It is currently Thu Dec 12, 2019 9:35 am




Post new topic Reply to topic  [ 43 posts ]  Go to page 1, 2  Next
 Project Otto. Z axis problem... 
Author Message
User avatar

Joined: Tue Aug 09, 2016 10:31 am
Posts: 43
Post Project Otto. Z axis problem...
Hey folks.

So I'm building a studio rig for my botanical time lapse.
I am running into something I could use some advice on.

My current vision is basically an overhead CNC machine. XYZ, Pan, Tilt, Roll, Focus.
A shooting envelope of X = 4.5ft, Y = 4.5ft, Z = 1.5ft. (roughly)
I want something where I can drop the camera anywhere in the set.

The overall frame would be 80/20 1020 series (not yet purchased)
X and Y will run on 1500mm ballscrews and bearings (not 80/20 bearings, but actual bearings)
P/T/R will be done with a Servocity PT-2100 with Roll attachment. (managed to get one with no motors, no electronics) and using geared steppers. Possibly harmonic drive gearing if i can find some for the right price.
Focus is controlled by a Lens Apparatus.

I already have the Camera, fresh laptop with Dragonframe, a Dragonframe DMC-16, Parts to build the power supply, ten DM432 leadshine drivers, PT-2100 w/roll, stepper motors, cables, connectors, and some experience in building these sorts of things.

There is one thing I keep banging my head on, and that is the Z axis. I am hoping for about 18 inches of travel on Z.

The studio has a low 8 foot ceiling. This is a bit cramped, but it is what i have to work with.

Getting 18 inches of Z travel with that PT-2100 using a ballscrew type setup is going to be a bit challenging. I could always lower the shooting stage, in order to help make room but I would prefer to keep things at a more comfortable height for working on the set. I estimate that using a ballscrew setup would mean the camera is already 3 feet from the ceiling at its top when I include the PT-2100. that's not terrible, but i would like to be able to get it closer to 2 feet.

Before I go ordering thousands of dollars of 80/20, ballscrews, bearings, etc, i want to have this problem figured out.

I am exploring the possibilities of a high precision scissor jack style setup. Does anybody have any leads on something like that? Or any other possible telescoping setups that will possibly work?


Tue Aug 09, 2016 11:01 am
Profile
User avatar

Joined: Wed Nov 11, 2009 10:45 am
Posts: 1685
Location: Merritt Island, Florida, Estates Unitas
Post Re: Project Otto. Z axis problem...
One way might be to stack two short dollies so they can extend twice their folded length. A more elaborate way would involve square extrusions with wheels between, like the telescoping antenna boom on News trucks or truck mounted mobile construction cranes. Take a good, close look at one of those for some ideas. Another possibility would be a robot arm with one or more elbows. This is essentially one half of your scissors jack mechanism. The entire arm can fold at the wrist, elbow, and shoulder so the straight parts between the joints wind up horizontal near the top of its travel and vertical when it is fully extended. This would let you have as much Z as needed without having to extend any of it upwards when you want to raise the camera up.

If you look up the tiny robot arm I showed on the forum a while back, The bottom of its crate that holds the dolly has threaded inserts which let it be suspended from 80/20 T-slot overhead. I gave the dolly wheels above and below the track so it could be suspended overhead and run upside down. It has total freedom of movement in all three dimensions and when it raises the camera up, nothing goes above the dolly. Hope that makes sense and is of help. I still have a few robot arm frames that I could put on ebay pretty cheap, They have the 18 inches of Z you want and maybe a little more. If you replace the claw with a camera, you get a pan-like effect where the base rotates and another pan and tilt at the wrist, all from gearing that comes with the arm. The wrist mechanism fits within the end of the arm and is way smaller that your PT-2100. Let me know if you want one. If you really like them, I'll trade you two robot arm frames with belts, timing pulleys, all the stuff that came with them for your PT2100, or can list one or both of them on Ebay for $500 each. They are the orange robots which are slightly larger and much stronger than the little one in my robot rig. Basically you will be replacing the servos with steppers, and replacing the claw with a quick release or camera mounting plate. I also have one arm exactly like the one I modified if you'd like it instead. I'm on expedition until Aug 21st, but can do it as soon as I get back if you want.

