Very rough Mongoose robot manual

This will be fairly detailed construction manual for the Mongoose. I'm posting it for anyone who has followed the design and would like to offer feedback. It's etremely rough (just a few illustrations & schematic at the moment)
But these things are selling fast at the only store that has them in Toronto

**broken link removed**

Here's a photo from the underside.
**broken link removed**

The chassis still looks great. Did you say that that was a local manufacturer? And I'm curious - did you end up just having to send them some sort of file exported from your designs in Sketchup?

I don't know if you've seen my latest posts on my recorder robot, but my next project ambition for application in future, similar robots is to design some sort of universal finger. I'm not exactly sure how I'm going to do that, but I know I'll eventually want a lot of them. For that reason I'm trying to find out how I might have them easily built (or parts made for my assembly) on a limited prototyping budget.

I wonder if you're interested in collaborating, even if just casually through this forum, on the design of a robot finger. It seems to me that it might be something marketable to your current and future enthusiasts - something that can simply, cheaply, and in a compact space quickly grip and carry an object. I know that if you already had them, I'd be buying them.

Sorry, didn't mean to hijack your thread - just thought it might be something of interest to you.

Some of what you need to address.

The instructions shows foam spacers on the outside of the motors. They do not show the spacer between the motor which is part of the kit.

The instructions show 4 screws going in each side of the gearbox but the gearbox only has 3 screws per side.

Is there any way you could add a breaking feature to the ball bearing wheel? I imagine you're going to have some center of gravity issues if you try to move this thing fast because you have no traction on the bearing. It'll work fine at low speeds, but to really use it to it's capability you'd need some way of controlling that last ballance point as a traction element, or some really well developed control loops..

3v0 said:

Some of what you need to address.

The instructions shows foam spacers on the outside of the motors. They do not show the spacer between the motor which is part of the kit.

The instructions show 4 screws going in each side of the gearbox but the gearbox only has 3 screws per side.

Click to expand...

You're right, I'll have to draw the spacer. The foam bits are simply open cell foam insulation (like you use in doors) I found after building a dozen of the double gearboxes a little snugness on the one side of each motor kept the pinion gear from binding with the crown gear.

On a side note a local artist wanted very long battery life, turns out you can just squeeze a set of C cell holders on the back.

Sceadwian said:

Is there any way you could add a breaking feature to the ball bearing wheel? I imagine you're going to have some center of gravity issues if you try to move this thing fast because you have no traction on the bearing. It'll work fine at low speeds, but to really use it to it's capability you'd need some way of controlling that last ballance point as a traction element, or some really well developed control loops..

Click to expand...

It's a very small robot and I've not had any problems with rapid braking/turing/reversing so far. With batteries mouted it has a very low center of mass. I'll have to record a video of it.

It's faster than a BOEBOT (Parallax) and although I'll have to do real world testing the math on it is.
9710 RPM motor under load, (12300 @ 3V no load)
https://www.pololu.com/products/tamiya/fa_130ra.pdf

So 9710 / 114.7 = 84.66 (RPM at wheel max)
And 84.66 * 3.14 * 58mm (Wheel diameter) = 22595mm per minute or 22.6 Meters per minute.

I do not have a Mongoose but I have used the same gearbox and wheels. They way Bill designed Mongoose it has a very low center of gravity and should be very stable. Also the caster contact area is a very small patch on the bottom of the steel ball. Not much traction.

In the spirt of lets do it anyway.... The top side of the ball is exposed. Put a small solenoid above it. The solenoid could drive the ball down off the 3 rollers and pin it against the cage.

Sceadwian said:

Is there any way you could add a breaking feature to the ball bearing wheel? I imagine you're going to have some center of gravity issues if you try to move this thing fast because you have no traction on the bearing. It'll work fine at low speeds, but to really use it to it's capability you'd need some way of controlling that last ballance point as a traction element, or some really well developed control loops..

Click to expand...

Presuming a 1/32 scale comparison to a life-sized car, I figured an equivalent speed of 50km/h. That's pretty fast compared to the cheap R/C off-road vehicle I picked up in a local, national-chain electronics store.

Sigh... seems like no one wants to give me the finger.

Wouldn't the ball bearing offer almost zero stopping? Even if it didn't turn at all the main drive wheels would merrily pull it along over most surfaces.

How much will this set me back?

The kit has an msrp of $139US (assembled & tested add $25)
You'll need a PIC programmer like the Inchworm to use it.

A top mount prototyping board is in the works (Pelican piggyback / prototype) with space for a ft232r, mini b usb to rs232 (ttl) that will hook directly to the mongooses serial port.