Tyres and Molding

[slurp]

Building on the theory put forward in the Friction page I've just posted a page on Tyres and Molding.

My thanks go to

• Phil Davis for his contribution based on Blade Runners Tyres, also appearing in Nuts & Volts Jul & Aug 2006
• James Carter for his Casting Tips based on Bannana3, also appearing in the Robot Builder Forum

Your comments and feedback are welcome.

best regards,
colin

[slurp]

Following an idea from Peter Waller's Insumountable I've picked up some washable (or re-usable) lint rollers from http://www.choiceful.com at £1.56 they're not expensive.

I hope to post some test results soon... maybe we'll get past some of Peter's advantage or at least see what some of his advantage was!

First impression is that these are _very_ sticky but my guess it'll not take long to get the dust to pass any "paper test" while retaining enough of an advantage - much as we've seen with Insumountable.

Best regards
colin

As per subsequent posts, the Choiceful roller didn't stand up to the job. I'd take the recommendation from PeterW instead:- http://www.caraselledirect.com/zcart.mo ... hp?PID=106

[benedict]

After posting a reply to the Thin Man thread, I've been doing some reading on tires. There's a whole lot more there than I was expecting. This thread in particular is pretty enlightening:

http://www.eng-tips.com/viewthread.cfm? ... 250&page=1

And here's a link to the discussion of Pacejka's Magic Formula, albeit for a racing simulator and not for actual tires:

http://www.racer.nl/reference/pacejka.htm

Tire geometry is a tricky problem. Treating it as a simple friction coeff x normal force equation apparently isn't entirely accurate, even in the case of a robot (i.e. no wet roads, oily conditions, oddball camber, frame twist, etc.)

One distinction I've seen tossed around a lot is the difference between global deformation and localized deformation on a tire. It's for this reason that larger contact patches really do equate to more gripping force, but it also means more rolling resistance.

Another neat one is that for a given contact patch size, a wider tire may have an advantage over a narrower tire because of the shape of the contact patch. It's wide rather than long, so the global deformation means less rolling resistance.

Inflation pressure comes up over and over and over again. For cast urethane tires that more or less equates to the hardness of the rubber being used. Softer rubber maps to lower inflation pressures. The benefit is a larger contact patch, but in the case of a pneumatic tire it also means less lateral stability and higher sidewall wear. I'm curious if those two can be separated when casting solid tires of two different materials. It may be possible to get a larger contact patch, but not allow much in the way of lateral instability. Testing!

Anyway, lots and lots of room to play. Now I need to pick up some urethane casting supplies!

I'm going to try digesting Pacejka's formula over the next few weeks and see if I can find any way to link enough parameters back to observables to stick it in the two wheeled robot spreadsheet. At least to play with. But if it's not possible to go from a part-in-hand (like those lint rollers!) and derive enough information to be useful, it won't be a useful extension of the model Colin already has posted.

Tom

[slurp]

I noted that the IMechE had a few papers on tyre dynamics that may be of interest - just read the abstracts but could get a copy of the papers relatively easily. Also going to dig out some of my old books as I've a few trains of thought to chase.

Over all I'd like to keep it simple, such that you get a ball park figure relatively easily. More than that becomes pointless as most of us loose the control in mixing compounds on a small scale. Hopfully our small solid tyre on a solid hub will ease the process, with thought to the elastic deformation that might occur.

While models for stock materials could be good testing probably throws the results anyway.

I'm not certain how much useful infor we can produce, for the moment it's a train of thought I find distracting.... not to say it won't feature in my reading tho'

regards,
colin

[benedict]

Over all I'd like to keep it simple, such that you get a ball park figure relatively easily. More than that becomes pointless as most of us loose the control in mixing compounds on a small scale. Hopfully our small solid tyre on a solid hub will ease the process, with thought to the elastic deformation that might occur.

Very very true. And yeah, the spreadsheet is already stretching things as it is. I still might pursue it just for grins.

And yeah, it's a distraction here, too. Which probably means I shouldn't hold my breath. I'm kinda wrapped up in some coding right now, and won't really be in a position to take on any serious mechanical projects for a while.

Tom

[slurp]

WHAT A RESULT!

Using the pull test I calculated the co-efficent of friction for the Tamiya Tyres as 0.84. I retested with a band of lint roller round the tyre but the co-efficent averaged about the same or just less. While the lint was gripping the surface it was sliding on the rubber tyre.

Some Super Glue later and the re-test - pull resistance jumped to give us a co-efficent of friction 1.13 about a 35% improvement.

Now the question... will it be enough to beat Peter? I guess I'll still have to find him first

regards,
colin

[slurp]

Just seen an interesting option for sticky tyres from Pete Miles, it was printed in Servo Magazine following an inquiry about a picture of his sumo robot.

Basically he's used standard R/C car tyres then coated them with a Room Temperature Vulcanizing (RTV) silicone gasket material. A thin layer was smeared on using a plastic spatula or lolly stick the night before a competition to give it chance to set enough to pass a sticky test. Like most other sticky tyres he recommended rubbing alcohol or using a baby wipe after each match to keep them clean.


The RTV Silicone Gasket material will probably cost you about £3 for a 3oz tube and could probably be found in most shops car or motor bike spares. My guess is this will give you enough for a number of competitions.

