[SCzEngrgGroup]

It sticks in my mind that some ethernet interfaces these days are adaptive: they don't care which pair is which. Rather cute engineering to be able to handle 100 MHz through switches like that, but it does solve a few installation problems.

Cheers
Roger

The Ethernet port on my laptop works like that, but the one on the machine does not.

I expect most modern (last year or two) PCs would be gigabit Ethernet, and would adapt. But I can't depend on that if the ATC is installed on an older machine, even my two-year-old Pro. Haven't tried it yet on the Pulsar, but I believe it uses the same MB.

Regards,
Ray L.

[SCzEngrgGroup]

It's funny how quickly we adapt to having "new technology". I've been using the ATC the last few weeks, and it's performed far better than I ever would have believed it could. Overall, it has worked surprisingly well, to the point I already take it for granted, and would be lost without it!

There have been very, very few problems, and all have been due to known deficiencies in the current hardware, or bonehead errors introduced when I've made changes in the firmware. The browser interface has worked perfectly, and has proven to be a very convenient way of controlling it. Once I learn a little AJAX, I expect to be able to significantly improve the capabilities of the browser interface.

I'm now in the middle of a new build of PDBs, and as soon as that's out of the way, I'll start building the first complete "production prototype" ATC. I will almost certainly make this one a 24-tool model, as it will take very little additional time and effort when I'm already making the parts for the 12-tool carousel - basically just make twice as many of most of the carousel parts. Once the production prototype is wrung out, I'll build a few beta units.

Regards,
Ray L.

[hayes]

How do we get on the beta list? I'm volunteering!

[SCzEngrgGroup]

Today I attacked a small project that's been on my ToDo list for a long time - converting my Harbor-Freight 12-ton manual hydraulic press to an 8-ton air-powered press. Here is what I started with, bought several years ago for about $70 on-sale:

12 Ton A-Frame Floor Shop Press

I recently purchased one of these:

8 Ton Long Ram Air / Hydraulic Jack

I grafted them together, like this:

Attachment 266472

The only modification to the air/hydraulic cylinder was drilling and tapping one hole in the base casting for a quick-release fitting. This allows for the addition of an external oil reservoir, mounted above the rest of the cylinder, which is necessary to make it operate upside-down. It can be operated either by air pressure, using the included air-motor pump (the big silver thing on the front, near the top), or manually, just like with the original short bottle jack. Maximum usable stroke is increased from a little under 8", to a little over 13". Since the return springs were not long enough for the longer stroke, I got two 14" dual-acting air cylinders, which will be mounted on either side of the ram, to push the piston back into the cylinder. I will also add adjustable limit switches, to turn the air motor off when it reaches a pre-set stroke.

I should have it finished up and working tomorrow AM.

Oh, and the ATC? It's been working just wonderfully! The last two little tweaks were:

1) Using a small screwdriver to carefully spread the TTS collet open, so it now naturally opens to a bit over 3/4" when the PDB is released. This makes it easier for the tool to get in.
2) Putting just the tiniest chamfer or bevel on the upper corner of the "ATC groove" on the TTS holders. Some of them have a very sharp corner, which would sometimes dig into the cam that locks the tools into the carousel. Just a quick touch with a file with the tool turning in the spindle is all it took.

With those two little changes, it appears to be virtually 100% reliable. I haven't had a failure in quite some time. I'll sometimes just fire it up running random toolchanges while I'm workign on other things, and it seems perfectly happy to do that for just as long as I let it.

In a few more days, I'll be starting to make the parts for the first true pre-production prototype ATC.

Regards,
Ray L.

[SCzEngrgGroup]

Some small projects accomplished recently:

Nearly completed the "power press". It's working, but I still have some modifications to make. The air motor is too slow, taking about a minute to run the full 14" stroke. By enlarging the piston in the pump I can make it much faster, at the expense of maximum force. Right now, it's good for 8 tons, which is WAY more than I really need. I'll likely reduce that down to about one ton, and go 8X as fast. I also need to swap out the air cylinders I use as "return springs". They are just a bit too weak to push that big cylinder back up. But, other than those two rather minor issues, I'm really happy with the conversion.

Attachment 267160

Next, I built myself a kind of universal electrical tester, using one of the gazillions of Arduino board I have laying around. It's a small box with an Arduino Mega2560 board in it. The Mega2560 has 75 I/Os. A few of these are used for a 4 line x 20 character LCD display, and a momentary rocker switch, which make up the "user interface". 69 I/Os are brought out to connectors, which can be connected to.... well, anything. I originally built it for testing the PDB and ATC cables, to make sure they were built correctly. It does a full open/short test, verifying correct continuity on all wires, in just a few seconds. MUCH better than a visual check. It will also be used for doing functional testing on electronic components (LCDs, motor drivers, encoders, etc) before they get assembled onto PCBs, or otherwise built into sub-systems. It will then to full functional testing on completely assembled PCBs and completely assembled PDB and ATC mechanical assemblies, before they go into burn-in testing. This will both vastly reduce the amount of time I spend testing, and re-working parts and assemblies, but also improve overall quality.

Attachment 267154 Attachment 267156 Attachment 267158

Regards,
Ray L.