Here is the video of it. Its very easy to get used to the unusual way the wrist moves. The pan and tilt interact when you are moving one motor at a time to program it, but they both move together when playing back the keyframed sequence in Dragonframe so perfectly level pans are easy. This video was made before the drivers were tuned, so the nasty noise when the dolly moves doesn't happen now. As with the pan and tilt at the end of the arm, coordinating movement of the shoulder, elbow and wrist lets the camera move straight up or down. Of course, moving in any way within its work envelope is possible. It sounds like this total freedom of movement in all three dimensions is what you are looking for. When suspended upside down above your stage, the arm and dolly could work instead of the whole X and Y gantry crane mechanism you are planning.



https://www.youtube.com/watch?v=xg4ZR3wpA9Q

Definitely post some pictures or video of your rig. Some of us like to see machines being built.


Sun Aug 14, 2016 8:56 am
Profile
User avatar

Joined: Tue Aug 09, 2016 10:31 am
Posts: 43
Post Re: Project Otto. Z axis problem...
Holy crap James, Color me impressed.

Let me look into it a bit. Its been a long day and I need to crash here soon but I am very intrigued.

I managed to get the driver / power box. I will post it on my blog tomorrow but here are a few shots. Sorry if they are gigantic.

So I picked up a server chassis to house all this in. I have 10 dm432c leadshine drivers with a pair of 36v 16a power supplies.
The back was a challange, I used aluminum stock to make a panel for all the 4 pin stepper connectors. A pair of 6x Rj45 jack plastic panels. I would like to have done those in aluminum too but these are going to work fine.

Wiring it was a bitch. took about 4 hours or so. few hours of drilling/tapping, bolting, soldering, cutting.

But it works! Like a champ! all 10 drivers are working and can run a motor.

I need to do some research on stepper selection. Im used to dealing with low powered ones that can be run off 12v and need to match something to these drivers a bit better than what I have one hand.

Image

Image

Image


Sun Aug 14, 2016 8:46 pm
Profile
User avatar

Joined: Mon Nov 22, 2010 1:18 pm
Posts: 1294
Post Re: Project Otto. Z axis problem...
Thats very neat see you have a DMC quite a investment . I enjoyed playing with the one i had unfortunately its gone to a customer now . Lets us know how it goes i didnt have any probs worked a treat very easy to use .
You not a Jack Ripper re branded by any chance ?
D1


Sun Aug 14, 2016 11:45 pm
Profile
User avatar

Joined: Tue Aug 09, 2016 10:31 am
Posts: 43
Post Re: Project Otto. Z axis problem...
Hey D1. Nice to see you.

That's me. The one and the same.
I couldn't remember the password for the other account, i have not been on in forever. I spent the past year filming for a documentary that is coming out next year and that had dominated my time. Now that I am paid up, I felt it was time to take the studio stuff to the next level. I have a running blog that nobody cares about at biolapse.com about all this stuff. I do plan to start posting here more often.

I really like the DMC-16. I am very excited about the DMX lighting controls I have some pretty interesting ideas i want to work on.

I got the 8020 framing ordered up today, ball screws, bearings. etc.

Oh and check this out. My buddy Kyle printed out a bracket for me to mount the lens apparatus direct to the lens rather than on a 15mm rail system to the camera to the lens mount, all of which can introduce a bit of play between the lens and the focus mechanism.

Image
Image


Mon Aug 15, 2016 9:54 am
Profile
User avatar

Joined: Tue Nov 24, 2009 7:55 am
Posts: 951
Location: UK
Post Re: Project Otto. Z axis problem...
Welcome back, used to enjoy your posts and especially your last time-lapse from the cellar.
An interesting bracket, does it load the lens mount excessively ? I guess only testing will tell.

D1, a bit more machining complete, part still sitting on the mill due school holiday duties.

_________________
Smile, it gives your face something to do!

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


Mon Aug 15, 2016 10:53 am
Profile
User avatar

Joined: Mon Nov 22, 2010 1:18 pm
Posts: 1294
Post Re: Project Otto. Z axis problem...
Ah welcome back Chris . Think you are going the right way with your approach to biolapse def the way i would go. Plant timelaspe prob one of the last interesting bastions of timelapse and it all self contained . Hey never need to leave the shed :D
On lens motors for last rig i made the client wanted a traditional gear rig . My first time not that impressed will stick to my belts def seemed to introduce more initial kick . This would be very noticeable but we discussed this before and some lens are worse than others.
Hi Mike dont worry it will be winter soon and you will be able potter to your harts content .
D1


Mon Aug 15, 2016 1:15 pm
Profile
User avatar

Joined: Tue Aug 09, 2016 10:31 am
Posts: 43
Post Re: Project Otto. Z axis problem...
Hey Mike!

