Can anyone help with the problem of tracks coming off. When my Rover 5 chassis was delivered the track wheels were slightly out of alignment because of the tension in the rubber tracks. I altered the angle of the gearboxes to reduce the tension but the problem persists.
As requested photos. I’ve programmed my mbed microcontroller and AD754410 h-bridge to go forward 1 metre and then spin 180 degrees and then forward and then repeat. After 5 iterations the tracks are falling off. Thanks for your interest
Hi Coleman
I already had the offer today from Julie at robotshop to replace the item.
Actually that is not the the tightest track configuration as I have already modified it (as I mentioned in the original post.) When it was shipped the tracks were in the tightest possible configuration which has bent the chassis out of alignment. If the item is always shipped in this way then the problem will always be present. A possible remedy for the future is to ship the unit with tracks unattached but you would have to negotiate with the manufacturers in order to achieve this. In reply to your suggestion, if I had wanted a different product then I would not have bought this one.
I wonder though does anyone understand the problem. Situation: tracks come off rover 5 chassis in course of normal operation. I’ve seen this mentioned on other forums and it seems to cause understandable irritation.
There are a few possible scenarios here.First , my particular item was faulty and robotshop have kindly agreed to replace it, customer happy. Second , there is a design fault in this item which future customers will experience,customer unhappy. Third , there is a design fault and it will be fixed so that future customers do not experience it.
Hi Coleman
I’ll wait and see if the replacement is any better. Its possible that many people use this chassis with wheels, then the problem would not be present.
Greg
Hello,
its possible one or both of your rubber tracks are not straight. Remove the track from the chassis and hold it up against a light background. Does it hang straight or is it curved ? if it is curved then it is a manufacturing fault and it will most likely always come off the driving wheel. On my Rover 5 one of the tracks was curved and twisted and it always came off. The manufacturer should supply Rover 5 with straight tracks or else replace them. My chassis was replaced by RobotShop with another which had straight tracks and now seems to work fine. If you look at my photos above you will see a curvature in the top track. Increasing separation and tension on the tracks did not solve the problem. Best of luck with your project.
I realize this thread is almost a year old, but I did not see a solution, so I thought I’d bump it.
I just tested my tracked 4WD Rover 5 this weekend and had the tracks come off when the Rover repeatedly turns. It is worse on carpet than on a smooth floor. The motors are speed controlled so that the ones on the same track run at the same speed. I also tried 2WD by disengaging the motors but the tracks still came off when the Rover turns. The wheel “legs” (arms?) are set about 3 notches from the fully extended position, so I may not have enough tension on the tracks. My Rover doesn’t appear to be warped (I keep the tracks off when it is not in use), and I don’t have much weight on it, just the minimum to steer and control the motors, plus 6 AA batteries.
I am planning to eventually replace the tracks with wheels but I first would like to get the tracks to work. Aside from increasing the track tension (which may warp the legs) and keeping the weight off, is there much else that can be done to make the tracks stay put? Is this an issue with other tracked designs, or is this particular to the Rover 5 due to the stretchy tracks?
Thanks in advance for any help on this issue.
Thanks for the tips. My tracks look pretty straight so I don’t think the issue is with track straightness. I set the angle of the legs to one notch from horizontal (previously set at 3 notches), then installed the tracks. The increased tension caused a very noticeable misalignment, similar to the photos above. I ran the Rover 5 a little and the tracks came off again while I was making it turn. The tension on the tracks was so much that it pulled one of the legs partially apart and stripped the thread. Increasing the tension doesn’t seem to work in my case.
Needless to say, it looks like I won’t be using the tracks. I will try some mecanum wheels next. If not for this issue the Rover 5 is a good starter robot platform to learn from. Hopefully Dagu will tweak their design to correct this and other (?) issues and make it even better.
Thanks for the manufacturer’s suggestions. I will try them if I can repair my stripped hole. However, I suspect the issue with the tracks coming off has more to do with the wheels becoming misaligned when the track tension is too high, or the tracks becoming loose when the tension is too low. So the solution is to either find a happy medium between too high and too low track tension, or to resolve the misalignment issue.
I recently discovered something on my Rover 5 that may explain the misalignment.
The Rover’s legs are very stiff (when the 2 halves are firmly attached to each other), and the method of anchoring the leg to the body is very secure. However, the plate that is anchored to the body is held to the leg by only 2 screws, and on one of my legs the threads of the hole of one screw was stripped. This is the 2nd stripped hole I found on my Rover 5.
With only 1 screw holding the plate to the leg, the leg moves more than normal. The excessive play causes my wheel to become noticeably misaligned when the track tension is increased.
I did not see any loose screws on the other 3 legs, but I did notice that a gap opens up between the leg and plate when the track is installed. This happens because the 2 screws are not optimally positioned to resist forces from the track, especially when the legs are in the horizontal position and the track is at maximum tension. For horizontal and near-horizontal leg positions, it is better to arrange the 2 screws in line with the leg, 90 degrees from the current position. If cost is not an issue, adding 2 more screws, for a total of 4 screws, might be even better as it will work for more leg angles.
