Thank you kindly mate. I'll dive in when I get time and see what's what. I've been thinking of trying my hand at the problem, but am glad someone sweated it out already.

Idea for steering
 

Hi all.
I've been following this thread with a great interest. I was wondering about something to make a "semi proportional" steering. We could use a motor for steering, and adapt the wheel of a PC mouse to the transmitter R & L contact. ie: each time the wheel turns to the right, an impulsion is given to the steering motor acting as a step motor that would turn the wheels to the right just a bit... I hope you see what I mean. That would give a more acurate controls to the wheels (while propulsion control could be done with your "potentiometer" system).
Tell me what you think about it.

OK, it's official "pull my finger out of my arse" time....

Expect results over the next week.

ph2t.

So, one month later, how is it going? :)

need for a mechanical mod for proportional steering
 

Hey,

I do a lot of electromagnetic actuation / oscillation for miniature movements - micro-robotics type stuff. I have a lot of practical experience with resonant oscillation of mechanical assemblies.

I think that the idea of a proportional controller emulator is probably the best way to get the effect, and it should be pretty straightforward to get proportionality from the drive motor.

However, the proportionality of the steering might be a bit trickier.

I would like anyone directly working on this project to read through this so that they are at least thinking about it beforehand.

The mechanical/manufacturing characteristics of bit steering assemblies do not look that precise to me in pictures (and I say that because I still have yet to receive the ones I ordered in the mail this week).

Since the steering assemblies are probably quite a bit different from car to car, with tight and loose tolerances, different wheel/tire setups, different springs, etc, the "resonant frequency" of the whole assembly will be different with each car.

In "stock" operation, the coils probably serve to overcome the restorative force from the central spring and slightly tip the balance past some breaking point, after which the motion of the car itself is enough to momentarily force the steering linkage to turn. It would be nice if we knew precisely what forces were applied by the coils under different electrical situations.

I bet that there is a fairly wide (but still critical) range of frequencies where PWM will work on the steering, but it may indeed be more of a coarse vibration around a central steering deflection angle, rather than a constant deflection.

I plan on taking some precise measurements of my bit chassis when it arrives, and I might make a more advanced chassis using my Prazi precision milling machine and lathe. I make mechanical parts smaller than watch works all the time, so bits should be a fun diversion. I also do some electronics work as well.

I will follow this forum. Just remember, nature abhors a change in flux.

Later,

Jesse, welcome to the forums, and let us know what insight you get once you receive your cars, sounds like you know your stuff! :)

thoughts about propo steering
 

I have been thinking a lot about the propo steering. I looked at some of the non-bit cars and decided I didn't like their setup.

I think I have a solution, but it will involve both electronic additions to the control and mechanical additions to the car. I don't like the idea of having to mess with the car, since ideally this would just be a sweet propo emu controller, but it is a different kind of setup than anything I have ever seen, so I think I will at least try it once.

I plan on making what I will refer to as a ratchet steering linkage.

It is basically quantized steering.

The steering mechanism will have a few stable positions that it can be in (such as dead-center, pegged at full left and right, and intermediate positions, although not necessarily spaced evenly).

This will be achieved by making a moderately tough spring with a rounded-off dimpled area on a flat end. This dimple/nub protrusion on the spring will be able to click into some holes (or perhaps depressions) that will be drilled into a thin disc of flat brass, aluminum, or stainless steel. It will look sort of like a record player stylus that can click into some pits on a tiny record.

To actuate the device, a forward or reverse pulse from the transmitter will hit a tiny DC motor with a small gear that meshes into a larger plastic gear that the metal disc with the pits or holes is glued to. This larger plastic gear can have the steering knuckles on it.

I have some 4mm dia by 10mm long ultra small pager motors that cost about a dollar each, which would be perfect.

I will post a schematic of the steering assembly soon.

To create the RF pulses from the transmitter, a computer mouse with a scrolling thumbwheel can be taken apart, and the scroll wheel can be built into the controller. The optical encoder disc setup can be rigged to send pulses on the Left and Right transmissions. The proper transmitted pulse length to kick the steering assembly over one click will need to be determined experimentally.

This isn't fully proportional, and it requires a lot of work, but I bet the performance and feel of the controls would be quite good, and I am sure that the whole thing can be made light, perhaps as light as the current coil setup. Also, we can eliminate those delicate wires to the coils. I bet those things are hard to solder. I have a giant 20lb roll of enamel wire that is finer than human hair. It makes a great low-current coil, but it is tough to solder.

One of the first mods I plan to make is to put pin connectors on multiple contact points of my receiver PCB so I can do quick connects to various electronic mods.

It shouldn't be a problem to get 20 different mechanical click positions into the steering mechanism, and they should each be stable enough that the force on the wheels as the car changes direction won't upset the balance too much. If the wheels start to straighten in an intense turn, just keep spinning that mouse thumbwheel as fast as you can. Unlike a standard trigger RC control, the controller steering wheel could be turned indefinitely in one direction, continuosly pulsing the steer motor.

There would, of course, be mechanical stops at the full left and right positions.


I will build it in the coming weeks, and report back. It might sound complex now, but I think I can design it so it is pretty buildable. Of course, people will need a scrap computer mouse to scavenge from, since new mice cost more than bits.

If anyone has any thoughts or knows if this has been tried, please post.

Hi :)
I had the same idea (about using a mouse's scroll button) but with a stepper motor. I don't know if there are any stepper motor small enough for BCG though. A stepper motor would be much more simple than the setup you describe I think.
Anyway, I'm so dumb in electronics that I never tried anything (but thinking never hurt) :)

The whole steering linkage thing is fine, but the ZZSE's already use it. It's a bit tricky, and I perpetually have to tune mine.
Why not use a variable resistor connected to a cap connected to the TX button leads to control it? The lower the resistance, the more the cap "fires", and the more the car turns. A similar system is used in RFFS controllers, and they work fine.
Alternately, just get a set of Cirrus Mj gear (~150.00) which includes a Rx, ESC, 2 servos. It's designed to use 3-4 cells, and, at 6 grams for the ESC,RX, and 1 servo, it should work nicely. Sure, it's expensive, but you'll have the best Bit in town.