Weight isn't bad. The bracket is printed and pretty light weight. The motor is the heaviest part of the assembly.


Tue Aug 16, 2016 7:01 am
Profile

Joined: Tue Apr 09, 2013 6:49 pm
Posts: 175
Post Re: Project Otto. Z axis problem...
Cool mount Jack. 3D printing is another piece of tech I would like to own.


Tue Aug 16, 2016 7:46 am
Profile
User avatar

Joined: Thu Mar 10, 2011 11:38 pm
Posts: 585
Location: Exmouth, Western Australia
Post Re: Project Otto. Z axis problem...
Hi Chris,
Great to see you back on Timescapes. I always thought you were the guy who really took this bull by the horns, going from 'Speedy' to specialising in world-class biolapse work via Project Chronos and all the work you did on astrophotography. That thread is still on here and I've pointed a few people at it over the last couple of years. And I do remember to drop in on your biolapse blog from time to time!

I can't offer much first hand info to help out with your current question other than to agree with James and point out how much stuff has been talked about industrial robot arms and DIY versions of same on some of the everlasting threads that have been running here for what seems like forever :D Just make yourself a VERY large cup of tea ( coffee, bourbon, whatever) and dive into the archive (especially any thread started by DISPLACEMENT1.) As you know, Timescapes always was a very technically orientated forum but the bar has been lifted a long way since you were last a regular contributor.

Regards

Kit


Wed Aug 17, 2016 4:29 am
Profile
User avatar

Joined: Tue Aug 09, 2016 10:31 am
Posts: 43
Post Re: Project Otto. Z axis problem...
Hey Kitwin!

Sounds like im back in the correct place. :)

The Chronos Project is pretty much over. It was supposed to be an online recipe for the DIY'ers. Neither Kyle or I were prepared for where it headed. It was a lot of fun for a while, but its a ton of work. I'm a single dad and I found I was spending more time building robots than hanging with my son. I already have 20 years in a well paying career that I enjoy and there was no chance of making a full time transition. Plus we were just building the same things over and over, and i like to build new things. Eventually we decided to stop broadcasting and let things sort of die out.

I learned a lot though for sure. In no small part to all of you folks on here. That arm that James put together is just awesome. While I have a pretty handsome budget for building this system, I am hesitant to make any big changes to my plans. Changes always cost money.

This new project is a lot of fun so far. I had been filming for the last year. I almost forgot how much I love building things. I am nowhere near the skill level as most of you guys, but i sure do a decent job of faking it. :)

To demonstrate that last point about how little of this I actually understand...

So i went from the a4988 allegro drivers using 12v DC 1a with a software program that was fairly rudimentary as far as stepper control, to a Laptop with Dragonframe, the DMC-16 controller, Leadshine dm432C drivers, with 36V 16a power supply.

Why the hell do my steppers still stall out at the same speed? I have this unknown i am trying to figure out. The relation between steppers, drivers, and how to properly mate and match them. But I'm having a hard time working that out. I have seen people get stepper motors just screaming on youtube videos. I did try various microstepping values and various amperage values with the dipswitches.

My new concern is my ignorance is preventing me from getting the most from the DMC-16. is the 200,000 pulse per second advantage more of a "keeping the speed while microstepping" thing?

I dont really need super high speeds, but I was hoping for some realtime speeds. From my initial tests I am NOWHERE near there. I have a CNC machine from Probotics that can moves its steppers WAY faster.

Any tips?


Wed Aug 17, 2016 10:52 am
Profile
User avatar

Joined: Mon Nov 22, 2010 1:18 pm
Posts: 1294
Post Re: Project Otto. Z axis problem...
Hi Chris their as been a discussion on the speed from various drivers and to be honest i gave up on the leadshines . I had the same prob as you so i have five in their boxes i just aint going to use . I found they just stalled out at quite low speeds . But their again other people say their good but my replacements are deff a lot faster than anything i could get out of the leadshines .
https://youtu.be/MI9t0XaxMa8
Stick with it till end
As you say its a shame to waste the plus rate of the DMC . Have you had any look with the limit switches on the DMC i just couldnt get them to work and i dont know why :(
D1

You tube doesn't seem to be working today


Wed Aug 17, 2016 12:57 pm
Profile
User avatar

Joined: Tue Aug 09, 2016 10:31 am
Posts: 43
Post Re: Project Otto. Z axis problem...
Thats interesting to know D1. I had not realized it. I will spend some time digging in the archives to educate myself on them.

