It is currently Fri Nov 22, 2019 4:28 am




Post new topic Reply to topic  [ 99 posts ]  Go to page 1, 2, 3  Next
 Selfmade motion controller (new version) 
Author Message
User avatar

Joined: Mon Jun 13, 2011 1:25 pm
Posts: 84
Location: Frankfurt, Germany
Post Selfmade motion controller (new version)
Hello,

the last few months I have been working on the new version of my motion controller and finally it is finished :-)

Features:
- Up to 6 axis
- no rattle noise, no annoying high pitch sound, completely silent stepper movement
- integrated DC-converter to power up your DSLR
- master / slave mode for triggering
- Linear movement with fade-in/out, 3-point move, 5-point move
- prepared for future app-control (bluetooth)
- solid motor plugs
- software completely written by me

Here are some Videos with the controller:

Silent movement of 3 axis:
https://www.youtube.com/watch?v=qtmdj5kaSxY

5-point move with pan, tilt and slider (stuttering because of low amount of pictures):
https://www.youtube.com/watch?v=ydtuHdkurgI

Slider mode:
https://www.youtube.com/watch?v=bmgYlrqN84k

More videos in higher quality will come soon :-)





Image Image Image
Image Image Image

_________________
PureMoCo - Camera Motion Control Systems:
https://www.facebook.com/puremoco
http://www.puremoco.com/

Photography-Website:
http://www.izphoto.de/


Fri May 08, 2015 11:25 am
Profile
User avatar

Joined: Wed Mar 16, 2011 2:58 pm
Posts: 1348
Location: Denver, Colorado
Post Re: Selfmade motion controller (new version)
Excellent job! Looks fantastic!

You going to produce them?

_________________
http://www.BioLapse.com
http://www.TheChronosProject.com


Fri May 08, 2015 11:38 am
Profile
User avatar

Joined: Mon Jun 13, 2011 1:25 pm
Posts: 84
Location: Frankfurt, Germany
Post Re: Selfmade motion controller (new version)
Yes, the first batch will be shipped soon ;)

Made in Germany :lol:

_________________
PureMoCo - Camera Motion Control Systems:
https://www.facebook.com/puremoco
http://www.puremoco.com/

Photography-Website:
http://www.izphoto.de/


Fri May 08, 2015 11:43 am
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: Selfmade motion controller (new version)
Hi

The control unit looks very neat and could clearly also be used with users existing hardware..

But may I comment that for English speakers the modes are not so clear?

Also the web site English version doesn't translate on my browser.

I used Google translate but it's funny English then... Sorry we are not so multi-lingual as most Germans... :(


Fri May 08, 2015 8:49 pm
Profile
User avatar

Joined: Mon Jun 13, 2011 1:25 pm
Posts: 84
Location: Frankfurt, Germany
Post Re: Selfmade motion controller (new version)
Hi, the english website will be coming soon ;)

At first: Master / Slave

Master: As every other controller, you can set the interval on the controller and it will trigger your camera. (Classic shot-move-shot)

Slave: If you are using you a app like qDSLRdashboard to monitor your pictures or even change exposure automatically, you can set the camera interval directly in this app. For Canon users this is not relevant. But with Nikon-DSLR it is not possible to connect your cam to such and app and simultaneously trigger the cam via cable. The camera wont trigger.

So the app will be triggering your camera and via a hot-shoe-adapter, the cam is connected to the controller. Every time the cam is triggered, this will be noticed by the controller and the shot-move-shot process will be triggered externally.

Moving modes:

(Procedure is always the same: At first you move your axis to an start position. Then move to the next keyframe and save this position and so on. On the last point, all axis will return to the start position and the programm will start.)

Linear: Just simple constant speed movement with no specials.(start point to end point, always same step amount)

Linear with fade-in/out: Same as above, but you will have a ramp-up in speed at the beginning and a ramp-down at the end. Those are fixed values within the firmware, but can be changed.

