Wickie´s 6DOF Sim

hi gadtet999

At the base the cardan joints have a distance of approx. 134 cm (long leg) and approx. 10 cm from joint to joint - see pictures.
At my current temporary upper platform made of wood the triangle has the same distances.
However, I am still not 100% sure if I will make the distances of the upper platform a bit smaller.
A smaller distance means closer pivot points and therefore smaller cylider strokes. This again means quieter running.
I still think about.........
Wickie

Translated with www.DeepL.com/Translator

Hi Peter, thanks for sharing your build. The actuator design you've developed is very attractive, both for aesthetics and performance (based on the initial tests at least). I'm excited to see the results!

I have a question that is a bit late; what CAD software are you using? And are your collision calculations simply checking for interference when moving the model to its positional extremes? I've been researching and it seems that one can either figure out the geometry with a guess-and-test method or go through an intense bout of mathematical calculations to find ideal geometry

I'm not part of the second group, har har.

edit: Sorry, something else I'm unclear about is the motors because I have a hard time finding torque specs. What is the rated torque on the 190kV motors you're using?

edit2: Found the answer! For a bldc and a good estimate, Torque = (8.3 * Ia) / kV where Ia is the amperage and kV is the RPM/V. Info here: https://things-in-motion.blogspot.com/2018/12/how-to-estimate-torque-of-bldc-pmsm.html. That would imply that a 195kV 1500W motor at 80A max is making approximately 3.4Nm @ ~6000rpm, which is fantastic performance for a $60 motor! Considering a 1605 needs ~1.2Nm to drive 135kg at 6000rpm to produce 1G of force and a linear speed of 500mm/s, well that's pretty darn good and a great case for these motors (as you pointed out earlier). Sorry for the ranty edits, I'm just very excited by all of this!

Hi Fargus
nice that you like my project - I try to do my best
The CAD system that I am using is the PTC Creo 4. We also work in our company with this system, so it was the easiest way for me to use this.
I checked the positions extremes with help of my CAD-model. Only in extreme extreme positions The platform could hit one of the two front actuators. But I have to remark two things: a) when you look to successful running sims like that from SilentChill you can see, that the movement are much (!) smaller than the extreme positions. So I intend to check the real limits with my ready simulator and will limit the travel of the actuator.
b) The second thing is that although I have build a provisional upper platform this doesnt mean the platform dimensions are fixed. I still think to make the triangle smaller. I think a lot about it !
Currently I am struggling with my limit-switches - they have been too cheap and are crap. So I have to buy new ones.....

A final word about the necessary speed of the actuators: Currently I am testing with aprox. 300mm/s and this is loving fast. Personal I dont believe 500mm/s are really necessary. But we will see. The other thing is noise. Although I use high speed ballscrews, the actuators are far from silent. The faster, the louder.
regards
Wickie

Ooh, that's powerful software! I'd like to use SolidWorks instead of F360 for the same reason as you (we use SW at work) but I'm not ready to drop some cash on it just yet. Reading through your build is a huge inspiration for the actuators I wanted to build, and now that you've put me on to the ODRIVE and outrunner bldc's I'm really falling down that rabbit hole.

Great point that the sim isn't necessarily going to use the full ROM, but I guess it's good to have extra motion for runout? Or perhaps, racing games you could lock the motion to smaller range for higher speed motion cues, and unlock 95% range for slower, larger motions for flight sims. I'm thinking of doing an under-seat platform that does smaller movements but takes up less space. It's all theoretical right now!

Speaking of theoretical, have you considered optical limit switches? We have a Strasys 3D printer at work and I noticed it uses optical interrupt limit switches in some places. No moving parts, no signal debouncing. This is only conjecture, but I'm sure it's doable.

As for actuator speed, is there a way to software-limit the max linear speed? It would be nice to design the actuator with some overhead, just in case.

Hello together
time for a short status of my project.
The last time I was busy with the construction of the electronic box and wiring up the system and the motor controls. It was much more work than expected, but had some good moments ofsuccess - thats what we need.