My other stripped hole (mentioned in an earlier post) was on a leg on the other track and occurred at one of the screws holding the 2 halves of the leg together. There are 3 screws for each leg, 2 near the motor and 1 near the wheel. The one that stripped was near the wheel. When this screw is loose the leg halves will separate and bend when track tension is increased.
Therefore, I believe that the stripped hole threads at both locations are the root cause for my tracks coming off when I increase tension. This may also explain why only a few Rover 5 owners are experiencing tracks coming off, because only a few will have stripped threads or loose screws. In addition to revising the screw positions for the plate to leg joint, it may also help to have longer and/or larger screws in critical locations (i.e., when there is tension on the screw) to reduce the risk of thread stripping.
I will attempt to repair my stripped holes, and if successful I will try the manufacturer’s suggestions then report the results. After this, I’ve decided that I will forego the tracks and start playing with the mecanum wheels.
Hopefully these comments will be useful to someone, or to the manufacturer if they are looking for ideas on how to resolve this issue.
I tried the manufacturer’s suggestions about correctly installing the tracks after I repaired my stripped holes, and found that it makes a huge difference in preventing the tracks from coming off. However, on my Rover 5 the tracks still came off, but only when it rotated in the clockwise (CW) direction. Both tracks rode up and over the inner rims. During my brief testing, in all other directions it appeared to work pretty well.
I then thought that increasing the track tension just a bit might make it work in all directions. So I adjusted the leg angle just 1 notch to increase the tension, then tried running again in different directions (i.e., forward, backward, rotate CW, rotate CCW). It got worse. Now the tracks came off when rotated in either the CW and CCW directions. I also noticed that the leg and wheel misalignment was greater. When I inspected the bottom of the Rover I saw that a gap opened up between the leg (with the repaired stripped hole) and the plate. My hole thread had stripped again.
So it appears that correctly installing the tracks does help to keep the tracks on the wheels if the misalignment is not too much, and that misalignment can get worse with increased track tension, especially when hole threads are stripped or screws are not tightened.
I will attempt to repair my re-stripped hole once more. Going forward, I will give up with the tracks and start working with mecanum wheels.
Hi Coleman,
I purchased my Rover 5 from Sparkfun back in April 2011. I’m pretty sure I’m outside of their warranty window. Thanks for the offer though, and for all your help on this topic.
Update 3/1/2013 - I contacted SparkFun about my Rover 5 issue and they offered to exchange it for a new one. I will take them up on their offer. From the prior post, it appears that RobotShop probably would have done the same. Kudos to both companies for great customer support.
Update 4/15/2013 - Received and tested replacement Rover 5 and did not have any issues with tracks falling off. New wheels have bulges at the center. Rover 5 was shipped with tracks on and with the legs 1 or 2 notches from horizontal. Testing was performed on a smooth hard floor and on carpet with the legs in the “as received” position.
Can you take a photo and post it here to show your current configuration?
Hmm. That’s the tightest possible track configuration. Do you think exchanging the tracks or the robot would solve the issue or would you like to exchange for a different product (assuming you are within the exchange period). We certainly apologize for any inconvenience.
The issue of tracks falling off the Rover 5 has not come up often (at least with RobotShop) and most people seem to be happy with their purchase. We will certainly keep an eye out of additional issues with this platform and if we see it persists, then we will report it to the manufacturer.
Aside from adding tension to the tracks, there is not much else you can do. The type of surface the robot travels on also has an effect.
We have contacted our supplier to see if they can offer any ideas / solutions.
This is the reply from the manufacturer:
If the robot is still under warranty, we can always explore an exchange (contact us via the Support Center).
This is a manufacturer design error. It did not correctly calculate the pressure of the tracks relative to the wheel mechanism, so they tend to misalign the mechanism and the tracks come out easily.
Another factor for the tracks to come off is at the wheel rim, because if the wheels are misaligned, the raised edges of the wheels make it come out quickly.
At the time, I fixed these problems by using a set screw to align the wheels with the tracks and I removed the wheel rims.
My Dagu Rover 5 setup is as follows, 02 gearbox motors, Sabertooth motor drive and 9ch radio control, running at least 5 years.
The cause of the bad wheel alignment on my rover is not as svevig described.
The stub on the gearbox where the wires exit locates in a hole in the chassis. This hole and the end of the gearbox mount are not properly aligned. This causes the gearbox to mount at an angle and that causes the end of the gearbox housing to have sideways pressure, possibly causing damage to it.
Solved this by enlarging the hole toward the center of the vehicle with a small half round file. This allows the gearbox to align properly. I printed a new fitting to force the stub into the required location.
You might be ok by putting washers under the metal gearbox retainer at the inside screw, forcing it to push the gearbox stub forward.