This year was out first year using Dual CIMs for each side of our 6WD, and it has left a sour taste in our mouths. It cost us both a higher ranking in Seattle, and playing in the eliminations on Archimedes. At the beginning of the season we were using 25 chain with aluminum sprockets. During competition we were both shrinking the sprockets and snapping chain at both the master links and cleanly at other pins on the chain. In st louis we replaced the aluminum sprockets with steel sprockets. We worked great through thursday and most of friday, and then our last match friday we snapped a chain at the master link. Then on saturday morning one of the pins in the middle of the chain had a clean, crystalized snap. Every time it was the chain that went from the transmission to the center axle (we use a live axle system). Does anyone have an explanation on why this is happening. We did some testing and realized that Dual CIMs was needed this year to get up the ramp, but its left a really bad taste in our mouths and makes us not want to use it ever again unless we absolutely have to.
The problem you were likely encountering is fatigue failure. You are not loading the chain to its tensile limit(although you may be) but you are coming close, and when you come close enough times the damage begins to accumulate and eventually it fails.
You should calculate the load you are putting on the chain(it will surprise you) and provide yourself a factor of safety on that so you do not exceed the abilities of the chain.
In your particular case I wager that if you were to increase the size of both of your sprockets (to maintain the same gear ratio) you would see far less chain failure. You are increasing the length of the moment arm by using a bigger sprocket, and this will reduce the tension in the chain, but provide the same speed and torque at the wheel. For example if you were previously using a 10 tooth sprocket and a 20 tooth one, change them out for a 20 and a 40.
Also, I have found and been told that half links are weaker than master links. Make sure to get a good brand chain, Diamond is a brand I trust when it matters. All #25 chain is not created equally, and I have never seen a use for #35 chain in a drive system, unless you cannot align your sprockets well enough to use #25.
Good luck
I’m real happy for you, and imma let you finish, but you’re about to make the worst mistake of your entire team career.
Honestly, the CIMs have been nicknamed the drive motors for a reason. Any team anytime (except 2009 at times) not using 4 CIMs in their drive is at a HUGE disadvantage in power compared to every other team competing on the field. If all else fails, your drivetrain will still work, and with 2 CIMs you’re running at half a drivetrain. Half a drivetrain vs. a full one will leave you in the dust, possibly with something broken.
I can almost assure you that your problems were completely unrelated to the use of 4 CIMs. The problem was your chains, your sprockets, anything else, but not the motors. The motors have no relation to the chain. They don’t make things go faster than they normally do. Switch to #35 chain. It’s stronger than #25 chain by a lot, and many teams (330 to name my favorite) have used them successfully.
This is from friend to friend, FIRSTer to FIRSTer. Making a decision like this will be as good as chopping your c-rio in two.
And I’ll ask you this: Why are you blaming it on the extra CIMs? Nowhere in your passage do you say why the CIMs themselves are causing the problem. Reevaluate the event, and what was directly involved in it.
I wouldn’t put the blame on the dual CIMs. Teams have been using that setup for years with few problems. (254, you want to chime in here?)
I’d look at your sprocket alignment first. #25 chain is quite finicky about how it’s aligned–get it off by too much, and you’ll have problems. Hit it right, and you’ll work for quite a while with no problems.
Then, I’d look at your chain tension. Are you really loose or really tight? Either one could cause trouble. (It could also be that you had a bad batch of chain.)
And, if all else fails… #35 chain. It’s more tolerant of misalignment, and tougher than #25 to some extent. But it’s also heavier. You could probably use it on the “weak” link and that should solve the problems.
Like I said, I don’t think it’s the 2-CIM setup, or inherently a problem with the #25. I think it’s alignment or tension that’s causing the issue.
I would recommend this tool:
http://www.team221.com/viewproduct.php?id=70
This allows you to build chain without the master link, greatly strengthening the chain.
Also, with #25 you need to tension the chain. When the chain is loose, the whipping action when the chain is pulled tight greatly increases the load on the chain. You don’t want the chain to suddenly accelerate and decelerate from it being loose.
Our thought is that we are putting too much torque on the chain when reversing directions. Either then or when hitting the bridge were when it usually broke. We have good alignment, and we try to have good tension but when looking at it today it looks like the toughbox mini’s keep bending then we put a drive load on them, which looks to be changing the tension. Currently we are using 16 tooth sprockets on both ends, but might increase that next year. We are also thinking about using direct drive, because it was only ever the chain from the transmission to the center that ever broke.
Sprocket size is not your problem. We are using 16T AM aluminum sprockets in our drive this year.
What brand/source of chain are you using? We do not use masterlinks anymore, but we rarely if ever saw failures due to them (if we did it was usually because of shoddy installation by us) and definitely never saw pins failing in shear. You could have bad chain, but far more likely you have improper tension/alignment.
I would bet that your problem is chain tension, either too tight or too loose. Many many teams use 2 cims and #25 chain on each side with no issue.