3-Point-Move: startposition, endposition and one keyframe between. Those 3 points are connected with smooth splines

5-Point-Move: Same as 3-point-move, but with 3 keyframe between start and endposition

(Fade-in/out will be added to 3-Point-Move and 5-Move-Point in the future)

_________________
PureMoCo - Camera Motion Control Systems:
https://www.facebook.com/puremoco
http://www.puremoco.com/

Photography-Website:
http://www.izphoto.de/


Fri May 08, 2015 10:09 pm
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: Selfmade motion controller (new version)
Hi Blue gene.

Thanks that is a clear explanation.

I wonder in keyframe mode do you have problems with overshoots? In the Utube demo it looked like possibly this was the case? I understand that it's not so easy in a "blind' system to see these things or adjust them, but it is possible to have interpolation algorithms which prevent overshoot. In my own software I normally use smooth interpolation with the addition of locking the slope at any vertex. Then later I developed a more complex version which controls the overshoots automatically - at some minor cost to smoothness.

Is it also the case with keyframe moves you can't really see the result until you shoot a pass? I guess faster playback is not an option at this stage?

Gerald


Fri May 08, 2015 10:42 pm
Profile
User avatar

Joined: Mon Jun 13, 2011 1:25 pm
Posts: 84
Location: Frankfurt, Germany
Post Re: Selfmade motion controller (new version)
I did not experienced any huge problems with overshots. In my opinion, as long as the movement is smooth, everything is ok. Some months ago I did an intensive testing on the interpolation and did not noticed any problems with it.


The Splines are calculated with this library:

https://github.com/kerinin/arduino-splines

I am using the Catmull spline:
http://en.wikipedia.org/wiki/Cubic_Herm ... Rom_spline

You are right, faster playback is not an option yet, because each point is calculated in real-time. And at this point the Arduino MEGA is clearly at its performance limits. For shot-move-shot it is no problem, but continuous motion for video in fast speed is not possible. I have to do some testing here.

_________________
PureMoCo - Camera Motion Control Systems:
https://www.facebook.com/puremoco
http://www.puremoco.com/

Photography-Website:
http://www.izphoto.de/


Fri May 08, 2015 11:26 pm
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: Selfmade motion controller (new version)
HI. I use something very similar to the Catmull Rom splines.. :)

If you ever see issues with overshoot (it will happen with these splines depending on the number of keys and their spacing) then you can use a monotonic type or variation.

From wiki...

"If a cubic Hermite spline of any of the above listed types (ie. catmull rom) is used for interpolation of a monotonic data set, the interpolated function will not necessarily be monotonic, but monotonicity can be preserved by adjusting the tangents."

Below you see an example from "Mantis" of smooth interpolation when two keys are quite close together, there is much overshoot. With (modified) monotonic the overshoot is controlled... :)

Attachment:
overshoot.JPG
overshoot.JPG [ 36.41 KiB | Viewed 19881 times ]


Attachment:
monotonic.JPG
monotonic.JPG [ 33.67 KiB | Viewed 19881 times ]


Fri May 08, 2015 11:49 pm
Profile
User avatar

Joined: Mon Jun 13, 2011 1:25 pm
Posts: 84
Location: Frankfurt, Germany
Post Re: Selfmade motion controller (new version)
My splines will behave this way ;)

I have extracted the spline generateion code to a C++-command line application. So I am able to generate each point and simulate a movement.

The black dots are the keyframes. X-axis is the current picture number, and Y-axis the amount of steps.

The blue line is the spline generate by the excel diagram function.

The red one is the spline from the Arduino code.

I put the first two keyframe close together (X: 200 and 250), as you can see, there is no overshot, but a not so smooth edge. I think your program will act the same way?

(This is the 5-point-movement, the first point and last point are generated depending on the second and 4th keyframe. You might notice that its some kind of mirrored. So there will be little ramp-up and ramp-down.)


Attachments:
5.PNG
5.PNG [ 58.8 KiB | Viewed 19867 times ]

_________________
PureMoCo - Camera Motion Control Systems:
https://www.facebook.com/puremoco
http://www.puremoco.com/

Photography-Website:
http://www.izphoto.de/
Sat May 09, 2015 4:37 am
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: Selfmade motion controller (new version)
Interesting.

I'm not sure what the excel curve is for - it goes backward?

At the beginning and end it looks like there is not zero velocity? I guess for stop motion this doesn't really matter.

