Help with hexapod robot needed

Can any experienced hexapod robot designer help me with following?

This is my situation:
My hexapod is performing all it’s supposed to, as long as it’s in “dry dock”
on body support with it’s legs in the air.
When put on the ground it wouldn’t lift itself up.

It has:
a) 18x JX PDI6221MG servos in it’s legs ( 20.32 kg-cm / 281.89 oz/in ).
b( just regular servo brackets, aluminum tubing legs ( it couldn’t get much lighter than it is ).
c) carrying just Arduino Mega and SSC32 boards.
d) 6V wall power source ( so it’s not even carrying a battery ).

I got already third set of servo motors, going higher and higher with torque,
but result is not getting better.
It totally doesn’t make sense to me, I’m suspecting my power supply setup may be the issue,
but I may be wrong on that.

I’d appreciate any help and/or idea on what I’m doing wrong.

_Honza

All down to current, I think.

Even a micro servo draws 730mA according to a source I saw recently, and 1A is a rule of thumb.

Let's see what current your servos draw at stall?

20 kg cm is 2 Nm
The servo speed I looked up is 0.16 s/60, call that 6 rad/sec.
So the power P = Tω is 2x6 =12W, which at 6V is 2A.

If you have 18 servos all trying to start at once, you need nearly 40Amps.

Does your wall charger thingy give that?

hi manor_royal, thanks a lot for your reply,

:slight_smile: … ehhhh … not, it says 1000mA on my charger.
it’s probably way too little, as I look at your calculations.

BTW, I do compliment to your knowledge.

So, you suggest I’d get appropriate power source ( giving 40 Amperes ).
I’ll look it up, assuming they sell anything as powerful as that.

May I ask you some more, since you know so much about this electricity issues?

  1. this 40A charger isn’t going to fry my servo controller, is it?
  2. do LiPo batteries give also this much Amps?

Thank you again for your help and amazing expertise.

1000mA is not even enough for one of those servos: barely enough for a small one.

Is there any way you can engineer the sequencing so that fewer servos are starting to move at once? (That will be a bad time, since servos need their stall current at startup… maybe you can “spread the load” so to speak.)

well, there are some motion sequences where it is possible to move legs one at a time,
but in most sequences all legs (all motors) have to be moving, at least to some extent.

I've just look at what there is when it comes to power supply options
and no matter if wall chargers or batteries, they all talk in milliAmps,
with some 4500 mA for LiPo battery being about the most I've seen so far.
With this I'd need like two batteries for each leg :slight_smile:

I'm totally lost in this.
Why I see on Youtube all kinds of hexapods running around with a battery (or sometimes with 5 AA battery pack), moving all legs simultaneously?
And most of them just can't have any torque weak motors, considering their constructions and proportions.

It's the startup current that's high: they don't draw that when moving except under heavy loads.

But I've never worked with a hexapod so I can't help with the practicalities I'm afraid.

What the "startup" means?
To move the servos at all very first time after power up?
In my case they move just fine, since the legs don't touch the ground during the first motion.
They just rotate to "lift up" positions.
It's after that, in lift up off the ground sequence, that motors fail and the hexapod collapses.

You've been great help to me manor_royal, and I do appreciate it.

Thank you again for your time and effort.