[RCaffin]

wot a lot a wires...
:-)
Very impressive.

I made up a very much simpler test box with just Ezy-Hooks and LEDs. Invaluable beast too: it lets me monitor BoB signals and Encoder signals - anything in the 0-5V range.
Can't live without T&M equipment. (But can't always afford the commercial stuff, which often is stuffed with useless features.)

Cheers
Roger

[LeeWay]

I swapped an air hydraulic jack out for the original on my 20 ton press a while back. I put a foot pedal actuator on it. It works great, but is on the slow side. I find that I can add block in there to prevent the jack from fully retracting and that increases the cycle times quite a bit. I don't need the full stroke on things I have used it for. The air assist jack was the same size as the original, so no mods to the frame were needed.

[SCzEngrgGroup]

Made some real progress on the first pre-production/beta ATC today. I've now got all of the re-designed metal parts for the new transfer arm drive assembly almost done, and they came out just about perfect. What remains is making the Turcite linear bearings, Delrin leadnut, and a few smaller plastic pieces for the two arm sensors. The linear bearings will be interesting. The three pieces that attach to/ride on the ground linear shafts will be assembled to each other. The Turcite bearings will be rough-machined, with under-sized bores. The bearings, and the bores in the bottom plate that the shafts attach to will be line-bored as a unit, to ensure everything is perfectly in-line. The bearings will then be hand-reamed 1-2 thou over, for a nice, free-sliding fit. Finally, the outside of the bearing block assembly (the two big-a$$ blocks of aluminum in the photo) will get face-milled on all sides, and have all its corners chamfered, for cosmetics. With a little luck, that could all get done tomorrow.

The new transfer arm and tool fork assembly is also designed, along with a new, smaller transfer arm drive bracket, and will be fab'd in the next few days. The new arm is several inches shorter than the old one, and much stiffer, as is the bracket. Once those are all done, I'll be able to mount the new drive assembly on the machine, and run it with the existing carousel, by simply moving the carousel moved closer to the column.

The new carousel assembly really has no major changes from the current one, just detail changes (slightly improve bearing system supporting the carousel, improved viscous damper, larger air cylinder for tool lock, and a few more sensors). It should come together very quickly and easily once the new transfer arm drive is checked out.



Regards,
Ray L.

[FineLineAuto]

Looking good Ray. Can't wait to get one.


Nate
Fine Line Automation
Home - Fine Line Automation

[SCzEngrgGroup]

Here is the nearly complete transfer arm drive. Just a few more small parts to make, and round up some missing fasteners, and it'll be good to go!

Attachment 268522

I always have pretty low expectations for "first articles", but this one has come out just beautifully. Making the Turcite linear bearings was a total no-brainer, and the fit is absolutely perfect - they slide almost effortlessly, with no detectable slop whatsoever. And the whole assembly is incredibly rigid. With luck, I could have it running on the machine by Fri. evening, with the new, much stiffer mounting brackets, and the new and greatly improved transfer arm and forks.

Regards,
Ray L.

[SCzEngrgGroup]

A bit more progress, though not as much as I'd hoped. I did something just blind-stupid yesterday, and broke my $200 1/2" hex broach. As a result, I won't be able to complete the transfer arm drive until a replacement arrives next week. But, with the exception of the tool fork assembly, and the sensors, the transfer arm drive is done, and working wonderfully. Today I got the arm itself made, as well as the mounting bracket:

Attachment 268776

Regards,
Ray L.

[RCaffin]

A photo of the broken broach would be educational, along with knowing how it broke.

I made my own hex broach out of air-hardening steel. Many small steps of diameter. Takes a bit of force to drive it thru'. Works tho'

Cheers
Roger

[LeeWay]

That is one thing I haven't done yet. Run a broach of any kind.

[kvom]

I've used keyway broaches a few times. I assume a hex broach is for cleaning up the round corners after milling the hex hole.

[SCzEngrgGroup]

And, here's the new transfer arm fork. VASTLY improved over the previous ones. It really holds the tools securely, it won't even allow them to "rock" when the arm is accelerating, and will do a better job of locating itself to the ATC groove in the toolholder.



Regards,
Ray L.

[LeeWay]

Is that part steel, Ray? looks great.

[SCzEngrgGroup]

Is that part steel, Ray? looks great.

Lee,

All aluminum, except the 3/8" alignment pin. In the production model, the two halves of the "fork" may end up being steel, I haven't decided yet.

Regards,
Ray L.

[kvom]

I'd like to get an idea of the target price, since it will depend on whether I decide to upgrade to a Pulsar vs. a Pro. Any thoughts on a range?

[SCzEngrgGroup]

I'd like to get an idea of the target price, since it will depend on whether I decide to upgrade to a Pulsar vs. a Pro. Any thoughts on a range?

Price will be around $4K. You just can't do a quality toolchanger for much less than that and make a profit.

Regards,
Ray L.

[tmarks11]

Price will be around $4K. You just can't do a quality toolchanger for much less than that and make a profit.

That seems like more than a fair price to me. Tormach charges $4300, and yours is more complex and has more features than theirs.

Bet you are burning the midnight oil trying to get a production-like model ready for Cabin Fever. The new parts look awesome.