[LeeWay]

It's a shame that it's not reliable!

It seems like a reliable tool changer would always be a money-saver when doing production. Sure, you're losing 2.4 hours to the ATC, but you're gaining ~0.25 hours *per job* of "not standing in front of the machine" time. If you could confidently spend those quarter-hours doing other productive stuff, you could make up for the lost efficiency.

You took the words right out of my mouth. If you have a reliable tool changer, it's almost like hiring another employee. One doesn't have to change the tools. Some of my jobs have 7 or 8 tool changes as well. In fact the parts that I have to make the most of are this way and I make those every week. I am working on budgeting for both the PDB and ATC when Ray gets them rolling. I like the idea of an employee that doesn't talk back and stays the hell off the phone.

[G59]

Wow, I'm lost here. Cycle time, and the chance it might bore to the table??????
Even manually changing tools, it could happen then. That is all Mach problems.
Get a different controller is only other option.
You all should be happy the ATC works even if it takes a 20 second cycle time.
The time it allows you to do other things is priceless.

[keen]

Very interesting thread - I had been thinking of a ATC ....but had decided, when that can be truly justified, that is the time to step up to a truly reliable VNC that can be trusted to run by its self making parts, so that I can concentrate on other things. If you cant trust your ATC to be reliable, then you may as well stay manual...I sympathise with LFR, its no good if it causes blood pressure rise! Keen.

[shred]

I use the ATC all day long to run small-batch production parts. ~6-10 minutes and 5-8 changes per. Once it's aligned and setup, it's plenty reliable for me to be doing other things, be it cutting more stock, second ops on finished parts, CAD work on the next one or just surfing the web.

The change speed is somewhat annoying and I fiddle with tools and operation order to minimize them, but by far the limitation on change speed is the Z rapid move to the change position and back, and the only way to improve that is either moving the ATC closer to the work (the mounting on mine has slots to do this although I don't) or finding a way to up the Z feed.

[treyjugson]

Very interesting thread - I had been thinking of a ATC ....but had decided, when that can be truly justified, that is the time to step up to a truly reliable VNC that can be trusted to run by its self making parts, so that I can concentrate on other things. If you cant trust your ATC to be reliable, then you may as well stay manual...I sympathise with LFR, its no good if it causes blood pressure rise! Keen.

you aren't reading what lrf said,

Now when I say reliability I am not implying that crashes are common because they are not.

there is a thing called preventative maintenance. to expect something to work forever without error is foolish. you change the oil, filters and belts in your auto from time to time don't you? or do you run it till it seizes then go back to walking everywhere? a few minutes a months could prevent hours of frustration later.

[keen]

you aren't reading what lrf said, there is a thing called preventative maintenance. to expect something to work forever without error is foolish. you change the oil, filters and belts in your auto from time to time don't you? or do you run it till it seizes then go back to walking everywhere? a few minutes a months could prevent hours of frustration later.

Yes but even occasional crash potential would stress me out. Do run a reliable ATC? I am interested in finding out how many do, and how many don't, not whether my plan is right or wrong. I am happy to stand corrected. Keen

[R.DesJardin]

LRF its unfortunate you are having issues I have one of the first ATC's and have next to zero issues with it. Just simple maintenance things but I am not afraid to go in the house while the machine runs. The advantage for me was while the machine runs I am running other machines or assembling product. This essentially doubles the money in my pocket per hour. I also have tried to fixture up so the machine runs for longer times by cutting more parts this lets me be even more productive in the shop. I also constantly tweak the programs to get the fastest machine time possible. Some programs I have completely rewritten to be more efficient to reduce machining time. Multiple vises and fixture plates help.

Regards and well wishes,

[philbur]

:drowning:

I can't follow your logic or your math.

Without the ATC you spend 3 working days changing tools in order to produce your hundred parts. With the ATC you have 3 working days to do something else productive. The logic is plain, the question is what value do you put on your time. If you put a none typical, low value on your time then the question would be "why did you buy the ATC".:drowning:

Phil

[LRF]

Rory,
Good to hear from you, hoping things are going your way.
For me, after ~40 years of being a class A toolmaker, tooling design & manufacturing engineer and most lately a Senior Process Manufacturing engineer, in the R&D development department, I am thinking about retiring, my goal being to move aside from the 8 to 5 job by 2017, if they let me go.

