PDA

View Full Version : Screw Drive garage door opener



Kevin McFarl
11-04-2004, 03:28 AM
Hello all i am new to cnc
I am just starting to get ideas of building one of my own
I was just wondering if a srew drive type garage door openner would work in place of a stepper or a servo motor.

You have almost every thing you need to start with motor,screw,carage nut ect.

Any info would be helpful

Thanks all
Kevin

Alan T.
11-04-2004, 10:08 AM
It would not be very accurate. Ton's of backlash. You would also have to change the motor drive.

Cliff_J
11-04-2004, 11:04 AM
Kevin, I've got one and it also binds pretty easy if not lubed up, not a great attribute on top of the loose tolerances everywhere and flimsy construction. To add an encoder to use as a servo drive seems a big waste.

You'd be better off with a DIY bearing like found all over on this site and even an allthread screw.

Kevin McFarl
11-04-2004, 11:35 AM
Thank alot.
I am new to all of this so i dont know what i am doing yet.
I have some more ideas i am thinking about.

lurch
11-10-2004, 11:10 PM
Keep the ideas coming, bad ideas should get shot full of holes quickly here(lots of experience to draw from). Learning from others mistakes is really easier than making your own. The ideas that make it through the culling process should be good, solid ideas. You may even come up with some out of the box new ideas, that change the way we do this CNC thing.....!

High Seas
11-11-2004, 01:50 PM
I'll let you all know in about 6 months or more. Got a couple screwdrives from a few different sources. I like the setup possibilities - there are some differences - I note the rail from Craftsman (Sears) has some backlash - but the Genie seems to have very little "slop". They have differen pitch (TPI) and a different setup for the drive. There may be some backlash- but should make a pretty nice foam cutting rig - gotta cut some wings when I return from downunder! Reports when I get to it. Cheers - Jim

Physisist777
01-10-2009, 02:56 AM
you guys are ridiculous....I bought a craftsman screw assembly off ebay and tested it myself. with a new plastic shuttle and a clean screw, I measured no discernible backlash or loss in precision. I used a 6" caliper to measure relative to a stationary object. for a hobby cnc, I think accuracy beyond that of my caliper, over a range of 6 inches is more than sufficient. the only caveat is the shuttle must maintain tight contact with the screw to insure accuracy. but the cool thing is that the track does just that, as well as providing a 3 sided inclosure and plastic bushings to eliminate lateral "wobble".

the one issue I see is there is low efficiency due to the plastic bushings. but accuracy is defiantly not an issue with these things; especially considering the inherent inaccuracies of the other components (skate bearings, etc)

so guys, if someone comes up with an innovative idea, don't bust there balls until you have personally verified your assumptions

jdell42
01-12-2009, 12:51 PM
you guys are ridiculous....I bought a craftsman screw assembly off ebay and tested it myself. with a new plastic shuttle and a clean screw, I measured no discernible backlash or loss in precision. I used a 6" caliper to measure relative to a stationary object. for a hobby cnc, I think accuracy beyond that of my caliper, over a range of 6 inches is more than sufficient. the only caveat is the shuttle must maintain tight contact with the screw to insure accuracy. but the cool thing is that the track does just that, as well as providing a 3 sided inclosure and plastic bushings to eliminate lateral "wobble".

the one issue I see is there is low efficiency due to the plastic bushings. but accuracy is defiantly not an issue with these things; especially considering the inherent inaccuracies of the other components (skate bearings, etc)

so guys, if someone comes up with an innovative idea, don't bust there balls until you have personally verified your assumptions

How did you test it? Assume you had some type of coupler to connect it to a stepper or servo?

NC Cams
01-12-2009, 01:25 PM
IMagine if you will a screw drive that ran a garage door up and down for 10-15 years. Most people never lube them, never clean them, As long as the door goes u/d, they simply push the button.

Localized wear is/would/could be a problem in the screw. SInce it is a relay drive, one could hardly easily turn it into a fast acting, bidriectional mosfet driven system. Maybe SCR's or TRIACS but not an easy H bridge like most DC servo amps use.

Doable? Yes. Qualified doable? defintiely.

OVersimplification is the root cause of too may "all you gotta do is" projects

Try using the search engine via 'garage door opener'.