For the Arduino are you maybe already locking the middle keys at zero velocity maybe? How does the curve look if you have between keys which don't change direction to the next key, but still force a change of speed? Here is an example to explain. The discontinuities are not so much, but there is a smoothing function if it is a problem...

Attachment:
Example 2.JPG
Example 2.JPG [ 39.49 KiB | Viewed 19859 times ]


Sat May 09, 2015 6:39 am
Profile
User avatar

Joined: Mon Jun 13, 2011 1:25 pm
Posts: 84
Location: Frankfurt, Germany
Post Re: Selfmade motion controller (new version)
Which "beginning" and which "end" are you refering to? The beginning is always at 0 and the end in this example at 1000. The point before and after are necessery for the calculation of the spline, but are not part of the movement.

Still no overshot, but a not so smooth edge.

But i found an other problem with this example: If you look between 0 and 400 you see, that the curve is almost at 0. Some of the values are even negativ. So the axis would at first move backwards before goein forward...

This can be fixed by setting the mentioned points to the negative value, but then there would not be a small ramp-up like intended. This is only a problem, if the slope gets high quickly in the beginning.


Attachments:
9954.JPG
9954.JPG [ 56.28 KiB | Viewed 19854 times ]

_________________
PureMoCo - Camera Motion Control Systems:
https://www.facebook.com/puremoco
http://www.puremoco.com/

Photography-Website:
http://www.izphoto.de/
Sat May 09, 2015 9:15 am
Profile

Joined: Sat May 09, 2015 11:24 am
Posts: 1
Post Re: Selfmade motion controller (new version)
Are those the Trinamic TMC2100 drivers from Watterott?

What is the part below each motor driver? The 16-pin part?


Sat May 09, 2015 11:37 am
Profile
User avatar

Joined: Mon Jun 13, 2011 1:25 pm
Posts: 84
Location: Frankfurt, Germany
Post Re: Selfmade motion controller (new version)
Yes, those are the SilentStepStick.

The ICs are optocouplers. To be able to switch off the drivers/motors and set different microstepping modes, those optocouplers are unfortunately necessary. Otherwise the PCB could be a lot smaller. The setting pins have 3 possible states: GND, open and VCC. So you can't just connect them directly to the Arduino-output-pin.

Further information
https://github.com/watterott/SilentStep ... ck_v10.pdf

So it is possible to set full steps, half-steps, Quarter-steps and 1/16-steps. If you use the SilentStepStick out of the box, you have 1/16. This is not acceptable if you use a stepper with a high gear ratio since many steps are necessery and it will take really long to move your axis.

_________________
PureMoCo - Camera Motion Control Systems:
https://www.facebook.com/puremoco
http://www.puremoco.com/

Photography-Website:
http://www.izphoto.de/


Sat May 09, 2015 2:44 pm
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: Selfmade motion controller (new version)
So yes overshoot in general is well controlled, but as you note there is still a little reversing at the start.

When I referred to first and last keys, I meant in reference to the actual move - ie. time zero and end time. I didn't actually realise your first and last points are control points.

If you make the position value for the first control point to the same as the first actual keyframe will this correct the undershoot? Or maybe even you need to make it negative? At worst you need to modify the slope (tangent / velocity) at the second key? Whichever works, you need to add as a rule, to correct this problem. I recall I had to make some complex rules to correct for all cases... defining the cases is the interesting part....

BTW - re the Trinamics - did you consider using jumpers to set them up? Or do you need to change settings from the Arduino for some reason?

Gerald


Sat May 09, 2015 5:44 pm
Profile
User avatar

Joined: Mon Jun 13, 2011 1:25 pm
Posts: 84
Location: Frankfurt, Germany
Post Re: Selfmade motion controller (new version)
To change to position of the 1st control point will definitely fix this problem. As you said, different cases would be necessery. I just have to try some versions of it.

Jumpers were my first approach, but after tested with the slider, i decided that it woul be more comfortable to change the microstepping during operation. If you use a 1:27 stepper for a 1,5m slider, it will end up in about 90k-100k steps. While setup, the slider would move very slow with 1/16. So I am planing to set the microstepping for setup mode to a lower microstepping (Full step or half step) and the running mode would be 1/4 or 1/16. (just calculated/interpolated the steps by the factor). So you will have a high resolution and smoother running in run mode and a fast moving for setup.

