Subject:
|
Re: SSClagorpion
|
Newsgroups:
|
lugnet.technic
|
Date:
|
Thu, 6 May 2004 10:55:49 GMT
|
Viewed:
|
13485 times
|
| |
|
|
In lugnet.technic, David Laswell wrote:
> In lugnet.technic, Steve Hassenplug wrote:
> > I thought about this. The two pieces that "lock" don't need to go beyond
> > their point of greatest extension. The reason for going beyond the PoGE is
> > to lock, so nothing else is required to hold it in place.
> >
> > However the pneumatic piston will be holding it in place (keeping it locked).
> > As long as the braces are in line, there is very little force on the piston.
>
> So what happens if one set of legs manages to lock, but doesn't want to unlock,
> while the other set of legs never locks at all? It's possible that the
> front/back leg pairs will behave differently if the back legs are longer like on
> a real scorpion.
If things do not want to unlock, the entire walking sequence stops. The whole
walking sequence requires that each set of pistons in turn completely transition
from closed to open, or open to closed.
>
> > Also, I'm pretty sure, as you said, the piston has the most leverage on the
> > braces (and the whole leg assembly) when it's close to locking. So if it
> > does have to lift any weight, because of flex in the other legs, that's the
> > best time.
>
> It's less because of flex in the legs (if one set of legs flexes, the other set
> should as well) than it is because a locked pair of legs will be slightly less
> extended than a pair that's just about to lock or unlock, which means the weight
> will transfer over to the walking legs just before they stop walking and again
> just after they start walking.
>
> Exactly when do the stationary legs switch over to walking legs? Do the
> previous walking legs have to fully settle into "stationary" position, or is
> there a slight transition phase where the one set is just about to complete its
> walking cycle as the next set begins its own?
Transitions happen when the switches of the pistons that are changing let
pressure from the center port to one of the two output ports.
In this particular scenario, the walking legs extend to the floor, causing a
timing piston/switch to change state. The timing piston does not represent a
muscle in the body like the pistons in the legs and hips. It is merely needed
to provide the correct sequence of the overall circuit. When the timing piston
finishes transitioning, then the legs that were stationary become the walking
legs.
Yes, the walking legs have to become stationary before the timing piston changes
state. The delay of the timing piston also helps assure that the walking legs
have had time to settle into stationary positions.
Before quad242 could actually walk, I had problems with races between the
walking legs becoming stationary and the stationary legs becoming walking. I
tried all kinds of boolean logic tricks to kill the races, but they didn't work.
In the end, I added the timing pistons (one per leg group) to act as race
killers.
Traditional pneumatic circuit designs have a central timing circuit which
controls the muscle pistons. The muscle pistons are uninstrumented and it is
assumed that they will expand/contract at similar enough rates to make things
work. This can lead to assymetries in the actuations of the muscle pistons,
because no two pistons expand/contract at the same rate.
My pneumatic designs use the muscle pistons as the timing pistons whenever
possible. This means that if four pistons are to change at the same time (as in
the case of putting the walking legs down), the transition is not complete until
leg1 and leg2 and leg3 and leg4 have completed the transition. Only then is
pressure applied to the next piston(s) in the sequence. The whole thing depends
on when the switches say the pistons are expanded/contracted vs. when the
pistons are actually expanded or contracted.
In this case, the accuracy of "actual lock" vs. "lock acquired" as indicated by
the switches is the issue. I'm certain there is a margin of error there, but
hopefully it will be negligable.
Kevin
|
|
Message has 1 Reply:
 | | Re: SSClagorpion
|
| (...) I don't know if this timing diagram helps explain the verbiage above. Forward slashes represent the pistons expanding, and the backward slashes represent the pistons contracting. Horizontal lines represent unchanging pistons. When all four (...) (20 years ago, 6-May-04, to lugnet.technic)
|
Message is in Reply To:
| | Re: SSClagorpion
|
| (...) So what happens if one set of legs manages to lock, but doesn't want to unlock, while the other set of legs never locks at all? It's possible that the front/back leg pairs will behave differently if the back legs are longer like on a real (...) (20 years ago, 6-May-04, to lugnet.technic)
|
300 Messages in This Thread:
(Inline display suppressed due to large size. Click Dots below to view.)
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
