DIR Return Create A Forum - Home
---------------------------------------------------------
input-inc
HTML https://inputinc.createaforum.com
---------------------------------------------------------
*****************************************************
DIR Return to: Questions and Answers
*****************************************************
#Post#: 25--------------------------------------------------
ERROR In My Own Design
By: Robert Snyman Date: February 1, 2013, 4:01 am
---------------------------------------------------------
The track tip action on my design is giving me a massive head
spin. I have redesigned and recalculated so many times I am at
that point to just leave it as it is. When the robot is in its
normal position and the track tip position he will be perfectly
balanced with the caster-A frame mech. But it`s between the
starting and end point where it all goes to hell...
When track tipping the caster-A frame will move down slower than
the track units causing the robot to tilt over backwards as it
moves up. Then suddenly the A frame will catch up quicly
bringing the robot back upright. And if he goes beyond the end
point the robot will tilt over forward.
I have seen mechanisms like that before when I was still in the
air force on the missile and fuel trollies but there the links
are balanced out equally. As for the cradle base mech the links
are not equally balanced over eachother if you know what I mean.
All I can say is Eric Allard is a genius and I take my hat off
to him
#Post#: 26--------------------------------------------------
Re: ERROR In My Own Design
By: J5GURU Date: February 1, 2013, 10:14 am
---------------------------------------------------------
Hi Robert,
Yes, The track Tip is a tricky piece of
Engineering and Geometry to get correct. At it's most basic
level it is just 2 parallelograms. One being the Track tip tube
lugs and upper rear A frame, the other is the upper and lower A
frame with the Caster wheel. All of those dimensions must be
perfect or you will have the exact problem you are which is a
non linear movement of the entire mechinism. We designed our
Track tip and rear caster linkage system to be fully adjustable
in order to be able to "dial it in" to a perfect movement and
function.
BTW.....While the Track Tip Geometry is Tricky.....The Lower
Torso parallelogram and Geometry is a nightmare!....LOL.
This seeming simple system is far more difficult to get correct
than most people realize. Send me as much data and measurements
of your current system and or post here for all to learn and I
will try my best to see if I can figure out where the problem is
and perhaps how to retrofit your design to bring it back into
proper operation. J5GURU@comcast.net
To everyone,
PLEASE NOTE! If you look closely at the trip pictures from
J5DRTDAY1 351.jpg - J5DRTDAY1 357.jpg, You can clearly see that
one of the 1/2" Track Tip Rods to the rear Caster upper A frame
is bent! This was a design flaw in the original Robot. Those
steel Rods and those SKF Actuators failed!!! We had to icrease
the Rods diameter and increase the SKF Actuators to a larger
size to correct the problem of the actuators failing under load
and the Track Tip rods failing! If you look through the pictures
and videos....you realize that the Robots Tracks should NOT have
moved from level to tipped when me and Eric put J5 over on its
side. This was only possible due to the failure of BOTH of the
actuators. Normally the Tracks remain static and maintain track
Tip posistion as the Actuators are self locking when unpowered.
The ones they used were far to small and cannot handle the
extreme forces of that mechinism. The Actuator's ACME ball and
screws failed.
One additional note, The original Track drive pads had exposed
pop rivet heads. While ok for filming...they will eventually
wear down and break..... They will also scratch or dent the Hell
out of any hard surface J5 may be traveling on. If your showing
your J5 at a show with hardwood floors you will likely cause
great damage to those floors when running J5. We have redesigned
and corrected that flaw by eliminating the Pop Rivets in place
of countersunk Button Head scocket screws which are below the
Track Pad Surface!
These are but only 3 minor examples of just how difficult and
time consuming it has been to redesign this incredible Robot.
Eric and his Team where truly ahead of thier time and as
talented as they come!
#Post#: 29--------------------------------------------------
Re: ERROR In My Own Design
By: Robert Snyman Date: February 1, 2013, 10:46 am
---------------------------------------------------------
Hi Terry
Been a long time
Yes I am having trouble with the lower torso mech as well. In
2009 I built a scale model of the robot out of presswood to test
the lower torso mech and it worked beautifly (after many
adjustments). But I can`t get it to work on CAD. The timing is
way off and I can`t seem to find the propper fastening points on
the cradle base for the swiveling parts and tie rods. If you
want I can send you the CAD files for the lower torso and cradle
base setup? I just think it`s better than looking at snapshots.
I would love to share my input with everyone but keep in mind
that my robot differs from the origenal. I`m going for a modern
look. But when I get to the software and programming I will let
everyone know. The university of Pretoria (university of
technology) will be involved with that development.
#Post#: 30--------------------------------------------------
Re: ERROR In My Own Design
By: J5GURU Date: February 1, 2013, 11:13 am
---------------------------------------------------------
Hi Robert,
No problem, Send what you got and we will figure it out. My
Email has a 10 Mb limit. I can work with SW files.
#Post#: 31--------------------------------------------------
Re: ERROR In My Own Design
By: Robert Snyman Date: February 1, 2013, 12:07 pm
---------------------------------------------------------
Hi Terry
Are there any other formats I can send you? Inventor does not
seem to list SW
#Post#: 32--------------------------------------------------
Re: ERROR In My Own Design
By: J5GURU Date: February 1, 2013, 12:18 pm
---------------------------------------------------------
Hi Robert,
I was referring to Solidworks file formats. Not familiar with
Inventor assembly or parts formats but any 3D CAD format it
exports to I should be able to open.....if worse comes to worse,
we will just have to send DXF files which both programs should
open without issue.
