It is currently Sat Nov 16, 2019 8:26 pm




Post new topic Reply to topic  [ 8 posts ] 
 Project Otto Testing. 
Author Message
User avatar

Joined: Tue Aug 09, 2016 10:31 am
Posts: 43
Post Project Otto Testing.
Well this has been a hell of a lot of work. Almost done, not quite out of the woods yet. I put another 20 hours in this weekend, i can probably do another 20.

The BCM has been integrated with the DF DMC-16. All steppers are wired up. All of them run under 60c except the Nema 14 used on the 3d slider. That little motor just needs very little power, i may have to log into the DM432c driver to get the power down low enough.

I spend some time testing the rig and by in large I have been very pleased.

Here are some results of motion testing. This was the first test


And of the camera running this test


I was pretty pleased by it. there is quite a bit of motion going on there.
Yesterday I ran the following, there was quite a bit less movement and some problems start to show up.


The horizontal has some pretty jumpy movement in it. I did manage to determine this was due to the rig, not the render.
The brackets that hold the stepper and ballscrew are not the best. There is a tiny bit of flex. With larger moves this does not seem to be an issue as seen in the prior video. However in this one the horizontal move jumps. I think the tiny movements are being held in the brackets for the motor and ballscrew like a spring from time to time, and it make take a couple moves before the elevator assembly finally moves. I had planned to replace those brackets, i am hoping the new ones alleviate this issue.

Here is a boring test video i too. that is standard graph paper, you can see the motion. There is tiny amount of this jumping on the Y direction, on the X there is quite a bit more.


I am looking into ways to brace the brackets on the Y axis, replace the current X brackets with some good solid 1/2 inch aluminum brackets that ill fabricate.

Open to any considerations, thoughts, critiques, and speculations on this movement and what it means.

here are some images of the X Axis brackets. (yes, i know, they need to be replaced. hahahaha
Image

Versus the aluminum brackets on the Y axis.
These are far more rigid than the flimsy X axis and I see less of the motion issues on Y, even though it moves the entire X assembly plus the elevator.
Image


Mon Nov 07, 2016 10:11 am
Profile
User avatar

Joined: Tue Nov 24, 2009 7:55 am
Posts: 951
Location: UK
Post Re: Project Otto Testing.
Hi, Chris, that first video clip looks very fluid, I see on your Facebook post you have an eMotimo rig set up, have you considered driving it via Dragonframe to give more flexibility of movement. I re-wired mine to a couple of external drivers and controlled it via Mantis2.

https://www.vimeo.com/86247767

_________________
Smile, it gives your face something to do!

http://www.vimeo.com/user1214873/videos
http://www.youtube.com/user/MrReggub


Mon Nov 07, 2016 1:20 pm
Profile
User avatar

Joined: Tue Aug 09, 2016 10:31 am
Posts: 43
Post Re: Project Otto Testing.
The first vid worked out pretty good, it is the 3rd one I'm concerned with, and the testing on the 4th.

these were all show with the new 8 "axis" rig that i built. I do have the emotimo and have connected it via USB and DF Slave mode to my laptop running dragonframe. No issues on that setup. :)


Mon Nov 07, 2016 1:25 pm
Profile
User avatar

Joined: Mon Nov 22, 2010 1:18 pm
Posts: 1294
Post Re: Project Otto Testing.
Hi Chris noticed you on line . Do i remember you saying you had some leadshine drivers and some 556 have you noticed any difference . I am a bit disappointed with my leadshines
D1


Mon Nov 07, 2016 2:15 pm
Profile
User avatar

Joined: Tue Aug 09, 2016 10:31 am
Posts: 43
Post Re: Project Otto Testing.
Either im not doing the leadshines right, or they flat out suck.

Example,
Leadshine set at 3.20A(Peak) 2.29A (rms)
CWD556 set at 3.2A(peak) 2.3A(rms)

The CWD outperform the Leadshines hands down every single time. I don't have specific numbers but when set the same from one to the other the CWDs just blow them out of the water with the resulting torque and high speed. The Leadshines stall out at maybe 1/3rd the speed I can get with the CWD's.

Now I admit, i am NOT logging into the Leadshines and tweaking the parameters. That is really outside my knowledge level though. Perhaps if i was stronger in my knowledge with them i could coax them to get better performance, but as it is, i removed 6 leadshines and replaced them with the 556. The 432's now provide the focus and mini 3d slider support, and 2 extra as spare channels to run a rotary table and the rail for the 4 axis setup.

The only place where i see any benefit to the leadshines? The motors are dead quiet with the leadshines VS pretty noisy with the CWD556's

I did read another thread where this was discussed, and it seems other people have had better results than I have. But my own experience seemed to echo yours.


Mon Nov 07, 2016 2:28 pm
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: Project Otto Testing.
Hi Bio

My conclusion after testing is the 432c is mainly suited for small motors when smooth performance is a priority. I have used them successfully for lens motors and small pan tilt heads.

