Total cost?

[joey_picus]

I'd like to make my own controllers one day but it's not really my department I am trying to have a diverse range of electronics without using any one controller in all my robots though. Diversity is important and should be encouraged in everything in life, and that goes for speed controllers too! The robots I have/am building at the moment have a good mix of Roryboards, Peter Waller's controllers, Scorpions, hacked brushless controllers, an old Sozbots, pretty much everything. I'm determined not to have all mine be the same internally!

[BeligerAnt]

If you don't want an "all in one" solution (and I can understand that) then there are several options for "rolling your own".

Simplest is to buy any off-the-shelf speed controller and connect it to your own receiver, but there is little learning to be done.

Next simplest is to hack some controller boards from standard servos. They're cheap and simple to hack, but the performance is mediocre to say the least. A good stop-gap if you're planning on moving on to something better.

Moving up the complexity ladder to a bit of software work there's the Pololu Baby Orangutan board. It's a small Arduino-like board that you need to program to convert receiver pulses to motor drive. More details here. You don't actually need to write any software but you do need to figure out how to get the program into the Baby O.

It's also possible to hack the software in some of the small, cheap brushless speed controllers to make them work with brushed motors. Details and open-source code here. This will probably involve figuring out details of the hardware, modifying the software and figuring out how to reprogram the hardware.

Finally you can embark on your own journey of discovery and design your own speed controller from scratch. This will involve electronics and software design, and although neither are particularly complicated there will be a steep learning curve if you are not experienced.

Good luck whichever route you choose!

[Rapidrory]

I'm just thinking how frustrating it would be if you had all the motors, receivers, batteries but couldn't use them until your learnt enough to make your own controller

That's exactly what drove me to finish developing mine! I had the robot finished and ready to go, it just needed a controller. I had to learn how to write firmware from scratch (and for some reason chose to do it in assembler... don't do this unless you hate yourself :L), and PCB design, and surface mount soldering, and basically most of the other skills required, but until I had it cracked I wouldn't be able to compete (I'm stubborn and refused to buy commercial ones :L).

[hobojoe]

Thanks for the replies gents, its a long fun road to run down. The fact that there is a "I really need this working now" option makes it all the better.
The fact that the default response is the kit is testament to the fact that it is bind and drive and that its solid. Having a backup plan is always good.

Rory if you don't mind, your kit, I assume its a controller that has been configured such that the channel mixing is done on the controller rather than the transmitter, is this correct? Its not just a simple, 2 ESC's receiving signals from PWM?

[Rapidrory]

Yeah, signal mixing is done on board, but it's as simple as adding the two control signals together for forward and reverse, and subtracting the two control signals to get steering..

Here's the development thread for my ESCs... dunno how helpful it is, but basically I think the moral of that story is keep it simple! Just don't copy anything off the first design as it was rubbish :L http://robotwars101.org/forum/viewtopic ... it=tinytwo

[mcscruff]

Its here!

uploading pictures