I do have use for several axis that will likely never hit very high speeds, so I can probably still make good use of most of the drivers, and have a few spares, but there are some that i will be replacing if i can get some better speed.

I have to day though the dm432c's are damn quiet :)

I have not tried the limit switches yet. I did manage to get my biolapse control module to trigger the DMC, so all my enviornmental and lighting controls are all in sync, that was fairly easy. Maybe ill spend some time this weekend playing around with the limit switches and let you know. I do plan to use them.

EDIT: Those are some gnarley speeds! Which drivers did you end up working with? i dont see any info in the video


Wed Aug 17, 2016 2:02 pm
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: Project Otto. Z axis problem...
Hi.

Without knowing more detail about the actual intent - have you thought about a Foba stand? It would greatly simplify your build and is very flexible. I have used them for animation work quite a lot . It's relatively straight forward to motorise the stand using belt drives. The lift is counterbalanced inside the column, so the vertical axis doesn't need too much power.

They are expensive new, but if you can pick one up second hand it is worth it...

The picture below shows one horizontal and one vertical axis. For an additional horizontal axis attach an industrial slider to the end of cross arm. The stand is very sturdy and moves around on a wheeled base. You can position the camera from floor level to over 2 meters depending on the column height, and if the head is under or overslung. Both motors in this example have planetary gearboxes - the lift is nema 34 while the cross slide is nema 23.

Attachment:
FobaSmall.JPG
FobaSmall.JPG [ 99.83 KiB | Viewed 15048 times ]


Re the Leadshine drives - they should work just fine for your pan tilt focus axes, and maybe for a small linear axis - but are a bit under-powered for bigger axes. I have found they work best with nema 17 motors. For other motors you may need to use a higher current / voltage DM series drives - you'll get the same smoothness with more grunt.

In general motor speed is improved by using a higher voltage supply, though the maximum you can use is obviously limited by the drive specification. Basically any motor will have more torque at higher rpm if the supply voltage is increased, while low speed torque remains constant. Some of the demo's you've seen may have been using 80vdc supplies or even more. I have seen a stepper do 3,000 rpm on my bench, though torque was not so stellar...

[edit] In response to your query about microstepping and pulse rate - another known property of stepper motors is that torque decreases with increasing micro-step rates. Typically you will find a good compromise of smoothness and power between 10x and 20x. (for this reason Gecko drives only have 10x microstepping) So most likely you will never need your 200khz unless you graduate to servo motors some day... :)


Thu Aug 18, 2016 12:00 am
Profile
User avatar

Joined: Tue Nov 24, 2009 7:55 am
Posts: 951
Location: UK
Post Re: Project Otto. Z axis problem...
Hi, Chris, Servocity pan, tilt and roll I modified to using stepper motors. my gearing is 600:1 so no high speed moves ;) More detail in the Vimeo description.


_________________
Smile, it gives your face something to do!

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


Thu Aug 18, 2016 1:58 am
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: Project Otto. Z axis problem...
Hi MIke I'm curious - why 600:1 ? Are you using it for astro-photography?


Thu Aug 18, 2016 4:02 am
Profile
User avatar

Joined: Tue Nov 24, 2009 7:55 am
Posts: 951
Location: UK
Post Re: Project Otto. Z axis problem...
Hi, Gerald, just a slow place of life here in the Forest ;) Original pulley ratio was 6:1 and I had 100:1 gearboxes at a reasonable price, running X10 stepping currently. Works fine for stop motion, macro work and even at "top speed" there are no jello effects. I might get a 5:1 gearboxes sometime when I get my global shutter camera system (Apertus Axiom), hopefully later in the year.