_________________
PureMoCo - Camera Motion Control Systems:
https://www.facebook.com/puremoco
http://www.puremoco.com/

Photography-Website:
http://www.izphoto.de/


Sun May 10, 2015 12:50 am
Profile
User avatar

Joined: Wed Nov 11, 2009 10:45 am
Posts: 1685
Location: Merritt Island, Florida, Estates Unitas
Post Re: Selfmade motion controller (new version)
BlueGene, Do you know when we can buy them? Seems I'm becoming a collector of motion controllers even if I have very little time to use them. Still hoping to get back to my Mantis controller if I can get the arm project, the wildly unsuccessful Stepout Kickstarter campaign and a new Dragonframe controller with built-in drivers done anytime soon. If you can get it to do realtime preview moves, that would be a huge benefit. DF won't do it unless you pay for a DMC-16 which is beyond my budget. (Thanks Gerald for letting me buy Mantis on the layaway plan)

Thanks for the link to the silent step stick. We've been working on a realtime focus motor and have done all sorts of stuff with the gearing, larger motors and high microstepping to make them quiet. We'll definitely give the silent stick a try. Definitely post something when you've tested them with lower microstepping. I'd be interested to hear if they remain silent in those modes. We've found that full step on normal drivers is pretty unusable because the motors make a lot of noise and vibration. We set microstepping to 1/4 as the default on a lot of stuff. Motors run a lot smoother and quieter at 1/4.

Your strategy for using Nikons is interesting. Thats been a problem for most cameras that do not comply with Canon utility program.

The no-overshoot keyframing is going to be very welcome once I get mantis running, and also if I get one of these. I haven't done much with the arm yet, but I think I see some of it with DF. I add an extra keyframe as it does an abrupt direction change if I want it to stop abruptly and go back the other way. Not ideal at all. This will be much better. Speaking of the arm, eight channels would be better, but I understand your aiming for a more typical rig and adding axes most customers will never use just adds unnecessary expense.

I'm giving serious consideration to using the stock DC gearmotors on the bigger robot arms we plan to set up for realtime movement with mantis because they are a lot quieter than steppers. I doubt the silent sticks will run steppers big enough to move it. (the steppers on current arm are four amp) They have encoders with four LED + Phototransistor pairs on the back end (Fairchild motors I think). Gerald, do you think these can be plugged into the Mantis encoder inputs? Does Mantis supply current for the LEDS or would I need to make some sort of adapter box?


Mon May 11, 2015 7:00 am
Profile

Joined: Sun Oct 26, 2014 5:19 pm
Posts: 62
Post Re: Selfmade motion controller (new version)
siencelookers, I´m using them, they are awesome and cheap!!!!
you only hear the fans (at slow speeds). you can make very very slow moves without a gearsteppermotor. it interpolates to 256 steps. best stepperdriver ever worked with..

have a listen:
https://www.youtube.com/watch?v=P3ebhi-vZRY

more great test-movielinks (scroll down to Reviews and Tests):
https://github.com/watterott/SilentStepStick


Mon May 11, 2015 7:42 am
Profile

Joined: Thu Feb 14, 2013 12:24 pm
Posts: 284
Post Re: Selfmade motion controller (new version)
Wow that looks so cool. I wish I could build that stuff!

I tried making a standalone focus controller and my stepper is pretty quiet, but not quiet enough for video.
I think I ended up with 1/4 stepping but motor was moving quite slow, maybe like 60rpm max with the sketch I was using.

Not me:
https://photoscs.wordpress.com/2014/03/ ... -joystick/

I really want a 3axis + focus / zoom controller.
Even though I would rarely use all 5 axis... most shots are fine with 2 or 3 axis, but sometimes for cool shots all 5 would be neat!