This discussion was started to discuss the ATC from a cycle time and a desire to make it quicker and improve it if possible.

I know how to multi-task and I do it assembling devices, surfing the web and watching TV and oh yes also changing the tools.

As for reliability and issues, you have seen some of them and you know of them because we talked on them while you were with Tormach. There is no one out their that is more supportive of the company because it has allowed me and my son to prosper quite well with our products while still collecting my day job income.

I did an estimate of how many times my ATC has changed tools. I did this by counting how many parts I have made and the number of tool changes needed per part. My ATC has cycled aprox 11000 times. In that period I have had about 6 or 7 dropped tools. The first 3 because of the set screw issue, which if I remember was finally resolved after I talked to you. The others because of the collet tightening up which you and I also talked about.

[SCzEngrgGroup]

I've designed several ATCs, and I can tell you that with the existing ATC architecture, there's not a whole lot that can be done to speed it up. Absolute best case you might shave off a few seconds per toolchange. But the cost of that would be increased wear, tear, and stress on the whole mechanism, which would hardly be worth it. Speeding up the carousel would likely require a larger stepper motor, and almost certainly a better coupling to the carousel. The gain, on an "average" toolchange, would be no more than 1-2 seconds, since the "average" seek will cover only about 2-3 "slots". You could move the ATC into position under the spindle faster, but they you'd have to stop it faster as well. This would likely require a larger air cylinder, and perhaps beefier linear rails and carriages. This would almost certainly result in more stress on the tool "forks", due to the rapid acceleration and deceleration, plus put much more stress on the mounting brackets, etc., which may have to be beefed up to prevent vibration and rattling. The PDB is already about as fast as it can be, so not much room for improvement there. You could perhaps move the head faster and gain a second or two. End of the day, you'd probably go from 20 seconds to, at best, about 15 seconds - a relatively insignificant change. Getting down to 10 seconds, would be HARD, as it would require major improvement in the performance of every, single operation in the toolchange cycle. This would result in a mechanism that is MUCH more expensive, and heavier, due to the much more rugged, high-performance components that would be required, and much tighter control would be needed to make it operate reliably.

Something else to keep in mind - TTS is highly sub-optimal when it comes to fast auto-toolchanges - that just wasn't a priority in the original design. "Real" ATCs require much more expensive toolholders, which make the toolchangers job MUCH easier, and more reliable.

My own toolchangers, a 10-tool system I built for my old Bridgeport-style knee mill (which is similar to the Tormach ATC), and the current 12-/18-/24-tool system for the Novakon Torus Pro and Pulsar also run about 20 seconds average toolchange time. Reducing that to 10 seconds would require near-total re-design of the carousel/chain and transfer arm assemblies, and would about double my BOM cost due to the added cost of the more powerful and robust motors, gearboxes, guides, coupling, supports, etc.. And reliability would absolutely suffer, due to the difficulties the use of TTS introduces. But, are you willing to pay twice as much for twice the speed? I'm not, and I suspect very few people would be. It makes no sense to spend nearly as much on the ATC as you did on the machine itself.

If you want fast toolchanges, buy a commercial VMC. But, be careful, because even many of them are not a whole lot faster, until you get up into the really $$$$ ones. And the really fast ones (like the Chiron below, which has a sub-two-second chip-to-chip time) tend to be rather finicky, and hideously difficult and expensive to repair when they break.

https://www.youtube.com/watch?v=97KQ5DKSuNw

Regards,
Ray L.

[CadRhino]

Words of wisdom. My vise setup has saved me more than once.
I look forward to transitioning to a pallet but then again if there were a crash it would happen fast.

I really dont trust mine to run with no operator, every so often I get to trusting Mach/windows, then it lets me down and does something definitely not in the program!! I read about it on other forums, but its not for me.......

edit: I do all my machining in the vise, in my opinion there is more time to shut it down if it should decide to bore into the table............

[R.DesJardin]

Rory,
Good to hear from you, hoping things are going your way.
For me, after ~40 years of being a class A toolmaker, tooling design & manufacturing engineer and most lately a Senior Process Manufacturing engineer, in the R&D development department, I am thinking about retiring, my goal being to move aside from the 8 to 5 job by 2017, if they let me go.