I know this has been thrashed about in the past year or so and done so successfully. Maybe it is time to do it again - door openers go bad in winter and it is the time of the season for this topic to resurrect itself.

Physisist777
01-14-2009, 12:33 AM
I am only using the screw from the garage door opener as a supplement for more expensive acme screw. (paid $30 for 9 feet) The track/screw comes in 3 foot sections.

I am directly coupling the 3' screw to a 425 oz/in stepper and driving at microstep precision. Therefore I tested the precision by driving a stepper a given number of steps and comparing the distance traveled to the calculated distance traveled. (the screw I have is 6 tpi) and repeat...

well, I will be finished with my rig in a few weeks; so I will let you guys know how the garage door screws works out. its a 3x3x12 burn table/engraver/mill

FandZ
01-16-2009, 06:25 PM
Does anyone have a link to a picture of the garage door opener you are talking about. I just don't have a clue what you guys are talking about but would like to know.

NC Cams
01-16-2009, 06:37 PM
Google as follows "screw drive door opener". THousands of hits emerge

Physisist777
01-16-2009, 06:39 PM
Yeah...when I get home, I will get a pic posted

FandZ
01-16-2009, 06:50 PM
Google as follows "screw drive door opener". THousands of hits emerge

I did O wise one, searched eBay too. I didn't see anything that would remotely look useful for a cnc.

FandZ
01-16-2009, 06:56 PM
Is this what you are talking about?

Physisist777
01-16-2009, 11:14 PM
Okay, here are the pictures I promised.

The first picture shows the screw inside its galvanized 14 gauge channel with plastic "bushings" every 8 inches to support the screw

The second picture shows one of the 3' sections with the plastic "shuttle" on the far right. The shuttle is 5 inches long. so, on one section, the effective travel length is 31.5"
- It is not plausible to use more than one section because the supplied junction has about a 2 degree slop. but 31" is plenty for most applications

The third picture shows up close of the 1/2" screw. The screw is 6 tpi, dual start. so, 3 turns moves one inch. with a 1.8 degree/step motor, that equates to a 0.0016 inch precision or better.

I see potential problems with the plastic shuttle wearing, or the plastic bushings wearing - which would both result in a lack of precision. but as far as screw wear...defiantly not. plastic on metal yields worn plastic. anyway, the screw is made of cold rolled steel and is force through progressive dies to form the threads. this is why there is a slight indention in each thread. but, considering how they are formed, I am fairly certain the screw reflects the accuracy of the die used to form it, and thus would be reiterated throughout the screw.

your picture Fandz, is probably an on-motor limit switch whereas the center screw is turned by the motor via worm drive and laterally moves a small metal strip up and down to engage the corresponding limit switch.
-you could theoretical build a compact limit switch for your CNC in this same manner. it would require gearing and probably be a pain in the ass. like mine would need a 6 inch 10-32 screw without gearing. But that is a discussion for another thread - I would be happy to start one if anyone is interested

FandZ
01-17-2009, 01:09 AM
Thanks for the picture. I saw them while I was searching but they were labeled as extensions and thought that couldn't be it. That screw looks great. Sounds to me you need to figure out a way to make a ball screw assembly for less. However, what kind of plastic is inside there? If it is like those derlin nuts I don't think you will have a problem. With a little lube those things are supposed to last a good while. I think the guy that owns cncdumpster used one for over a year under hard loads and when he took it off it was still in great shape. I read that somewhere on their website.


Below is what I drew up real quick a solution. It would be composed of angle steel, skate bearings, and some bolts and washers. The outer skate bearings would dig into the groves and serve to screw forward and back and stop backlash since they would be taunt on opposite sides. The inner bearing would roll on the outside of the threads and in conjunction with the outer bearings keep everything on track. It would work but I don't know how the friction on the inner bearings would be. Food for thought really.

I would like to sit and figure out how to make an inexpensive ball screw assembly. That would be very beneficial to a lot of DIY'ers.