Story time: In the past, we used to build H frame modules that consisted of an H shaped piece of welded aluminum (with 2 cross pieces), a gearbox, and 3 dead axles wheels. They were a nightmare for a couple of reasons. One was that they were warped severely during welding, bent outwards, and then occasionally cut up and rewelded. It was awful. Our chain wasn’t even remotely close to tight due to incorrect center to center distancing, made worse by warping and horizontal misalignment, and compounded by chain stretching. We put delrin blocks all over the place to tension the chain, but it ended up following a serpentine path, tight on one side of the sprocket and loose on the other, was really inefficient from all the rubbing, and liked to break. We did this for years and years, with the same results, more or less. So in planning for this year, you wouldn’t believe how hard it was to get the drivetrain crew to build a fairly standard west coast drive. They were absolutely terrified of the chain breaking every match, citing the H frames breaking chain in the past. But we pressed on, built the WCD, and we didn’t have a single problem with it.
We also managed to break 9 drive axles one year (which we solved by removing a CIM), but that’s a different issue.
We are using this chain
Our chain from our center wheel to our outer wheels never broke once. These were tensioned using cams. Our centers were tensioned by moving the transmission up with washers to raise the distance. We use e-clips for alignment, so that doesn’t change. The only thing i can think of and see is that the transmissions are bending under load, which could be changing the alignment and tension.
Those AM were the same sprockets we used at the beginning, and i can get you a picture of one that is round and is missing teeth.
We ran into a frustrating problem similar to the one you are describing in San Diego that probably cost us the regional. We also use #25 chain from AndyMark but have switched to #25 from Mcmaster-Carr after getting some advice from The Holy Cows at SDR.
During the regional at least one chain would break and end up on the field, usually the one from our transmission to drive wheel. This happened at least 7 times at that regional.
Upon closer inspection it almost always broke at the masterlink. It took quite a bit of time to figure out what the problem was. There was a number of things that went wrong.
1: Shaft collar on the transmission output shaft was too close to the chain. If you run the wheels manually it may look fine. But flying at several hundred rpm looks quite different. The shaft collar was basically flicking the master link lock off the masterlink and then the tension on the chain bent the master link pins making the chain looked like it broke under load.
2: Make sure you install the masterlink locks on properly. Another reason why the locks flew off so easily is that we weren’t installing them till they “clicked”. This was a bit tricky for us to tell, the locks can go on and then stay on for a manually inspection (turning the wheels with your hands) but may fly off under faster speeds and higher loads. For both Mcmaster-carr and AndyMark links you have to make sure the lock is installed well, otherwise it will fly off during a match and then destroy your chain.
3: This was probably not the reason our chains failed during the regional, but we made sure it would not be a problem. Our bearing block bolts were not countersunk, we had a 1/4-20 button head sticking out and with the smallest amount of space from the chain. We added another spacer so that it was a good 0.125" away from the chain so it wouldn’t nick the master link.
Not too sure how you broke your sprockets, we abused our AndyMark ones to the max when we ran our robot with slack chain during early testing, we never broke any gears or sprockets and I’ve seen many teams with 7075 sprockets that never failed them.
Hope this helps, it really sounds as though installing the locks on the masterlinks or having something too close that was contacting it may have been your problem. It was quite a puzzler for us to solve.
We didnt have a shaft collar on until st louis, so I do not know if that is an issue, but we will look for sure.
For anyone that is wondering, we use a 6wd and live cantilevered axles with 3 chains coming off of the center axle. 1 going to the front, 1 going to the back, and the one closest to the bearing going straight up to the transmission. We have a 1:1 gear ratio coming off of the standard toughbox mini ratio. All sprockets are 16 tooth steel mcmaster sprockets. We also use the KOP wheels, so it is geared for about 9.5 feet per second.
The dual CIM’s didn’t fail… your chain and sprocket failed.
Had you switched to #35 chain and sprockets… well… who knows what could have happened. We used #35 chain and aluminum sprockets for years and never had a problem. In fact our Breakaway 'bot, with dual cims and a 14:1 gear ratio, just spent two days of continuous driving doing demonstrations, has given “robot rides” to people weighing well in excess of 200 lbs, and can still climb ramps in excess of 45 degrees. The chain and sprockets show no signs of wear.
However… looking forward… use the largest sprockets that you can. The larger the sprocket the less the load in the chain, and the more teeth on the sprocket to transfer the load.
Also consider always having one wheel on each side direct drive. Get the extended shaft for the toughbox, and the keyed mounts for a wheel, and bolt that wheel to your gearbox’s output shaft. No matter what happens to your chains you’ll still be able to move. Our drivetrains have been 100% reliable since we shifted to direct drive.
Jason
P.S. Our drivetrains have been 100% reliable… but during Lunacy we had an unexplained control system breakdown in the quarter finals at Seattle… which adversely affected our alliance, captained by a then rookie team 2898. It was an awful way to end the weekend… little did I know what it would take to get someone to investigate FMS reliability issues…
It sounds like you have only one chain running from your 2 cim gearbox to the center wheel, and then other chains run from that center wheel to the front and back correct?