_________________
Smile, it gives your face something to do!

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


Thu Aug 18, 2016 4:35 am
Profile
User avatar

Joined: Mon Nov 22, 2010 1:18 pm
Posts: 1294
Post Re: Project Otto. Z axis problem...
Hi Chris this thread contains a lot of discussions on drivers viewtopic.php?f=48&t=13181
Like you i noticed my mill motors could go a lot faster then those driven by the leadshines so i just decided to by more of those driver https://www.cnc4you.co.uk/Microstepping ... olt-CWD556 They work very well with nema 17 and 23 dont get hot and run smooth . Apparently they are their own design but made in China as most things are .
By way as anybody noticed the Dragonframe forum seems to have disappeared :(
As for microstepping i have ventured up to 32 times but that was with the DMC you do need a high pules rate . No noticeable loss of torque with the 556s
D1


Thu Aug 18, 2016 12:45 pm
Profile
User avatar

Joined: Tue Aug 09, 2016 10:31 am
Posts: 43
Post Re: Project Otto. Z axis problem...
Geraldft,

That is a pretty slick looking system. I had actually never heard of that type of lift. Ill do some research into it, but Yikes, price is crazy!

MikeA,
Boy that looks familiar :) 600:1 is a heck of a gear reduction :) I picked up three 50:1 steppers. Unfortunately the 8mm shaft is big for the gears they provided, i need to pick up an 8mm bit and bore them out before Ill have mine running.

D1,
Thanks for the lead on the 556's. I ordered one up to play around with. Ill spend some time in that thread as well, i appreciate the link! :)

I also ordered a very short Z axis assembly from ebay. It will fit nicely in the cradle of the tilt on this rig. It has 4 inches of travel. I really want to experiment around with the 3D function of Dragonframe :)


Thu Aug 18, 2016 2:17 pm
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: Project Otto. Z axis problem...
Bio. If you can't find a Foba for the right price maybe you could make something similar? It's basically a long vertical rail with a cross bar. My impression was you wanted something that would sit on a scissor lift to gain access higher up - whereas this could do it all in one. Or maybe integrate it with your present idea in some way? In older days similar rigs were called "crucifix" rigs and also rode on a dolly base. MRMC Animoko is designed on similar lines. If your studio setup was permanent you could also consider a gantry rig?

Mike - I understand sometimes you make things out of what you have. Still, you could use it for some star tracking.. :)


Thu Aug 18, 2016 3:18 pm
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: Project Otto. Z axis problem...
Bio - if you are interested in stepper theory you can read some very detailed information here..

http://www.geckodrive.com/support/step-motor-basics.html

An important point to consider is the effect of drive voltage vs torque.

Attachment:
TorqueVsVoltage.JPG
TorqueVsVoltage.JPG [ 30.24 KiB | Viewed 15036 times ]


In basic terms the voltage effects the rate at which current can be applied. At low speed the current can easily reach the current ceiling, which you set by the dip switches on the drive. As the speed increases there is less time for the current to rise so it becomes more limited and torque decreases. Using a higher voltage allows more current to flow at higher rpm. Unloaded the motor may still reach the same speed at either voltage, but under load will stall more easily.

Gear ratio is also important. A lower ratio can often give more top speed since you are working the motor in the range where it has more torque. The trade-off is that at low speed the motor steps might become visible as a vibration at the camera. As an example, in the case of your leadscrews, you will gain more speed by using a 2cm pitch rather than than 1cm, since to achieve the same speed you only need half the rpm.

Regarding microstepping rate. Manufacturers commonly show data for motors at full steps. The reason for this is standardisation of data but also because they know the numbers look better. In practice we know that a stepper motor running at full steps is pretty ugly, so microstepping is necessary. Yet above a certain rate it has no tangible benefit, so you should make your own tests to determine the optimum. How high you can go without losing power will also depend on the quality of the motor and how much load the motor is driving. Gecko mention "linearity" which can become a significant issue at higher microstep rates.

My own tests have shown that power does fall off with increased microstep rates depending on the specific motor. For example I have some Oriental motors which respond better at higher microstep rates than the common or garden chinese variety. Yet even then I can't really detect any advantage in the smoothness above 16x. In fact resonance effects become more significant, which is where the Leadshines help more due to the anti-resonance circuit. In the case of the DM432 this anti-resonance may actually rob some power which may explain D1's issues. Oddly the larger DM556 and DM856 drives do provide more power than the DM432 at higher rpm - this may be due to slightly different anti-resonance circuits or because they have different output transistors which are more efficient.

