[SCzEngrgGroup]

Here's a short video of the ATC carousel doing random "seeks". Performance is about what I expect from the production system. As you can see, it's pretty quick. This is running with a small, under-powered servo motor, due to the incorrectly built encoders on the motors I expect to actually use. I hope to be able to kludge together an encoder to get the real motor going in the next couple of days.

http://youtu.be/v7kQPxKUQJ0

Regards,
Ray L.

[Bubba]

If thats the underpowered motor, I can just imagine what the proper power will do to it!!!!

Looks fantastic.

[Titaniumboy]

Ray,

That is pretty darn slick. NIce job!

I'm still not sure how the tools will be removed and replaced in the carousel, but it should become clear as your work continues.

The embedded video says "This video is unavailable.", but works fine if I view it at Youtube.

Titaniumboy

[keithmcelhinney]

That thing got a Hemi?

[ivans]

Very nice Ray. Aren't servos grand for applications like this?

Sent from my SCH-I545 using Tapatalk

[FineLineAuto]

Looks really nice. Much better than the ATC designs on other machines in this class.


Nate
Fine Line Automation
Home - Fine Line Automation

[rlockwood]

To hell with class, do I need to record a video of the 1.5 seconds a fadal takes to move from one cartridge to the next? Going from 10-1 takes something like a painful 30 seconds.

Nice freaking work man!

Sent from my Nexus 5 using Tapatalk

[johnedward]

To hell with class, do I need to record a video of the 1.5 seconds a fadal takes to move from one cartridge to the next? Going from 10-1 takes something like a painful 30 seconds.

Nice freaking work man!

Sent from my Nexus 5 using Tapatalk

is the fadal moving a small empty rack like shown here? i'll hold off praise until i see how fast/accurate it is with a full load of tools. what happened to the geneva mechanism?
john

[rlockwood]

is the fadal moving a small empty rack like shown here? i'll hold off praise until i see how fast/accurate it is with a full load of tools. what happened to the geneva mechanism?
john

Empty or full (on the fadal) doesn't really matter. Runs a fixed speed motor and geneva gear. Painfully slow.

[johnedward]

ray used to use a geneva mechanism for his atc, wonder why he didn't here? painfully slow? so your comparison of his present rack to one using the geneva makes little sense, how it compares to a similarly rotated design would be more logical.
john

[rlockwood]

The majority of ATCs used to use Geneva mechanisms, and now use servos in their place.. Its likely he changed for the same reasons everyone else did.

Sent from my Nexus 5 using Tapatalk

[TiagoSantos]

given how cheap and cheerful servo motors are now, I don't think it makes much sense to use a geneva mechanism anymore. Even with a rack full of tools, I'm sure Ray won't have any trouble finding an affordable servo that can spin the tool rack that fast.

Of course the other consideration is Ray's solution for holding the tools in the rack. With the typical hobby grade ATC's we've seen before (and Tormach's..) the tool isn't securely held to the carrousel. Spinning it this fast would be a good way to fling a bunch of tools all over the place

[bozidar22]

The servo looks like one of those pololu gear motors used in robots. I can't imagine it lasting very long.

Could be wrong though. He may be just using it for testing.

[Titaniumboy]

Just spent the last couple of hours completing my El Cheapo Coropolast enclosure. I fastened the panels together using eight #8 machine screws and nuts along each vertical corner (I will swap out the nuts for Nylocks....). I found the top frame was not really necessary, so just put a single 1"x1" stick across the top front, to square up the opening. I added 1"x1" sticks along the vertical edges of the front opening, to stiffen them up as well. The bottom ends of these are notched, to go around the lip on the chip tray, which also provides the vertical alignment of the panels, ensuring the opening remains square. If it ever needs to be removed, I can simply remove the screws fastening the stick across the top, and each half simply lifts/slides out as a unit. Total time to remove or re-install: under 10 minutes.

I must say, I could not be more pleased with how this has worked out, and wish I'd done it a long time ago. I can't see much reason right now why I'd want to build anything more "permanent", given the 10X or more cost differential. This is very functional, dirt cheap (about $50 in materials), and is structurally quite sound. As an added bonus, the white panels have improved lighting on the machine significantly. Total build time was probably about 3 hours. If it ever gets damaged, it will cost only a few dollars, and a few minutes, to replace the damaged part, and be back in business.

Next up: Do the same for the Pulsar!

