All for my new robot project - a Hexapod. I did a quick 3d renedering on how it is going to walk. If I have anything of interest to show I'll make a new 'robot' entry :)
"BUT, drawing 600-900mA from your pack is going to cause a voltage drop on your PIC and it will brown out."
First, the uC won't be a PIC but a Atom Basic Pro, and indeed I will power the uC with a 9V battery and the servo's with 4 AA cells (maybe a 7.2 pack in the future).
Noted. Force of habit. When you're used to the best, it's hard to accept that some folk still use inferior technology :-)
I hooked up all 20 servo's and moved them at the same time, everything worked fine, so no problemo. If it turns out that it will reset, I wil buy more batteries :)
Of course you can make them move, but can you make them all lift 100g at the same time? Let me put it another way: I can make the windscreen wiper motor of a car move from 4 AA cells, but I doubt it would carry around a 190lb adult.
I don;t think you need worry about teh servos "reset". The analog control board will definitely brown out under load, but because it's analog, it'll pick up again right where it left off and you'll probably not even notice.
You know what? You might be better off with four DDs! Here's the paradox: you'll probably need more power to lift the batteries!
wow! so many plastic bags all together!!
I am interested too in which servo controller you'll be using for your project
4 DD cells, 20 servos, I like it :)
Shouldnt this be a blog-post? Well, whatever!
What is your servo controller?
Will your hexa be blind?
How are you going to hook them up? I have some thingeys to strap onto servos turning thingey, if interested!
Dude, those are AAs. Not DDs. Don't you know anything about bra sizes?
On that point, assuming these are stock R/C servos, they will draw about 300mA each under load. That means you need more juice. Assuming you lift 3 legs at a time off the ground. Let's guess they run 100mA each with no great load. you're into 600mA. Let's assume you drive forward with three legs at a time. Not at full stall (obviously). You're into 900mA. 4 good AAs are only going to offer 1500mA. (OK so far.) BUT, drawing 600-900mA from your pack is going to cause a voltage drop on your PIC and it will brown out.
I found this the hard way with my biped. Only powering 2 servos at a time the uC would reset. Could I suggest you consider a R/C racing pack (7.2V rechargable NiCd)? I would also consider a drive circuit with, maybe, 4x 7805 regulators whose INPUTS are in parallel (for God's sake don't put the outputs in parallel - unless you want the smoke to get out). Three of them would supply each alternate pair of legs on the tripod (assuming you adopt a dual-tripod gait) and one for the CPU.
Look at my servo controller board. See the 3x voltage regulators at the back and the SODDING BIG HEATSINK? You think that's for decoration? :-)
Experiment: drive one servo from your controller. Write a program which increments a counter each time the MCU powers up. This counter should be stored in EEPROM. Put a load on the servo (by grabbing it with your fingers). Grab the servo a few times and then read back the EEPROM contents. You will have found that each time the MCU resets due to brownout, the counter increments.
Ask me questions.
NB - "turney thingies" are knowns as "horns".
It would seem from the picture, that he has a 9V battery to power the uC :-).
Another way to get the power needed for all the servos is to get a DC-DC converter like the one at http://focus.ti.com/docs/prod/folders/print/pth08t220w.html. It delivers a max of 16 A.
In Denmark "horns" are known as "turney thingies"! :-D