It is currently Fri Dec 06, 2019 2:16 am




Post new topic Reply to topic  [ 12 posts ] 
 New Tricks and Developments 
Author Message

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post New Tricks and Developments
Look here for new developments and to post requests...


Fri Jan 24, 2014 1:38 am
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: New Tricks and Developments
As of 2014

1. Mantis has DMX lighting control via Open DMX.

2. Direct import of Maya .mov files with option to convert virtual to real axes for boom lift rigs. *

* This is a precursor to full virtual axes mode within Mantis.

Meanwhile an experiment to see if Mantis can run a SpiderCam style rig.



Not suggesting this will work for stop motion or timelapse, even with a stabilised head there will be some drift?

... Alternately the "cables" (strings) might be replaced with rigid linear drive rods for a more solid rig.... ?


Fri Jan 24, 2014 1:45 am
Profile

Joined: Tue Apr 09, 2013 6:49 pm
Posts: 175
Post Re: New Tricks and Developments
Hey Gerald,
3.2.5 is awesome! Quick question, I have a DMX controller from DragonFrame but Mantis doesn't recognize it. I have the FTDI drivers loaded but nada. Anyway I can get these two to play nice together? The DDMX uses standard DMX protocol I think. Any suggestions?
Also, rebuilding my jib and playing with Maya for virtual moves. Very cool!


Fri Jan 24, 2014 5:01 pm
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: New Tricks and Developments
Hi Doug.

Mantis only recognises an "Open DMX" USB controller. I have no idea what interface the DF DMX uses but it's definitely something else. Maybe if you know then pass it on.

You can buy an Open DMX controller for less than $100. The original is made by Enntec in Australia, but I've seen it sold around the globe... there are also a few clones but I've not tested them...

There is also a "Pro" version but that will not work with Mantis. I thought I'd support the basic model first, and it seems to work ok...

Preliminary testing shows that it works best in real-time with incandescent dimmers. LED light sources can have flicker problems in real time, a bit like the problems you can have with fluorescent lighting at the wrong fps. These issues are not caused by the DMX module but by the lamp dimmer, as they work at different frequencies to dim the lights. And since LED's have a very fast turn on and turn off times (like a strobe) then your camera can see every little change in the voltage.

For animation no problem - you can use just about any light source... there are lots of crazy cheap DMX Chinese RGB disco lights around for example...


Fri Jan 24, 2014 7:00 pm
Profile
User avatar

Joined: Mon Nov 22, 2010 1:18 pm
Posts: 1294
Post Re: New Tricks and Developments
Hi Gerald like the spider rig as you say using it as a camera platform my be problematic but it would make a great model mover . Got a friend who works for Stage One a big scenic company they do the Olympic stuff . Lots of winches computer controlled really amazing stuff .
Will post a link to their site you might find it interesting if not feel free to delete it
Cheers D1
http://www.stageone.co.uk/kit/innovation/qmotion


Sat Jan 25, 2014 4:12 am
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: New Tricks and Developments
Hi Yeah kind of similar - but "SpyderCam" was the original US system devised for film production. "SpiderCam" is used mainly for sports events and is German.

Both use exactly the same control program as my baby one. I mean the algorithms to solve XYZ space to the cable drives.

What they add mainly is scale. Massive winches on towers. Then they use an expensive stabilised camera rig. (On the cine rig it's a Libra) Then they need to transmit live data - SPydercam uses fibreoptics travelling along the support cables as well as power since it's for broadcast and has to last for a long events.. SpyderCam just sends wireless video assist and records hi-def on the camera which is battery operated.

Even with stabilisation there is drift so accurate moco is not really possible, though maybe for timelapse it could work with a bit of post fixing up to deal with any problems. It would open up the possibilities for some amazing sweeping timelapse shots.. !


Sun Jan 26, 2014 1:48 am
Profile

Joined: Tue Apr 09, 2013 6:49 pm
Posts: 175
Post Re: New Tricks and Developments
OK, so now that i am on my computer I can see the video. Very cool. I can think of a great drinking game you could use that for:)


Mon Jan 27, 2014 11:37 am
Profile
User avatar

Joined: Tue Nov 06, 2012 9:00 am
Posts: 60
Post Re: New Tricks and Developments
geraldft wrote:
Look here for new developments and to post requests...


Hi Gerald,

I've been working with your rig on an 8 axis setup that's using several flavors of leadshine drivers. In the current version you can set pulse width to a few separate values for all the axis at once; I found it would be more useful to be able to set the pulse width for each axis and to be able to set it to more than just 4 separate values.

The reason I mention this is because I have two different model drivers that mention a 2.5us min pulse width; it seems that one of them is happy with the Mantis supplied 2.5, whereas the other isn't. So in the current application I have to slow all the steppers down to say 3us. Being able to set, for example one axis to 2.6us may resolve this while keep all the other axes up to speed.

Thanks!


Mon Feb 03, 2014 3:58 pm
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: New Tricks and Developments
Hi

Unfortunately it's technically not possible to assign different pulse widths to different axes.