I hope this is not too much information - but I find some basic understanding of theory helps when searching for that elusive speed. :)


Thu Aug 18, 2016 4:48 pm
Profile
User avatar

Joined: Tue Nov 24, 2009 7:55 am
Posts: 951
Location: UK
Post Re: Project Otto. Z axis problem...
Hi, Gerald could use for star track but have one of these with a Gemini controller: http://www.orionoptics.co.uk/MOUNTS/losmandygm8.html

D1, DF forum has been missing for at least a couple of weeks maybe more, pity because there was some interesting info on it.

_________________
Smile, it gives your face something to do!

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


Fri Aug 19, 2016 12:36 am
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: Project Otto. Z axis problem...
Re DF forum. I guess there was a lot of redundant info there and use of the forum seemed to have dropped off...

Meantime I should add some remarks in reference to my previous mega stepper drive post.

In summary, there are many options for drives. I mainly refer to Leadshine because I am very familiar with them.

In cost they sit at the lower end of industrial quality drives. The DM432 is their smallest and lowest spec model in the DM series, which have anti resonance circuits. For bigger motors such as nema 23 and 34 this smoothing is quite important, since these motors suffer from more vibration than smaller motors.

For higher speeds you should choose a drive which can run from a higher voltage supply. That will allow better performance from any drive, including the ones D1 refers to, and other similar models from Chinese manufacturers. Geckodrives are also a good option since they have an 80vdc max current. If you have the cash then Oriental Motor are probably the best.

On my rigs I use mainly the DM856 and the higher current EM series. I also use EasyServos for driving more demanding axes.


Fri Aug 19, 2016 4:59 pm
Profile
User avatar

Joined: Mon Nov 22, 2010 1:18 pm
Posts: 1294
Post Re: Project Otto. Z axis problem...
Think the DF forum suffered badly from spam . But it is a shame thats its gone their was some useful information on their and i was the first post which i was always proud of :D
Hope Timescapes doesn't give up one day it would be a waste of a huge resource and what would we do with our spare time . I wont be joining Face book !
On motor front we seem to be creating a new challenge for our self's now we got rig basics sorted . Speed is the new challenge lets face it was never really a issue for timelaspe but think we moved past that one .
If i build another rig i will try and reduce the gearing as much as i can . But as Gerald says theirs a point when the steps will become apparent on slow moves . Just a matter of testing and deciding what you are personally happy with . People have different standards and applications so theirs no set rule
D1

Edit
Hi Chis this may interest you

Bit simalar to Geralds studio stand


Sat Aug 20, 2016 12:59 am
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: Project Otto. Z axis problem...
Hi D1.

Good find - that is basically a homemade Foba with a rotation on the base...

Ideally you might want a two speed gearbox..? :) But in the absence of that, the Easy Servos are a good option if you don't want to go all the way to real servos, given their attendant problems of tricky tuning...


Sat Aug 20, 2016 1:59 am
Profile
User avatar

Joined: Tue Aug 09, 2016 10:31 am
Posts: 43
Post Re: Project Otto. Z axis problem...
this is all some awesome information, i appreciate everyone taking their time :)
its gonna take a bit to digest through all of this :)


Sat Aug 20, 2016 12:49 pm
Profile

Joined: Wed Jul 16, 2014 5:42 pm
Posts: 7
Post Re: Project Otto. Z axis problem...
Biolapse wrote:
this is all some awesome information, i appreciate everyone taking their time :)
its gonna take a bit to digest through all of this :)

Given the size parameters you specified, it would seem like these folks have already created something very similar.



They're probably not moving as much weight, nor have to be as precisely repeatable, but it's an interesting setup.


Sun Aug 21, 2016 12:42 pm
Profile

Joined: Tue Apr 09, 2013 6:49 pm
Posts: 175
Post Re: Project Otto. Z axis problem...
Yeah, the DF forum was getting hit a lot by spammers. Oh well. Come on DMAN join us! Join us! If this forum ever went down I would turn my personal website into a moco forum. Can't be that hard, can it? All to keep D1 happy:) Too bad about your Chronos project Chris but you have made the right decision. Family first, as they say. DMAN, my FB spy's are having a difficult time penetrating your shed! You must have some kind of cloaking device;)


Cheers,

OH, and on the Leadshine thingy, I do find I can push more power to my motors than I could with my BEDs and they still run very cool. Can't offhand say what my micro step is at but i think 32? Also, the one thing I remember that really screwed me up was not remembering to reset the drive after changing motors! That had me pulling hairs til Gerald reminded me. I kept switching out motors trying to figure out why my motors were not performing. The DUH moment! I know there is quite a debate about different drives but I am just happy mine are working the way i need them to.


