Re:Hexapod design drawings

[Appropos of hexapods, it seems like the main problem with them, aside from
the difficulty of writing the code to tell them what to do, is the way that
they cut, in a sort of sweeping motion. (Not that I've seen any in action,
but this is how it seems like they would work.) It seems like the machine
would have to be really big to cover an equivalently big workpiece, with a
proportional added expense. Has anybody ever made a hybrid machine; using a
small hexapod as the business end of a conventional bridge or gantry style
router? This would give a lot more flexibility to this sort of unit, with
more ability to get into undercuts than the usual 5-axis set-ups, along with
being able to cover more territory. Writing the code would be a
head-scratcher, for sure, but programmers always like a challenge, right?]

Andrew Werby
www.computersculpture.com


Doug Fortune <pentam@...> wrote:

Subject: Re: Re: Hexapod design drawings

rab@... wrote:

>
> Doug,
>
> How are the joints for backlash ?

The socket universals that I purchased, have no backlash.
And of course being designed for high torque (a hundred or
more foot pounds of torque) they are up to the task (for a
small machine).

> Can you pre-load the axis on a hexapod design ?

That is the basis for a hexapod's phenomenal rigidity at
such a light weight!

Of the six actuators, three are always in tension, acting
against three in compression. Coming to a triangular
mount point above, the pairs

AB CD EF


instead are paired differently at the bottom (tool platform):

BC DE FA


so if AC&E are in compression, and BD&F are in tension at
the top, then each actuator under tension is matched with
one under compression. So it is also at the bottom (ie
B under tension is paired with C under compression, etc).

Keeping this state of affairs of course is one of the
constraints of the hexapod controller software, and
hence very difficult.

- - -
Some of you are no doubt thinking that under great tool
pressure, all of the members will be under compression.
In fact (or should I say, I understand) that the whole
structure would then relax (causing positioning error)
when the compression is relieved.

Instead, the tensions and compressions are pre-set to
far overwhelm the tool pressures, and hence there is
insignificant relaxing (dimensional change) when the tool
pressure is relieved (helping maintain tolerances).

Or so I am led to believe......

Doug Fortune
http://www.cncKITS.com