#Post#: 38--------------------------------------------------
Re: ERROR In My Own Design
By: J5GURU Date: February 1, 2013, 5:35 pm
---------------------------------------------------------
Hi again Robert,
It's at times like this that this project and "leaked" former
Team Member J5 information and there designs REALLY sucks! I
hate being the bearer of bad news.....
I did get your files ok and I see your problems all to clearly.
One of the problems is that over the years former Team members
or other persons that somehow got ahold of a copy of the 2004
Trip Videos and Pictures have Cadded there own set of Designs
with the best of intentions. Then other good people such as
yourself try to build J5's based off of that information. We
have yet to see any plans or Designs other than our own
Input-Inc designs that are anywhere near correct.
You are yet another victim of someone elses mistakes,
oversights, assumptions and lack of the most basic adhearence to
the 2004 documented measurment information. Two of our newest
Team members, Joe and Drew had to throw away every part they had
made and start completely from scratch because they also were
building from similar designs that were not just a little bit
wrong....but rather was in another Galaxy!
This is one reason I personally tried so hard to keep the
information secret and only having the correct designs come from
the Team.
Please see the attached Photos with the Correct designs
superimposed on the "not so correct" ones.
I wish I could help you out, but the fact is that your trying to
make the original J5 Geometry, Electrical and designed for
mechinisms work in a new design and shape it was never intended
to be in. I don't know if it could ever be made to work or not.
The design you worked from also has internal sheet metal plates.
Your source apperently did not think to make sure that the
batterys would actuall fit before they designed them in.
I also took note that all of the J5 Aluminium body structure
materials are Metric. That alone completely messes up the
ability to fit everthing together as it needs to be. The Track
drive Boomerang plates in yours are 8mm thick when 6mm thick is
closer to what is supposed to be (0.250"). The internal width
of the Cradle base is to wide by 0.310"....Again...not your
fault. The entire Robot was designed with little tolerence and
most of the time even a small deviation from what they did means
that something won't fit together or work.
My favorite thing to say about how the original was designed is
"It works.....But its so close to not working......that it's not
funny".
The long and short of your situation and perhaps others in the
same boat is simply this, continue as you have and try to make
it work, or start all over with our designs that are not only
accurate but checked and tested to make sure they actually work.
The Input-Inc Team is truly sorry for your situation.
#Post#: 42--------------------------------------------------
Re: ERROR In My Own Design
By: Robert Snyman Date: February 2, 2013, 5:48 am
---------------------------------------------------------
Hi Terry
Thank you for getting back to me. Like I said, I am not trying
to build an exact replica of J5, but clearly my designs need
some tuning up. I in fact did not receive any information from
an out side source. What you see is what I got from viewing
photos from input-inc as you progressed just to get me started.
I Then whent on designing the parts with a few changes to fit my
own views, and everything above the waist is all Robert. My
robot`s head differs from the original by miles. That was a
personal choice I made. Yes, all dimensions are in mm because
in SA we only work in metric with designs and I went with those
sheet metal thickness just to be sure that the plates are strong
enough. Don`t worry about battery space, they do fit. I do
apologise if I upset you in any way. It is not my intention to
make a profit with my designs nor to give out any information of
my own robot that could act against your work with the original
J5. And my robot still falls under the copy right protection of
the original ;) so no worries.
There was a time when I was offered to be apart of the input-inc
team but I regrettably declined. I started a new job at that
time and did not have the time to do my own thing, but I see now
that I should have jumped at the opportunity and I probably
would not have the problem I am stuck with now.
Thanks again for trying to help me out
Robbie
#Post#: 43--------------------------------------------------
Re: ERROR In My Own Design
By: J5GURU Date: February 2, 2013, 9:31 am
---------------------------------------------------------
Hi Robbie,
Hey, no apologies nessesary dude! You have upset no
one.....least of all me :)
I had just made the incorrect assumption that someone gave you
those designs and represented them as accurate to the Original.
Keep up the good work and keep us posted of your progress! I
look forward to your version and what interesting designs and
ideas you come up with. If myself or the Team can be of any
help to you with questions you have please don't hesitate to
ask!!!
We will likely start rolling out a great majority of our J5
structural designs from the ground up to the lower Torso Waist
Motor assembly within a few weeks. That will show you exactly
how the entire mechanical system works and you can then modify
it to suit your individual build needs. It has many design
enhancments and safety features that the original robotic movie
prop did not have.
One such safety feature is a Waist Motor over rotation cutoff
system I designed and named "The Jesus Switch". It allows for
360+ degrees of Waist rotation but in case of malfunction it
kills all power to the waist Motor while at the same time
applying an "electric brake" to stop the Waist from continued
over run. I felt that this safety feature was such a critical
necessity that I designed it to be fully automatic with a built
in double redundancy. The last thing you want is for J5's entire
upper body and arms to be spinning out of control because a
single waist Motor feedback Potentiometer wire came
loose....LOL.
Our Team has also designed in some Track Tip limit "hard stop"
points into the design...again, a single Actuator limit switch
Wire becoming disconnected would be disasterous as the Track Tip
Actuators have more than enough power to destroy the Cradel Base
and mechanics if they did not stop when they are supposed to!
#Post#: 78--------------------------------------------------
Re: ERROR In My Own Design
By: alfaster78 Date: February 11, 2013, 3:35 am
---------------------------------------------------------
Hi for everyone!
Anycase the caster frames mounting holes must be inline and
vertical (to the horizon) for mechanically correct
implementation the hinge joint (i mean the hinge joint between
the base and the caster spindel) like Eric did. But this is
purely my subjective opinion. I know that on the original J5 the
Cradle base bottom plate was not parallel to the horizon
(difference of about an inch, I don't remember the exact value,
look at the videos Terry did) I think, we did the right thing by
correcting this small flaw.
Thanks!
*****************************************************
DIR Next Page