Sun Aug 21, 2016 2:49 pm
Profile
User avatar

Joined: Tue Aug 09, 2016 10:31 am
Posts: 43
Post Re: Project Otto. Z axis problem...
Going to mount that little guy in the cradle. :)
Image


Mon Aug 22, 2016 3:37 pm
Profile
User avatar

Joined: Tue Nov 24, 2009 7:55 am
Posts: 951
Location: UK
Post Re: Project Otto. Z axis problem...
An interesting addition, another in the Z axis on top or a dual mount to convert to a Z axis could increase your move options.

_________________
Smile, it gives your face something to do!

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


Mon Aug 22, 2016 11:35 pm
Profile
User avatar

Joined: Tue Aug 09, 2016 10:31 am
Posts: 43
Post Re: Project Otto. Z axis problem...
This will be used for stereoscopic work. I would like to find another similar to that with about 18 inches of travel. I may have to build something.

Not sure the best way to tackle it. I think it will probably be a ball-screw type setup. Unless i can find a better way.


Tue Aug 23, 2016 5:54 am
Profile
User avatar

Joined: Tue Aug 09, 2016 10:31 am
Posts: 43
Post Re: Project Otto. Z axis problem...
I am considering this. it would allow for 20 inches of vertical motion which would be plenty for the type of work I do.
max vertical payload is about 80lbs.

http://www.aliexpress.com/item/linear-m ... 134.L2lceg

I am certainly open to other ideas. I cant help but think some sort of a pully system could possibly get more of a drop as long as there is some supportive structure.

I almost wonder if something like this would work. Get 3 sets of 15 inch tubes in a triangle shape, and use a pully system with counterweight to raise/lower.
https://alcobrametals.com/guides/telescoping-tube

or maybe better yet, with the configuration i may be able to put another 3 foot slider up on top of the gantry that has the cart attached to a cable. That can sit horizontal and use a pulley at the end to drop the cable down. Then i just have to find a way to stabilize the pan tilt. That gets rid of the need of a counterweight and i should still get very precise moves.


Tue Aug 23, 2016 7:37 am
Profile
User avatar

Joined: Wed Nov 11, 2009 10:45 am
Posts: 1685
Location: Merritt Island, Florida, Estates Unitas
Post Re: Project Otto. Z axis problem...
Gerald made an excellent suggestion in the other topic. That router I built is basically the X-Y gantry you talked about at the beginning of this discussion. If there is room for it overhead, you could extend the Z-slide and put your P/T head on the end. You can cut the gantry plates on your fireball router since they are open source. The V-slot extrusions and v wheels make a surprisingly strong movable frame. Openbuilds.com has linear actuators similar to the one you were looking at, but they use a "C-Beam channeled T-slot extrusion" which is stronger, and their kits are cheaper than the actuator you were looking at. They also have lighter belt drive linear actuators.

Going with the telescoping tubing would be a bold move. No telling what unexpected behaviors might occur. I'd be very interested in how that works out if you go that route. I think I'd try the scissors jack/robot arm plan before attempting the telescoping tube, mainly because there have been more DIY builds with that general design, so more examples to draw ideas from.


Thu Aug 25, 2016 6:28 am
Profile
User avatar

Joined: Thu Mar 10, 2011 11:38 pm
Posts: 585
Location: Exmouth, Western Australia
Post Re: Project Otto. Z axis problem...
Re steppers and drivers:
I bought a Leadshine DM865 to compare it's performance on my CNC router compared to the $12 TB6600 based drivers I'd been using before. I also changed from using a string of car batteries to a switch-mode PSU.

The Leadshine made the motors buzz very loudly at about 1kHz with the PSU but not the batteries. After running the auto-tune function on the driver the buzz changed to around 60Hz. The other drivers didn't seem to notice the difference. The main thing this tells me is never to forget the influence the power supply has on the operation of any piece of equipment.

I haven't used the PC based tuning software available for the Leadshine but suspect that best performance will not be obtained without it.

My current projects are a GPS regulated wooden clock for me and a semi-automatic weaving loom for my wife. I'm using steppers in the loom and having similar issues with motors stalling before getting the speed I want. Accelerating through the mid-band resonance area seems to be the critical issue.