Mon May 11, 2015 8:16 am
Profile
User avatar

Joined: Mon Jun 13, 2011 1:25 pm
Posts: 84
Location: Frankfurt, Germany
Post Re: Selfmade motion controller (new version)
@James:

They are ready to buy already, I just have to print the housings (Ultimaker 2);)

As already said, fast realtime movement seems to be a problem, respectively I did not focus on this point yet. I will try to get it working, but for now, this controller is intended to be used for shot-move-shot.

If you look at the youtube-videos about the SilentStepStick or the manual, you will notice, that this "silent-mode" only works with 1/4 and 1/16-steps. Also, the torque is not the same as with the standard-mode. kruppTown posted the right link, at the bottom you can find many compare-videos to the Pololu drivers.

The slave-method for Nikon is from Gunther, the man behind LRTimelapse:
http://lrtimelapse.com/tutorial/true-ho ... o-ramping/

_________________
PureMoCo - Camera Motion Control Systems:
https://www.facebook.com/puremoco
http://www.puremoco.com/

Photography-Website:
http://www.izphoto.de/


Mon May 11, 2015 9:09 am
Profile

Joined: Sun Oct 26, 2014 5:19 pm
Posts: 62
Post Re: Selfmade motion controller (new version)
bluegene, i found even in the "normal mode" the stepper is pretty silent at slow speeds.
you could use slow moco at live-interviews or wildlife ;-))

I use the "VCC_IO open Quarterstep 256 µSteps"
http://www.trinamic.com/products/integr ... mc2100-bob


Mon May 11, 2015 9:34 am
Profile
User avatar

Joined: Mon Jun 13, 2011 1:25 pm
Posts: 84
Location: Frankfurt, Germany
Post Re: Selfmade motion controller (new version)
Since I did not equip CFG2 with VCC, it is only possible to use the red marked modes. Otherwise another optocoupler would be necessary for each driver.


Attachments:
767.jpg
767.jpg [ 81.18 KiB | Viewed 19739 times ]

_________________
PureMoCo - Camera Motion Control Systems:
https://www.facebook.com/puremoco
http://www.puremoco.com/

Photography-Website:
http://www.izphoto.de/
Mon May 11, 2015 9:55 am
Profile
User avatar

Joined: Wed Mar 16, 2011 2:58 pm
Posts: 1348
Location: Denver, Colorado
Post Re: Selfmade motion controller (new version)
Neat info on the silent step sticks, i may have to bag a couple of them and try it out.

Its strange, digging through the info on this stuff i see many people making comments about how obnoxiously loud 3d printers are.

our Zortrax m200 is virtually silent. My CNC router, and my partners Taig Mill are WAY louder.

_________________
http://www.BioLapse.com
http://www.TheChronosProject.com


Mon May 11, 2015 10:07 am
Profile

Joined: Sun Oct 26, 2014 5:19 pm
Posts: 62
Post Re: Selfmade motion controller (new version)
bluegene, sorry, it´s confusing. I work with CFG1=Open and GFG2=GND --- 2(Half-step) with 256 µSteps..
just testing and soldering ;-))

btw, I dont´use any optocoupler at my chipkit with the TMC2100 and it works like a charm.

will I have some ugly moments in the next future???


Mon May 11, 2015 11:28 am
Profile
User avatar

Joined: Mon Jun 13, 2011 1:25 pm
Posts: 84
Location: Frankfurt, Germany
Post Re: Selfmade motion controller (new version)
The optocouplers are only if you want to change the microstepping modes with the microcontroller, otherwise you would not need them.

I assume that your CFG1 and CFG2 are just hard-wired?

_________________
PureMoCo - Camera Motion Control Systems:
https://www.facebook.com/puremoco
http://www.puremoco.com/

Photography-Website:
http://www.izphoto.de/


Mon May 11, 2015 11:39 am
Profile

Joined: Sun Oct 26, 2014 5:19 pm
Posts: 62
Post Re: Selfmade motion controller (new version)
yes, they are hard-wired. thanks for the infos!!


Mon May 11, 2015 11:44 am
Profile

Joined: Thu Feb 14, 2013 12:24 pm
Posts: 284
Post Re: Selfmade motion controller (new version)
Oh, actually I saw this a few months ago:

http://hackaday.com/2015/01/24/new-part ... er-motors/
https://www.youtube.com/watch?v=33jQ0P7SMJA

In the hackaday article it said "However, this stealthChop mode drastically reduces the torque a motor can provide. 3D printers throw around relatively heavy axes fairly fast when printing, and this motor driver is only supposed to be used at low or medium velocities."