This discussion was started to discuss the ATC from a cycle time and a desire to make it quicker and improve it if possible.

I know how to multi-task and I do it assembling devices, surfing the web and watching TV and oh yes also changing the tools.

As for reliability and issues, you have seen some of them and you know of them because we talked on them while you were with Tormach. There is no one out their that is more supportive of the company because it has allowed me and my son to prosper quite well with our products while still collecting my day job income.

I did an estimate of how many times my ATC has changed tools. I did this by counting how many parts I have made and the number of tool changes needed per part. My ATC has cycled aprox 11000 times. In that period I have had about 6 or 7 dropped tools. The first 3 because of the set screw issue, which if I remember was finally resolved after I talked to you. The others because of the collet tightening up which you and I also talked about.

Lynn,
I fully understand your intentions but I was hoping to help enlighten a few that may not be familiar with reducing machining costs and improving profit. Some of this stuff is very new to them.
I am doing good here, my business had its best sales year ever last year so that was awesome and this year looks to be better yet.
Side Note: How do these threads always end up with comparisons/reference to the other brand? :tired: Just saying.

[LRF]

Thanks Rory, glad to here of your success.
To your comment, the Tormach meets all my capacity needs at this time. When the company needs a VMC I will decide that, then.

[Ken_Shea]

LRF,
Here is another possible help.
Look up PappaBear10 here on the CNC forum and email him. He is about a GURU with Mach3. While not knowing how much access you may have to that code he may and be able to modify it to increase the rapids on the tool changes. Actually PappaBear10 (Scott) developed the first working prototype 12 station ATC for the Tormach a few years back (Cannot believe it was 7 years ago) . He did all the hard work, interface, mechanical and electronic design and I did all the machining of components. Apparently It was not completed in time to make Tormach happy and could well have been other reason but it was all for naught at that point. It would not surprise me one bit if the current design does not incorporate much of his work. If it can be modified he would be the man to be able to do it.

Ken

For a nostalgic look back
https://www.youtube.com/watch?x-yt-c...-ts=1421914688

[SCzEngrgGroup]

LRF,
Here is another possible help.
Look up PappaBear10 here on the CNC forum and email him. He is about a GURU with Mach3. While not knowing how much access you may have to that code he may and be able to modify it to increase the rapids on the tool changes. Actually PappaBear10 (Scott) developed the first working prototype 12 station ATC for the Tormach a few years back (Cannot believe it was 7 years ago) . He did all the hard work, interface, mechanical and electronic design and I did all the machining of components. Apparently It was not completed in time to make Tormach happy and could well have been other reason but it was all for naught at that point. It would not surprise me one bit if the current design does not incorporate much of his work. If it can be modified he would be the man to be able to do it.

Ken

For a nostalgic look back
https://www.youtube.com/watch?x-yt-c...-ts=1421914688

You're not going to get Scotts help with anything in any way related to Tormach. He is not at all happy with them....

Regards,
Ray L.

[Ken_Shea]

LOL, yeah, your're right on there.

But he's a good guy and this would not be for Tormach directly. Alsio don't think Tormach would increase the rapids for the ATC for reasons previously mentioned, it seem obvious that they could if they wanted.

.

Ken

[shred]

All the ATC-specific things -- tray in, tray out, rotate tray, PDB, etc aren't much slower than the HAAS machines I've worked with.
The difference is all in the Z movement. 90 IPM vs 1000 IPM.

[Ken_Shea]

More thought on the issue.
Don't know what size of component he is machining but if one could and did go to a double vise set up and machine two at a time (really as many as possible) this would essentially be the same as halving tool change times.

Ken

[Ken_Shea]

It doesn't look to be hitting 20ipm, be interesting to time/distance the tool change Z movement.

[WOTDesigns]

I'd be willing to bet that they are just using a macro to control the rapid speed all of which were able to be modified via windows browser in the mach/tormach file. For example the delay after spindle on was irritatingly slow since I always start more than 3 inches above the part which is plenty of time to get the spindle up to speed without sitting still first. I just found the "m3" macro and changed the delay time per 1000rpm to a lower number.

I'll bet the A T C Macro can be adjusted as well.

Brian
WOT Designs