Physisist777
01-17-2009, 02:20 AM
Those extensions are the same thing, but you do NOT want those. they are 20ish bucks for only 18 inches. you can buy a real ACME rod for that. So, stake out ebay for new or lightly used Garage door tracks. I paid 30 bucks for a nine foot track. I think the plastic is Derlin or something similar...I would have to contact the manufacture for sure though

Thats a pretty fancy sketch up...but a few things about your idea (I know its a quick sketch, so take my criticism lightly)
1) I dont really see any way for backlash compensation (I do see one possible way - but its not very good) ...so everything in the assembly would have to be nuts on (no pun intended)
2) your top bearings would need to be aligned with the pitch of your screw for optimum operation - not impossible, but a pain in the ass to do accurately
3) your design incorporates way too many variables. each component had its own inherent slop which exponentially exasperates the backlash. for instance - the precision of the bearings, the "snugness" of the bearings to the thread, the clearance of the bolt and the bearing, etc.
4)with the outer bearings digging into the threads - you have a right angle fitting inside a circle (kinda), so you are only going to have one point of contact at the very top of the thread. very good considering friction and efficiency, but will result in quick wear of the screw - the bearing is much harder than the screw and with rapid direction changes under load, will destroy the top edge of the screw.
consider the surface area of the point of contact; very small...and then consider the peak force applied to the POC during a direction change while cutting deep. this force would be massive. you could calculate it if you wanted...but the point is - Fast wear

albeit, a great idea...I just see a lot of technical issues that you could probably fix, but would be a huge pain in the ass

I think the simplest method would be doing the cutting board trick. where you get a plastic cutting board and sandwich the screw between. is that what you where referring to with the Derlin nuts?

I have a few ideas for easy Ball nuts...but I need to draw them up first.

BillTodd
01-17-2009, 06:56 AM
Sounds to me you need to figure out a way to make a ball screw assembly for less.

See:A different (better) kind of lead screw (http://www.cnczone.com/forums/showthread.php?t=13593) and
Roller Nut projects (http://www.cnczone.com/forums/showthread.php?p=552830#post552830)

FandZ
01-17-2009, 10:28 AM
See:A different (better) kind of lead screw (http://www.cnczone.com/forums/showthread.php?t=13593) and
Roller Nut projects (http://www.cnczone.com/forums/showthread.php?p=552830#post552830)

Those are great links but the level of those projects looks like something you'd build on your second build. I don't see how you could easily make them with a drill press and hand tools, for a novice builder like me. but if I had a CNC machine going, it would be cherry.

BillTodd
01-17-2009, 01:21 PM
Those are great links but the level of those projects looks like something you'd build on your second build. I don't see how you could easily make them with a drill press and hand tools, for a novice builder like me. but if I had a CNC machine going, it would be cherry.

You'd need a lathe or access to one to make the rollers etc. But, honestly the hardest thing is retro-fitting them to the machine. If you can design the machine with roller-nuts in mind (e.g. allow space for the nut) then the difficulty is halved.

Think about it this way; You are probably going to have to make a new nut for your double start thread at some point (either because of wear or excessive backlash), that's going to be hard (harder in my opinion than making a roller nut) and/or expensive, so if you plan now for a roller-nut at least you'll have the option to try it later.

BillTodd
01-17-2009, 01:29 PM
The outer skate bearings would dig into the groves and serve to screw forward and back and stop backlash since they would be taunt on opposite sides. The inner bearing would roll on the outside of the threads and in conjunction with the outer bearings keep everything on track. It would work but I don't know how the friction on the inner bearings would be. Food for thought really

Could work, look at this thread:http://www.cnczone.com/forums/showthread.php?t=66381

FandZ
01-17-2009, 05:34 PM
I've already started building my machine. I've got the table done and the Y axis rolling. Seeing something roll for the first time really pumped me up...

Anyways, I've left plenty of room for future part upgrades. But I've already reinvented the wheel a few times, I'm probably going to go with an acme rod and derlin nut to start with. Later on I may look into a more exotic option.

I do appreciate threads like this that make you think outside of the box.



Could work, look at this thread:http://www.cnczone.com/forums/showthread.php?t=66381


You'd need a lathe or access to one to make the rollers etc. But, honestly the hardest thing is retro-fitting them to the machine. If you can design the machine with roller-nuts in mind (e.g. allow space for the nut) then the difficulty is halved.

Think about it this way; You are probably going to have to make a new nut for your double start thread at some point (either because of wear or excessive backlash), that's going to be hard (harder in my opinion than making a roller nut) and/or expensive, so if you plan now for a roller-nut at least you'll have the option to try it later.