Attachment 214904
Attachment 214906

Ray,

Did this end up being the final version of your coolant enclosure? Did you have to seal up between the Coroplast panels, or between the Coroplast panels and the chip tray?


Titaniumboy

[SCzEngrgGroup]

Ray,

Did this end up being the final version of your coolant enclosure? Did you have to seal up between the Coroplast panels, or between the Coroplast panels and the chip tray?


Titaniumboy

I sealed nothing. The only leak I have, is where the tray was damaged in shipping, and I re-welded a seam, but didn't get it fully closed up. And that only leaks when I aim the washdown hose directly at that seam - there are no leaks when actually running the machine. The enclosure has worked better than I ever dreamed. The only change I want to make is to get rid of the shower curtains closing off the opening, and replace them with Plexi sliding doors.

Regards,
Ray L.

[Titaniumboy]

Ray,

So you didn't use any sealant between panels? How the heck do you not have any leakage?

In your post, you mentioned that you used vertical 1x1's along the front. What did you use to attach the Coroplast panels to the sides and rear of the chip tray?


On another note, do you have room in your luggage for five coolant pumps when you go to Cabin Fever?

Titaniumboy

[SCzEngrgGroup]

The Coroplast panels overlap several inches at each corner, so coolant never gets far enough around the corners to leak. The panels are simple resting on the inside of the chip trays, so no sealant is needed anywhere - the chip try walls are plenty tall enough to keep coolant from getting out. I can remove the one 1x1 that runs across the top of the two front panels, then lift the entire enclosure off in two pieces.

Regards,
Ray L.

[SCzEngrgGroup]

Things continue to move along very nicely on the ATC. I've completely eliminated the pneumatics, and I'm now confident I've found suitable motors/gearboxes for all the various drive mechanisms. I should have samples to confirm performance early next week. I've made the encoder selections, and figured out where/how to mount them - one of the trickier aspects of the whole design.

I've also just completed the transfer arm drive design. It's slightly more complex that I'd hoped for, but I am confident it will work really nicely, and provide all the performance we can safely use. If I can get all the necessary parts and tooling, I should have the first prototype built in the next week or so.

Attachment 229094

Looking at the expected times for all the various steps of the toolchange sequence, I expect total worst-case toolchange time (for the base 12-tool configuration) to be under 10 seconds.from spindle stop to spindle start. The hardware, and firmware, will be capable of going much faster, but we'll have to wait and see if that makes sense, from a safety standpoint.

Regards,
Ray L.

[sagreen]

Just spent the last couple of hours completing my El Cheapo Coropolast enclosure. I fastened the panels together using eight #8 machine screws and nuts along each vertical corner (I will swap out the nuts for Nylocks....). I found the top frame was not really necessary, so just put a single 1"x1" stick across the top front, to square up the opening. I added 1"x1" sticks along the vertical edges of the front opening, to stiffen them up as well. The bottom ends of these are notched, to go around the lip on the chip tray, which also provides the vertical alignment of the panels, ensuring the opening remains square. If it ever needs to be removed, I can simply remove the screws fastening the stick across the top, and each half simply lifts/slides out as a unit. Total time to remove or re-install: under 10 minutes.

I must say, I could not be more pleased with how this has worked out, and wish I'd done it a long time ago. I can't see much reason right now why I'd want to build anything more "permanent", given the 10X or more cost differential. This is very functional, dirt cheap (about $50 in materials), and is structurally quite sound. As an added bonus, the white panels have improved lighting on the machine significantly. Total build time was probably about 3 hours. If it ever gets damaged, it will cost only a few dollars, and a few minutes, to replace the damaged part, and be back in business.

Next up: Do the same for the Pulsar!

Attachment 214904

Attachment 214906

Regards,
Ray L.

Ray, I have the coroplast and was going to build an enclosure like yours today. Why didnt you score the back, and bend the coroplast as opposed to bending and fastening with #8 screws?

Thanx,
Scott...

[SCzEngrgGroup]

Ray, I have the coroplast and was going to build an enclosure like yours today. Why didnt you score the back, and bend the coroplast as opposed to bending and fastening with #8 screws?

Thanx,
Scott...

Because each panel is longer than the Coroplast sheets, so there HAD to be seams somewhere. The easiest way to do it was to put seams at all the corners. And, Coroplast bends very nicely in the direction of the corrugations, not so nicely in the other direction. So, on mine, all the panels are oriented so the corrugations are vertical. On a smaller machine, no reason you couldn't do as you suggest.

Regards,
Ray L.