However - If your drives will all work at 3usec then this shouldn't create any real problem. This provides a maximum theoretical pulse rate of 166khz. (In fact it may go a little higher but you'll need to test to confirm.)

Meanwhile - Even a motor micro-stepped to 5000 steps/rev can do nearly 2,000 rpm at this rate, which is very fast for most stepper motors. (Little nema 17 and 23's have some chance of this speed but certainly not nema 34 motors.) If you do need higher rpm then reduce the microstepping a little to find the sweet spot. Above 3,200 you will be unlikely to see much difference in smoothness of the motors. (In fact Gecko will tell you that 2,000 steps per rev is as good as it gets... funny though - that's as high as their drives go.. hmm)

Also - if you do run a motor at more than 5,000 steps per rev - I can predict that it will lose power and will not rev as fast in any case. Any microstep rate above 10,000 is entirely mental. I'm not sure why they have these settings - maybe just because they can?

If you find anything meantime to the contrary - please let me know... :)

It might be possible to have more incremental pulse widths - but I'm not sure if this would solve your problem?

G


Wed Feb 05, 2014 1:35 am
Profile
User avatar

Joined: Tue Nov 06, 2012 9:00 am
Posts: 60
Post Re: New Tricks and Developments
geraldft wrote:
Hi

Unfortunately it's technically not possible to assign different pulse widths to different axes.

However - If your drives will all work at 3usec then this shouldn't create any real problem. This provides a maximum theoretical pulse rate of 166khz. (In fact it may go a little higher but you'll need to test to confirm.)

Meanwhile - Even a motor micro-stepped to 5000 steps/rev can do nearly 2,000 rpm at this rate, which is very fast for most stepper motors. (Little nema 17 and 23's have some chance of this speed but certainly not nema 34 motors.) If you do need higher rpm then reduce the microstepping a little to find the sweet spot. Above 3,200 you will be unlikely to see much difference in smoothness of the motors. (In fact Gecko will tell you that 2,000 steps per rev is as good as it gets... funny though - that's as high as their drives go.. hmm)

Also - if you do run a motor at more than 5,000 steps per rev - I can predict that it will lose power and will not rev as fast in any case. Any microstep rate above 10,000 is entirely mental. I'm not sure why they have these settings - maybe just because they can?

If you find anything meantime to the contrary - please let me know... :)


G


Hi Gerald,

Thanks for getting back to me. Indeed, you are right, I didn't actually run the real-world numbers to see what 3us equated to - 2000RPM at a microstep of 5000 is plenty. It's been fun working with our rig and we can't wait to get all 8 axis humming at the same time. By the way, nice jib rig!


Quote:
It might be possible to have more incremental pulse widths - but I'm not sure if this would solve your problem?


it may be nice to have them increment a bit higher. we do have a set of digital leadshine drivers that require a 10us pulse width (ugh, should have read the spec sheet a bit closer). so doing something like 2.5us, 3us, 5us, 10us, 15us might be nice. sounds like it's just a few extra lines of code? :geek:


Wed Feb 05, 2014 6:51 am
Profile

Joined: Wed Feb 15, 2012 3:36 pm
Posts: 408
Post Re: New Tricks and Developments
I don't know if this will help, but I have a mixture of Leadshine drives for my rig, the majority are rated at a Pulse Input frequency of 200khz, some of the smaller ones at 166, and the bigger ones at 600Khz

So if I run the 600Khz at 166Khz, I'm not going to get the higher revs.

Solution: use the multiplier inside the software. Change the Nominator to 2 and the Denominator to 1. Then you can run your 500Khz drive at 200 and still get the high revs.

Not sure how it works, but it seems to work!


Mind you, I'm not sure if the stepper software has this facility, certainly the servo software does.

And I measured the revs with a tachometer before and after.

Edward


Wed Feb 05, 2014 7:54 am
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: New Tricks and Developments
PhotoCs : 4 usec is the maximum possible with the Kflop board. So far I haven't found a driver that won't work at this setting so shouldn't be a problem... Inverting the pulse can also help if the drive is fussy. Minimum can go less than 1usec but no need to go there I think... BTW to change these numbers requires code inside the Kflop which means another .out file to update it when booting. So it's not just a matter of adding code to the software... I set a spread that works for all drives I have tested so far.

Edward : Yes servo drives often require a multiplier - since pulses per rev is tied to the resolution of the encoder. For stepper drives it's simpler - usually a dip switch to change the microstep ratio... note that the steps per rev rate always assumes a motor with 1.8deg/step (200 pulses per rev) . Note - Some motors can be smaller steps such as 0.9deg, then the value for the same setting is double the pulses per rev.

Boom lift rigs seem to be multiplying... Edward has a nice one and so does Doug kropla and D1. All works in progress but state of the art with this type of rig is moving on thanks to you guys.. :)


Wed Feb 05, 2014 5:56 pm
Profile
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 12 posts ] 


Who is online

Users browsing this forum: No registered users and 1 guest


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