Which I was concerned about for focus motor because it's already so small, reducing torque would be a problem.

How are you finding torque reduction with the driver?


Mon May 11, 2015 1:08 pm
Profile

Joined: Sun Oct 26, 2014 5:19 pm
Posts: 62
Post Re: Selfmade motion controller (new version)
I don't use it in stealth mode, it's just a great driver. have to try it and: read the manuals from trinamic - nice ideas are built within the chip-software. I like the interpolation from half steps to 256 µ steps - that's the interesting point. also the power and temperature management is very clever - you can get pretty much torque from the driver, if you don't choose the stealthmode. try it, it's less than 10 euros


Mon May 11, 2015 1:59 pm
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: Selfmade motion controller (new version)
Hi Bluegene

Thanks for sharing the info on the Trinamics. They do look very good. Internal safety features too.

The only worry really is the step input doesn't seem to be opto isolated, hence potential for damage if the chip fried and motor pwer went back to the step generator. A good mini-board would be one that had opto inputs. I guess many solutions will appear to support this chip, so I will keep an eye out.

Also I will be happy with jumpers to configure this... Changing microstep ratio on the fly is not necessary with my application because very fast clock rates are possible. Though it is not clear from the specs what is max step input rate for the TMC2100. I suspect it's high but how high?


Tue May 12, 2015 12:42 am
Profile

Joined: Thu Feb 14, 2013 12:24 pm
Posts: 284
Post Re: Selfmade motion controller (new version)
Anyone in US know if the chip on digikey is the same as the watterott one?
http://www.digikey.com/product-search/e ... A-T&v=1460

I am in NA so I'm not about to order from EU :)


Tue May 12, 2015 7:05 am
Profile
User avatar

Joined: Mon Jun 13, 2011 1:25 pm
Posts: 84
Location: Frankfurt, Germany
Post Re: Selfmade motion controller (new version)
I did a quick test with the 1/16 spreadCycle and the 1/16 stealthChop:

I could not notice any difference in torque (but in noise^^). I would even say, that the stealthShop is slightly stronger. Maybe 1/4 will behave otherwise, but is see no reason to not use stealthChop.

Maybe someone can also perform a test regarding torque of both modes?

_________________
PureMoCo - Camera Motion Control Systems:
https://www.facebook.com/puremoco
http://www.puremoco.com/

Photography-Website:
http://www.izphoto.de/


Tue May 12, 2015 10:42 am
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: Selfmade motion controller (new version)
ScotchTape - I am speculating a bit, but when they did the tests for Utube the operating voltage was fairly low. 24v via a dc - dc converter. I suspect this approach is not optimal as stepper motors don't always like working so well with some switch-mode PSU's. In any case the the motors will go faster with more voltage... it's rated up to 46vdc.

BTW - on DIgikey I can only find the surface mount chip, not the silentstick board.. it's very cheap, but you will have fun wiring it up...!

PS - Watterott quote EU 50 for shipping to Australia!


Tue May 12, 2015 4:12 pm
Profile

Joined: Sun Oct 26, 2014 5:19 pm
Posts: 62
Post Re: Selfmade motion controller (new version)
gerald, in the TMC2100_datasheet.pdf on page 10:

CFG0: SETS CHOPPER OFF TIME (DURATION OF SLOW DECAY PHASE)

this pin is not wired at the watterott-chip, but might be important for "fine tuning" - so it would make sense to wire the chip yourself..

maybe I'm wrong ;-))

link:
http://www.trinamic.com/products/integr ... er/tmc2100


Wed May 13, 2015 12:49 am
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: Selfmade motion controller (new version)
Thanks I looked. I also noticed Trinamic are now selling their own "BOB". Maybe this is slightly better? But ultimately it seems optimum to make your own motherboard like you. I'm looking at some high-speed opto chips to use for step/direction input...