Funny where curiosity takes you.

Kit


Thu Aug 25, 2016 5:36 pm
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: Project Otto. Z axis problem...
Hi Kit. A transformer supply is the most reliable and least likely to cause issues (after a battery.. ). If you do use a switch mode supply it should be one designed for use with stepper motors. In one case I used an old laptop supply and had some issues - so I added a big capacitor to the output and that did help. I've never heard a 60hz hum from a motor, normally you might hear much higher frequencies, in the dog whistle range.

Software tuning can help with midband resonance, but sometimes it's just the motor is difficult. If the resonance is bad enough to stall the motor then it sounds a bit extreme - maybe the motors are not quite right for the job or the acceleration rates are too high. Maybe also the gear ratio is not high enough?


Thu Aug 25, 2016 11:46 pm
Profile
User avatar

Joined: Thu Mar 10, 2011 11:38 pm
Posts: 585
Location: Exmouth, Western Australia
Post Re: Project Otto. Z axis problem...
Gerald,
A mains supply is always better than batteries when available, but an old-fashioned unregulated supply is likely the best option for powering steppers. Absolute voltage is not critical but the ability to dump a bucketful of Coulombs into the motor at short notice when required is definitely a plus point. The cost of the required capacitors for a really potent power supply is a limiting factor for us impoverished amateurs (cue violins) but a switch-mode supply is merely the cheap, light-weight option for this application, not the best.

Kit


Fri Aug 26, 2016 2:49 am
Profile

Joined: Wed Jan 20, 2010 4:50 pm
Posts: 561
Post Re: Project Otto. Z axis problem...
Hi Kit "the ability to dump a bucketful of Coulombs into the motor at short notice" - that was why I mentioned the battery option - batteries are excellent at doing just that! You could always leave one on trickle charge? Otherwise surplus equipment is a good source of parts. Audio amplifiers usually have decent capacitors and transformers.. :)


Fri Aug 26, 2016 3:23 am
Profile
User avatar

Joined: Thu Mar 10, 2011 11:38 pm
Posts: 585
Location: Exmouth, Western Australia
Post Re: Project Otto. Z axis problem...
Gerald,
You have to work hard to get a lower source impedance than a lead/acid battery. It's keeping 4 of them fully charged that takes some management. Unless you build a 55 volt charger, which I haven't.


Fri Aug 26, 2016 3:55 am
Profile
User avatar

Joined: Tue Aug 09, 2016 10:31 am
Posts: 43
Post Re: Project Otto. Z axis problem...
Well here it is. The limited space Z Axis solution

I am using 4 pairs of heavy duty 500lb drawer bearings stacked into 4 rails. Those bolt between a 4x4x12 Aluminum square tube and an 8x8x12 square tube.

This actually turned out much better than I had expected. When fully extended there is approx one millimeter of play when it is horizontal. I think when it is hanging straight down it is going to work very well. It is still not done, I plan to prime and paint the tubes with some engine enamel to make them look a bit nicer. I have a few more holes to cut out, and i need to add a flange to the top so it will connect to the rails of the overhead system. The 8020 came in a few days ago, lots of beautiful black anodized aluminum. :)

Here it is collapsed. 14 inches long.

Image

Here it is extended, 38 inches long. 2 Solid feet of motion, and the camera will sit around upper chest height when fully retracted.

Image

Cant wait to try it out. The way things are going this may be up and running sooner than later.

Cost was not too bad. about 150 for the aluminum, another 280 for the bearings.


Fri Sep 16, 2016 3:02 pm
Profile
User avatar

Joined: Wed Nov 11, 2009 10:45 am
Posts: 1685
Location: Merritt Island, Florida, Estates Unitas
Post Re: Project Otto. Z axis problem...
WOW! Nice elevator. Have to admit, I would not have tried that myself. Once again your outside the box design strategy has created a unique solution. I'm amazed there is so little play. You're probably right, when it is vertical, gravity is going to hold it straight. Its perfect for your plants because the X and Y movement is not going to involve a lot of acceleration and there is plenty of settle time between exposures. I can't wait to see the finished rig.


Fri Sep 16, 2016 6:45 pm
Profile
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 43 posts ]  Go to page 1, 2  Next


Who is online

Users browsing this forum: No registered users and 4 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