does that we didn’t expect was limit how we could attach structures to the drive rails. We couldn’t use the standard Versa mounting holes because bolts/rivets would contact the chain. As long as you design around it, it’s not that big of a deal.
First off, assuming you are tensioning your chains and assemble them correctly, you will not break a #25 chain in an FRC drivetrain without huge shock loads. With the available COTS components (sprockets, bearing blocks, etc.) it is possible to build a very robust #25 chain drive with minimal effort. The rated breaking strength of #25 chain is 930lbs, which is very difficult to see in FRC unless you have extremely high shock loads (such as ski-jumping defenses and then landing in 2016) combined with a large wheel and small sprocket diameter.
Second, #35 chain has a width of 0.456" so you will not fit two in a 1" OD tube of any real wall thickness. This means you will have to use a wider tube. With a 12T sprocket, the total height of the sprocket with chain will be around 1.66", which will comfortably fit in a 1/16" wall tube but might not fit in an 1/8" wall tube depending on how much you drop the center wheel(s).
Third, when you put the chain in the tube, you can’t use the top and bottom face of the tube to attach anything since the rivets/nuts/bolts will hit the chain. You can either mount from the side or weld attachment points for mechanisms.
Overall the chain-in-tube method is very robust if you use #25 chain, and the new double sprockets and bearing blocks make this easy to build. The chain is kept safe from debris, and the packaging around the drivetrain improves. Build using #25 chain, watch your build quality, and utilize the wealth of advice from CD, and you will not break a chain.
We already drilled a set of holes on the top and bottom of the section we planed to Chain in tube at .25" from the edge. (Looks like Yahtzee with fives).
Drive train is 6 mini CIMS on two speed ball shifters with 2-4 6" VexPro higrip and 2-4 Omnis. (Total 6) JVN calculator has the adjusted speed at ~ 7 and 17 ft/sec.
The rated operating strength of 25 chain is 200lb, which entirely possible to see in an FRC drive train. As a matter of reliability, I make a habit of not exceeding manufacturers ratings.
The ANSI standard for #25 roller chain is a minimum ultimate tensile strength of at least 780lb. A quick search for specs by chain manufacturers shows them to all be fairly close to this (The first link I clicked showed 930, the number I used in my first post). Different chain manufacturers have different working load or power ratings/rating curves, which include different lifetime failure modes. The failure modes at high load include link plate failure, but also pin impact load and pin wear from continuous use (two failures which FRC robots do not run enough to reach).
Quick calcs show the maximum load on a drive chain to be about 340lbf assuming 6" wheels, 150lb robot, 1.1cof, and 18t sprockets, where each chain takes the full load of one side of the drive. Most WCDs direct drive one wheel, so loads from that wheel would not be included in this. This still leaves a safety factor of at least 2 before breaking the chain, more if you don’t have all wheels going through this chain. The chain will wear more due to the increased load, with a life expectancy of 15,000 hours, even halving the life should be orders of magnitude more than an FRC robot will see.
Also, this page has a lot of useful information on roller chain design and wear:
We ran #35 chain in tube last year with 12t* (double) sprockets in a 40x60x2.5mm tube, exact C-C without a tensioner. It took an hour or so to assemble, and didn’t have any problems or need adjustment all season.** It may be a bit harder to find metric tubing near you, but if you can the sizes work well.
All that being said, we decided not to use chain in tube again this year. We started seeing wear on our offseason robot that used it, and we want to make sure that it will not fail mid-competition because repairing it would have required us to take apart nearly the entire robot. Using chain in tube also requires some interesting mounting methods to make sure the bolts/rivets don’t interfere with the chain runs, which we didn’t want to deal with this year.
*IIRC, might have been 15t
**On our competition robot seeing 2 discticts, DCMP, and CMP and on our practice robot with 100+ hrs of driving
Turns out that for us #35 with 12t sprockets in VersaTube and VersaBlocks just won’t go. We are running to the edge with #25H. I’ll let you know how it goes in March.