Wed May 13, 2015 4:12 am
Profile
User avatar

Joined: Mon Jun 13, 2011 1:25 pm
Posts: 84
Location: Frankfurt, Germany
Post Re: Selfmade motion controller (new version)
Here is a little preview of the assembly and operating manual:

http://www.puremoco.com/downloads/docum ... ontrol.pdf

http://www.puremoco.com/downloads/docum ... manual.pdf

_________________
PureMoCo - Camera Motion Control Systems:
https://www.facebook.com/puremoco
http://www.puremoco.com/

Photography-Website:
http://www.izphoto.de/


Thu May 14, 2015 8:24 am
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: Selfmade motion controller (new version)
Hi Bluegene

I take it that your optos are only used for changing modes in the TMC? I intend to isolate the step direction inputs from my controller. For this it requires high speed optos. I also read that there are possible heat issues if I use the TMC with it's internal logic regulator. I note that your system uses a 12v motor supply so maybe this is not an issue? However I intend running a supply voltage of 36vdc and motor current of 1.5amp. This may require an external +5vdc dc-dc converter. Did you test the TMC at other voltages? Thanks Gerald


Thu May 14, 2015 4:42 pm
Profile
User avatar

Joined: Mon Jun 13, 2011 1:25 pm
Posts: 84
Location: Frankfurt, Germany
Post Re: Selfmade motion controller (new version)
Yes, the optos are only for changing state of CFG1 & CFG2 (microstepping) and CFG6 (Enable, Disable, current ramp-down). Since you have to switch between 3 states (open, GND, Vcc) there is no other solution.

The TMC2100 is only rated up to 1.2 amp (1.7 amp peak).

No, I only used 12V since it is a common voltage for power banks.

I am also using an external +5V DC-DC converter (LM2596) to generate the 5V from the 12V battery. (Even if you shut down the logic driver supply of the SilentStepStick, the stepper will be enabled in stand-by, you have to use CFG6 to disable power from the stepper)

_________________
PureMoCo - Camera Motion Control Systems:
https://www.facebook.com/puremoco
http://www.puremoco.com/

Photography-Website:
http://www.izphoto.de/


Thu May 14, 2015 10:54 pm
Profile

Joined: Sun Oct 26, 2014 5:19 pm
Posts: 62
Post Re: Selfmade motion controller (new version)
hi gerald, I'm using a "avaya psaa60w-400" (40v 1,25A).
I am afraid to go closer to the 46V.. but 40V is good, more powerful than 24v!!
put a bigger aluminium cooler on the TMC2100 and it will not stall and stay cool.
the 5v come from the chipkit.
like bluegene said, put a switch to CFG6 as an emergency shut off switch


Thu May 14, 2015 11:39 pm
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: Selfmade motion controller (new version)
Thanks all.

I had a reply from Trinamic and they said it's ok to use the internal 5v regulator so long as use is "intermittent" and a decent heatsink is applied...

Meantime I do have a 48v - 5v converter if I feel the need... Still waiting for my TMC but looking forward to trying it... :-)


Fri May 15, 2015 1:23 am
Profile

Joined: Sun Oct 26, 2014 5:19 pm
Posts: 62
Post Re: Selfmade motion controller (new version)
very good!!! please, gerald, make a test with 46v (I think you have it in your laboratory) , maybe then a higher speed is possible!!?? :-))))


Fri May 15, 2015 1:42 am
Profile
User avatar

Joined: Mon Jun 13, 2011 1:25 pm
Posts: 84
Location: Frankfurt, Germany
Post Re: Selfmade motion controller (new version)
I did a quick test with my pan-tilt-head to demonstrate the different motion movement modes:

Linear with fade:
https://www.youtube.com/watch?v=Rw_xNGC4EEU

3-point-movement:
https://www.youtube.com/watch?v=i7SjDwMl4O0


Futher videos will be added soon ;)

_________________
PureMoCo - Camera Motion Control Systems:
https://www.facebook.com/puremoco
http://www.puremoco.com/

Photography-Website:
http://www.izphoto.de/


Sun May 17, 2015 9:47 am
Profile
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 99 posts ]  Go to page 1, 2, 3  Next


Who is online

Users browsing this forum: No registered users and 5 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