If so I think that is why you are seeing issues, with only 16 tooth sprockets your chain tension will be extremely high, and you have only one strand of chain to deal with it. If you were to run chain from the motor gearbox to the front, and another to the back, with another to the center wheel then you will have 3 chains to transmit the total output of the motor to the wheels, which will (in the case with all wheels having the same traction) reduce the tension in the chain to 1/3 of what it is now.
Also as a reliability concern, I always run separate chains to the front and rear wheels from the transmission, If one chain fails for whatever reason you still have a somewhat functioning drive train. It takes a double chain failure to stop that side.
I did a bit of math, and assuming the following
2 CIM motors in stall, 16 tooth sprocket, and 12.75:1 gear ratio on your tough boxes, you are applying 848 pounds of force to the chain and the sprocket teeth. Diamond brand roller chain in #25 size is rated to have a tensile strength of 875lbs, so it really is not much of a surprise that the chains failed at the master link, which is typically weaker than the chain it holds together. If you change to a 32 tooth sprocket the chain would only experience 427 pounds of force, which is about a 2x factor of safety.
We do plan on trying to switch to direct drive. We want to prototype that this summer.
Another thing I have done for this years demos is make it so only 1 CIM is enabled until we hold a button on the controller, which is when the 2nd CIM is enabled. This will both reduce the stall torque on the chain, and make the drive less jumpy.
And that rookie team is now having issues this year 
sst,
I have seen many teams use your method to tension chain from the transmission. Often the washers allow the mounting for the transmission to shift, bend and twist under the load of reversing the drive. Without pictures this is pure conjecture but when the trans moves, the chance for the chain to climb up on the sprocket increases. Something has to give. I would bet that a close inspection of your sprockets with a bright light will show some unusual telltale marks. That is where the chain break occurred when it was caught on a high spot or tooth during the stress.
One of the primary rules on FRC2168 is that we have to move. If you can’t move you can’t do anything.
With that we also believe that testing new things in the drivetrain in season is not allowed.
Our 2012 drivetrain consisted of:
- 8WD
- Non-Direct Drive
- AM Shifters (2 CIMs Each)
- 5"x1.5" Colson Performas (on WCP hubs)
- Dead Axles (3/8" Grade 8 Bolts)
- C-C distance for exact chain length
- Spacer/Tensioner Combo (1" Nylon rod stock with non-concentric thru hole)
- #35 Chain
- AM Aluminum Sprocket
We built the drivetrain during week 3 of build season, and haven’t done anything except check tension and grease the gearbox.
Is this fairly old school? Yes. Could we switch to #25 chain? Yes. Why haven’t we? Because we know that our tolerances are not always the best, and #35 chain is really forgiving.
Based on your posts, I strongly suggest designing for c-c center distance and integrating a different tensioning method. Most problems people have with #25 chain is misalignment and improper tensioning.
Yes, best purchase I made all season. Worth every penny.
While that’s cool, it’s not fixing the problem. You need to understand that the CIMs are in NO WAY related to your chain failure. They’re irrelevant to the topic. My advice: Keep all 4 CIMs running all the time, and be careful when using chain.
That’s my advice. Take it or leave it as you choose.
Explain the drive being jumpy?
For demos, you could switch to a less grippy wheel, so the drivetrain won’t be stressed.
I don’t believe you want to drive two motors that are connected together with different values. That will stress the driven motor more. We also had a situation where we tried to calibrate one of the victors on a drivetrain with the other disconnected, and blew out the disconnected victor, but I’m not sure that’s related, but given the price of victors it’s an experiment I don’t want to repeat.
If you really want to try driving just one motor, I would either drive the second at a ‘reduced’ PWM value so it will be coasting, or make sure the brake jumper is not set on the coasting motor.
Can you post a picture of the chassis? Is it possible with the cantilevered axles there is flexing that is causing the mis-alignment? Are the chains near the frame, or on the other side of the frame/bearings?
The jumpiness is when we go to move, the robot lurches forward and sometimes the front wheels come off of the ground. Also when reversing directions the robot almost always goes onto the 2 back wheels.
When in the 1 CIM mode, the other CIM is set to coast mode via CAN, and receives a 0 output. so it is just freespinning.
We have a pillow block in the middle with a bearing on both sides. There is 2 inches of axle coming out of both sides. On 1 side is the wheel, and the other side are the 3 sprockets. The sprocket connected to the transmission is the one closest to the bearing, with the sprocket actually touching the bearing.
The only time it could become misaligned is if the transmission bend and torque down, which would move the output shaft slightly.
And the reason we are thinking 1 CIM vs 2 is that last year we used the exact same drive setup except only 1 CIM per side, and never had any issues at all. As driver I loved the 2 CIMs per side, but the chain breaking caused so many issues that i want to figure out what caused it. And everyone on my team is thinking overtorqueing the chain.