Yesterday I was able to operate two of my actuators with the FlyPT Mover interface for the first time. It's going quite well already - but I still have a lot to learn!
What unfortunately worked very badly were my reed limit switches (see pictures). I've been searching for errors for three long evenings/nights now and found out that the reed switches are crap. So I bought 12new ones today. They work quite well now. But as it is, with every problem solved two new ones appear. It still seems that the two limit switch circuits influence each other. Sometimes an actuator just doesn't get through the calibration sequence
Next days I have to fokus more on my Christmas presents ....so a few days break with my sim

(Edit: The noise of the high quality ballscrews is not as low as expected but is ok. Seems it isnt worth to spend so much money.......I'll make up my mind when everything's ready.)
Wickie

The progress is super fun to watch. Good job hanging in there with the stupid bugs. Thanks for the honest preliminary opinion on the ball screw noise. I'll be watching for the full opinion later.

Hi all,
after some really bad evenings/ nights of endless searching for bugs I finally managed to make my 6 actuators running and connecting them with the mover interface (this was the easy part )
Finally I found out, that my new power supply is not able to handle my 6 high power BLDC motors at a higher speed. For this reason the video was made with a verry low speed and travel way of the actuators - but at least it works and I wanted to share this Moment with you mates.
This was one of the most important milestone I hoped to reach one time
Hope now the sleepless nights become less....

Regarding my power suply I ordered three car batteries that I will connect in a serial way to get the 36V.
Now I'm finally taking care of the upper platform…..

Wickie

Hi @PeterW hows about trying these PSU's

https://rover.ebay.com/rover/0/0/0?mpre=https://www.ebay.co.uk/ulk/itm/263896525393

I would stay away from batteries to be honest

Thats a good choice silentchill. I found those on my local ebay for $90AUD delivered. I might get one for 'just in case'
Another option is this link below but I dont have any experience with this one
https://www.banggood.com/3000W-AC22...-p-1530723.html?rmmds=search&cur_warehouse=CN

Whats bad with batteries? Ok, they are big and last not forever, but they can deliver hundreds of amps.
Will put them in the basement and drill two big holes for the cables direct to my living room...

Ya I'm curious about this as well, have you run into issues before?

@PeterW assuming its 3x 12V your using you may want to consider also putting 3x smaller 12V chargers with one on each battery to keep each battery healthy voltage wise. Cells run in series can end up experiencing mismatched voltages over time which results in a constant undercharge and constant over charge of at least 2 of the cells ultimately meaning much lower life span/sulfation (if lead acid) etc but if you have an independent maintenance charger on each individual cell that problem goes away.

Hi @WalkerYYJ

I thought to use the batteries as a kind of buffer for my Meanwell RSP1600W Power supyly. The Idea is somehow only halve baked and seemed to be the last hope reaction to get a suitable supply....
During my tests the last days I saw, that even a high quality1600W PSU cant manage to drive alone two BLDCs that I use with Odrive at high speed.
This was before I saw the 6000W power supply mentioned by SilentChill......
Now the money is spent and I will try the batteries. To avoid unequal loading I could use a banancer/equalizer...
Wickie

Hi all,
after some days vacation I am back to my workshop - yessss
So the last two days I started to build my upper platform. I am using stainless steel sqare pipes 45x45x2mm (aprox. 8m)
There are still some reinforcements and brackets missing, but the main shape is visible....
Now I hope my bucket-seat will arrive soon to check the dimensions and make a "test drive"






Wickie

Looking great !

I'm catching you up you best hurry hahah

Yeah, looks like a couple of 6dof are about to be finished the next months...

The race begins

You guys with your old Stewart Platforms, should be in a museum ... TIC .

Hi all,
some little updade....
Nearly finished my upper platform.
I had to make some minor changes because my bucket seat is a little different than it was in the CAD model...

Ah...and tis is my bucket seat... a Cobra Nogaro from England.
It was love from the fist moment..

Covering the woden platform floor with carpet was easier than expected, so we have been able to fix the bucket seat and the wheel today for first tests.....its sooooo comfortable .

Nice dude very nice !!!