For bigger motors moving larger chunks of metal, you should use the DM556 or DM856 - which will perform similarly to your clone drive.

Logically two drives set to the same current and with the same supply voltage, should perform the similarly. In practice this is not always the case. Perhaps the reason is more headroom in the larger drive? Or maybe the current settings are actually not the same? We can't be certain. BTW you didn't mention the running voltage for these drives? Motor power is proportional to supply voltage as well as current.

But that aside, it looks like vibration is your limiting factor? That isn't surprising given how long that dropper arm is. I suspect even if you stiffen the brackets, there is likely to be some vibration inherent in the structure. Meantime I'd also shorten the top bracket - try and keep the drive screw as close as possible to the driven part - otherwise it will be transmitting a twisting force as well as bending the over-long brackets.

Edward will likely comment here - he has spent a large amount of time perfecting the art of smooth macro shots... from using sophisticated servo motors to experimenting with gearboxes and drive systems. It's not something for short pockets.. :)

But perhaps you are mainly using real time to assist programming, and can shoot the actual moves in stop-motion?

Alternately, you could use reduced constant frame rates such as 1-3 fps, and thereby retain motion blur.


Mon Nov 07, 2016 4:55 pm
Profile
User avatar

Joined: Mon Nov 22, 2010 1:18 pm
Posts: 1294
Post Re: Project Otto Testing.
Hey Chris forgot to say your shots look great when their transferd to 3d thats going to be quite a epic flight . Planet Earth Two is just airing over here if ya get a chance to see it theirs some amazing stuff in it . Its what the BBC do well nobody compares i dont mind paying my licenses fee for it .
On the driver front think we have both reached the same conclusion .When you say 556 is it these or did you find another supplier https://www.cnc4you.co.uk/Digital-DSP-S ... olt-CWD556 But what ever the case they just easy to use and they will run tiny 17s to Nema 23s no problem .
I currently have a leadshine connected to a Nema 23 it just about works but doesn't half get hot . On the other side i can happily leave a 556 for days and the motors hardly get warm . My rigs do get left on for days at a time so i want to feel confident i aint going to burn the place down .

D1


Tue Nov 08, 2016 2:03 am
Profile

Joined: Wed Feb 15, 2012 3:36 pm
Posts: 408
Post Re: Project Otto Testing.
Hi Bio,

My main interest is in macro moves, similar to what you are doing. I have done a lot of testing and I'm still not quite there, although getting there slowly.

The best thing you can ever do is test each axis in isolation, this way you can identify the culprit. Even when movement seems perfectly smooth to the eye, as soon as you film at close range, the most minute vibration is augmented and annoying. I prefer to start this systematic detective action from the pan/ tilt, then, once you are sure that pan/tilt and focus are super smooth, move on to the other bigger axes, which are likely to give you the most problems.

For pan and tilt, you have to ask yourself, do you need all that speed? 4secs per 360 for macro? The answer is surely not! You will find it extremely difficult to build a machine that does it all, the fast moves needed for slow motion, and the super subtle moves for macro. I am not saying it's impossible, but sometimes we set ourselves goals that are practically impossible to reach, particularly on a budget.

In my case, my first pan/tilt was the most successful right from the start, even using the same little Leadshines. But I had a 100:1 reduction and the motors were running fast, making a hell of a din. I used Harmonic drives for that. This reduction and faster motor speed eliminated the typical vibration zones of a drive, like you have in the video.

Then I went for servos, a complicated and expensive solution but very good in terms of smoothness. I used 50:1 harmonic drives. This is my heavy duty head that weighs 5Kg. No vibrations there and it can do at least 2secs per full turn, and probably faster, but that's the realm of the speed maniacs. :lol:

Lately I've gone for lighter heads, as I've realised that the less weight you have at the very front, the better. So I have built a new prototype, this time using the wonderful SilentStepStics and Nema 14 (and even Nema 11) and a 50:1 HD reduction. It's not fast, but plenty for macro. And this drive is very smooth and silent, but has little torque at its smoothest setting, and it may be too slow with your particular controller. It's pretty fast with Mantis though. I expect this head is going to be around 1.8Kg. Currently it is 2.1Kg, but this is with Nema 17 motors, which I will be changing as soon as I can move to my new workshop.

For the long axes, in my case the 2m track I use AT5 belt and Delta Servo drive and motor reduced 25:1 with a precision planetary gearbox. The problem is the actual track, Hiwin type rail seems to transfer the little vibrations of the ball bearings inside the blocks all the way to the camera. Therefore I am next going to try good old plastic wheels on round track, as I think it will be the smoothest solution. I think that round track and bearings like you have is a little coarser than Hiwins for macro stuff.

Edward


Tue Nov 08, 2016 3:14 am
Profile
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 8 posts ] 


Who is online

Users browsing this forum: Majestic-12 [Bot] and 9 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB © phpBB Group.
Designed by Vjacheslav Trushkin for Free Forums/DivisionCore. pozycjonowanie