I have an BRAT upgraded with the 475 servos. I have tried to adjust the configuration (min/max angles, etc.) as accurately as possible, but I still don’t have it walking well using the demo walking sequences. It is close–I think I just have to make a few more adjustments. But as I work with it more, I wonder it the demo sequences that are supplied need to be adjusted for the more powerful 475 servos.
For example, even the very simple “bow” sequence (standing position, bend 90 degrees at the hip) causes my BRAT to fall on its head because when the hip servos bend they appear to produce so much acceleration that the force causes the BRAT to tip forward. If I run the bow sequence slowed down, using Advanced play, then it works just fine.
Do you think I may have to adjust all the demo sequences for the upgrade BRAT? If so, what would be the best way to do that?
I don’t think the differences you are seeing is due to the 475 servos being used. It’s more likely some other mechanical difference. Are you using the same size or style battery pack? Have you changed any other mechanical configuration? When all of the servos are set to 1500 does the little guy stand straight up?
Oh and I found that by putting small felt pads on each corner of the foot makes the BRAT walk very straight. Tried this Friday night and was blown away at the difference.
I have the standard battery pack that you sell (6v) for the BRAT. 4 Cells in a row and it is located the usual place (U-bracket). Held in place in the U bracket with some velcro.
And, yes, when I set all the servos to 1500 the bot stands erect. It is not exactly straight, but very, very close. I am not sure how to explain this, but if I go into the servo confguration screen and move the sliders just a little bit from where they are set at 1500 so they all are at 0 degrees then the bot is perfectly straight. In other words, 1500 gets the bot standing straight, but I have set the zero points for the servos correctly so that when the servos are at the zero point, the bot is exactly straight.
Thanks for the tip on the felt pads. I was wondering about that. I notice the smooth aluminum feet to seem to slip on some surfaces. I wonder it little rubber pads or the little press on hemispheres (as on cabinet doors) might also be good.
So why do you think a simple routine like “bow” causes my bot to fall over? It sure seems to me that it flips its head down so fast that it just keeps on going and falls over, if you know what I mean. Is there a way to adjust the speed of a move (I am sure there is.) I could see if that fixes the bowing problem.
Felt provides a little slide. Using rubber bumpers is a little too sticky. Like Bowling with tennis shoes. But it depends on what you are trying to do.
Yes there is a speed or time value associated with every move. The manual cover it so I will let you study it and ask if still not clear.
Well, I did some more experimenting last night and I found that even when I simply move the BRAT to the bowing position (hips 90degrees) my BRAT will tip over. Perhaps it is the weight of the serial cable.
But the goods news is I got the little bugger walking. Very cool and a little creepy at first. It does seem to come alive.
I also played with storing the walking sequence in the GP sequencer and playing it from there and that worked just fine.
Ah ha! Yes we used a modified serial cable. We carefully removed the plastic covering, and all wires except ground, xmit and rx. This minimizes the effect of the cable. Sparkfuns bluetooth modem works well too.
