I don’t have an awful lot of payload space remaining on my scout, so I’m hesitant to add another power source if I can avoid it.
Power is provided by 5x AA NiMh and seems to be sufficient to run everything (12x5645s in the legs, 2x 422 for pan/tilt) and all of the electronics. Power for the I2C bus is regulated by a Dimension Engineering 5v switching regulator, which does have integrated filter caps (guessing .1uF).
Everything is fine whilst the legs are engaged, however, when a pan/tilt cycle is started the I2C devices continually reset. The same occurs when I disbale the legs. The control electronics (ABB+Coridium/SSC32) have no issues and continue to work perfectly.
At the moment, I’m guessing its noise.
So… Should I just bite the bullet and try find a place to mount a 9v battery somewhere for the I2C bus, or try adding a few more filter caps (.1uF or larger ones)?
Sorry, I don’t have any answers, but I have a question about your bot.
How much noise do the 5645 servos make when your scout is walking? I’ve been wondering if they are much louder than a metal gear analog servo suchas the 645…
I was actually meaning it was electrical noise causing the problems. I think I’ll go with a few extra filter caps of various sizes first.
These are my first servos, so I can only compare them to each other, and they have a completely different gear train and gear material.
The 5645MG is a digital version of the 645MG, I should imagine they sound about the same. That’s just a guess, though.
He’s not all that good at walking yet, we haven’t got much past crouch/stand/lean/shuffle, though he is speaking and does sport a nice 160x128 display…