From the pictures Pete was using Permatex's Blue RTV Silicone Gasket which got me thinking about alternatives. The data sheet for this one talks about a cured Shore A hardness of 32 while with similar cure times the data sheet for High-Temp RTV Silicone Gasket shows a hardness of Shore A - 18. Potentially this softer compound may give slightly better traction.

I have used Blue Hylomar when I spent more time tweaking motor bikes than robots but beware it's not suitable for this application - it doesn't set or harden! Hylomar do make RTV Silicone, most of their products appear to be around Shore A - 25 but one or two are harder. The point is there are various alternatives out there, take your time and research your options... it might give you another small edge over the opponent.

regards,
Colin

[benedict]

Since I'm still interested in molding, but not at all eager to tie up all my shop cash on urethane casting rubber, the RTV approach looks very very attractive. (VERY attractive.) So do you know what would be involved in adapting this to the molding process?

I guess one point I need to find out about is whether it requires oxygen to cure. If you shoot a closed mold full of RTV, does it kick? Or are you going to have a mold filled with goop?

Thanks for the post, Colin!

Tom

[slurp]

Since I'm still interested in molding, but not at all eager to tie up all my shop cash on urethane casting rubber, the RTV approach looks very very attractive. (VERY attractive.) So do you know what would be involved in adapting this to the molding process?

If your mold is open to the air at the top I can't see a great problem. The data sheet for the Permatex materials say it's activated by moisture in the air. I wonder if the amount of surface area factors on the cure rate, this stuff is usually put down in a small bead for gaskets after all.

That said it's effectively closed from the air when bolted down. It has a small volume with a large surface area, tho' only the outside edges see the air - the rest is close by the mating surfaces.

If you shoot a closed mold full of RTV, does it kick? Or are you going to have a mold filled with goop?

Well the tube of RTV Silicone isn't exactly expensive to trial, you don't really need a good mold in the first instance - just something rough sized.

regards,
colin

[benedict]

Good point. I'm off to the hardware store to pick up some epoxy for potting underwater cameras, so I'll pick up a tube of RTV to play with. Should be fun!

Tom

[slurp]

Just so it can be found if someone comes looking for tyres... from the ExSpurt thread:-

ExSpurt's wheels and tires. http://brooksbots.com/ExSpurt%20Tires.htm

I find it interesting that you saw the bounce on the Shore A 10 tyres, Tom Benedict had commented on that with one of his 'bots previously. I do wonder how much of this is a function of tyre depth combined with hardness.

I guess there's a long way for me to go from my best co-efficient of friction at 1.13 to your tyre at 2.5!! Seems like a very real limit to the budget approach... must get back to trying RTV Silicone.

best regards,
colin

[Rick Brooks]

I find it interesting that you saw the bounce on the Shore A 10 tyres, Tom Benedict had commented on that with one of his 'bots previously. I do wonder how much of this is a function of tyre depth combined with hardness.

The limit for soft tires is bounce. I've been fighting that limit since 2003. I think that tire depth is a factor. ExSpurt has the thinnest tires that I have molded at .150" (3.8mm) and can use Shore A 20 without bounce. Most of my other sumos will bounce on Shore A 20. (Executioner also uses Shore A 20 tires without bounce(.25" or 6.35mm thick).) I haven't experimented specifically with tire depth, but plan to some time in the distant future.

I guess there's a long way for me to go from my best co-efficient of friction at 1.13 to your tyre at 2.5!! Seems like a very real limit to the budget approach... must get back to trying RTV Silicone.

See the page on Exert-O'Meter for details on measuring push and results. http://brooksbots.com/Exert-O'Meter.htm I have seen results over 3.0, but can not consistently repeat that level of traction. I think that my sumos are consistently around 2.0 to 2.5. Of course, throw a little dust on the sumo ring and all traction goes away.

By the way, in the sumo ring I try really hard to avoid head to head pushing battles. I would rather push the other robot out from the side or behind.

Rick Brooks

[benedict]

Hey Rick,

Have you tried a two-shot process for making tires? I'd be curious what using two compounds, one inner and one outer, would do to tire bounce.

I'm still up to my eyeballs in other projects, and haven't tried shooting my own tires yet. Looking forward to it, though!

Tom

[Rick Brooks]

Tom,

Have you tried a two-shot process for making tires? I'd be curious what using two compounds, one inner and one outer, would do to tire bounce.

I have thought about several different types of tire construction. Most of which would be very hard to construct. Remember that the tire material that we are using was intended for mold making and designed not to stick to anything. It doesn't even stick to itself once cured.

So far, all of my theories on tire bounce have been completely wrong. My latest theory was stated by Colin about tire depth. The two compounds would be one way to achieve less effective tire depth while keeping the wheel size constant. I would just make larger diameter wheels with the same OD for comparison.
Rick

[Rick Brooks]

Tom,

Wait, I just had a thought. Perhaps your suggestion for two compounds is for a softer compound on the inside for more flex and a harder compound on the outside for traction.

I assumed a harder compound on the inside for more stability and a softer compound on the outside for more traction.

I guess that there are two different ways of looking the two compound tire.

I think that the bouncing comes from tire flex and softer tires flex more than harder tires. But, remember that I've been wrong more times that I've been right on tires.

FYI, these tires are very sturdy. I've had them cut and had them loose chunks, but I've never had a tire separate or become unusable. The first two tires that I molder years ago are still in use on Sticky and Excuse.

Rick