I believe that it is usual to have drive dogs with BT30 so spindle orientation (at least to 0 or 180 degrees from the standard orientation) will be required. Looking at the Tormach site their old BT30 spindle supports drive dogs.
Printable View
I believe that it is usual to have drive dogs with BT30 so spindle orientation (at least to 0 or 180 degrees from the standard orientation) will be required. Looking at the Tormach site their old BT30 spindle supports drive dogs.
Embrace the healing power of "and" :)
I think of this in terms of two key metrics: total operator time per part, and how long the spindle is on in between operator actions. For a large operation or long production run, the difference between 30 and 90 seconds of operator time per part may be very significant. For a sole proprietor, the difference between needing to swap tools every 5-10 minutes or being able to walk away for 30+ minutes may be a lot more significant.
I made a few runs (totaling about 60 each of 4 parts) for a customer without an ATC. I did a few sets in the vise as a first article test, and when he said he'd want a couple dozen at least, I made a fixture that could hold ten each of any of the parts using Mitee-Bite hex clamps. The time to design and make the fixture was considerable even amortized over 200+ parts, but I couldn't have done that number of parts without it.
Still, because the run time of each operation was relatively low (~1 minute per part per tool), I basically ended up standing in front of the machine the whole time. When the customer came back and said he wanted 100-200 sets, I told him it was probably time to find a "real shop" as that would have meant a full 40+ week in front of the machine, and the money wasn't enough to make that worthwhile for me. An ATC wouldn't have reduced the total run time, but it might have allowed me to do anything else in the shop, so I might have taken that on.
As it happens, I *just* got an ATC delivered this week and am hoping to get it up this weekend. I bought this largely as a late Christmas present to myself, and will be curious to see how much it affects my work. These days I do mostly one-off parts for my own projects, so there's fewer opportunities to save time with multi-part fixtures.
Hi Cliff, you asked my opinion in another thread but I thought it would be better here.
I don't own an ATC, but if I used my mill for business I would almost certainly have one. Their price (and shipping costs) is the only reason I don't have one now. Some have reported issues with reliability but I suspect you'd be able to sort anything out. If you pick up an older version make sure you upgrade to the latest PathPilot version.
Instead of 3 jaw chucks I'd probably have made a fixture with mitee-bites to hold far more parts, and for the parts you showed in the video this should be fairly straightforward. Once set up, a combination of a fixture together with an ATC would allow you to machine for stock with very little effort. I believe individual tools can still be changed manually if necessary, but would more likely be the exception than the rule.
I'd go for the new 12 tool version (which some have reported to be usable on non M series mills) just in case, but the 10 tool version is most likely more than adequate. You know your jobs best, so it should be fairly easy to estimate the maximum number of tools you'll need.
Creating a fixture is always an option, but using an ATC is only an option if you own one!
Step
The M-derived #39009 ATC is now the only one offered on Tormach's store for 1100 Series 3 machines. I never saw or ran the earlier version so can't comment as to differences beyond capacity, but mine has handled at least twenty tool changes well since I installed it a couple weeks ago :)
While it took me a little while to figure out the flow logic of the system at first (the documentation is a little confusing as written), I like the way the whole system works. Ultimately, I think the weak link in the system is the PDB, and that's the same as it's always been. I felt OK with that as I do mostly lighter milling and have never had a problem with tool pullout.
I have never used multiple offsets, I should check that out maybe??
I make multiple parts 4 or 5 at a time on one fixture and one offset, I manually change the tools, really not a big deal for me.
I draw up the parts in CAD as multiples then CAM up the file and machine it, I use each tool for each op on all the parts, then change the tool and do the same, some have 9 or ten different tools, I do this all on a G54 offset.
But then I am relatively new to CNC and there are many things I need to learn...........
For tools, you don't want to use the G54-type offsets, you will want to use the Hn Tn G42 M6 type offsets.
There's a table on the "Offsets" page in PathPilot that lets you measure the length of each tool (with a tool setter/gauge) and the mill knows where the tip of the tool is once you change to it.
(Assuming you have the correct tool offset in effect when you set your initial zero.)
The TTS system is designed to make this work -- the ground flat on the holder slams against the nose of the spindle, so you get a good reference every time. (The tool setter measures distance from flat of holder to tip of tool.)
Of course, if you do R8 collets and take the tool/collet out of the spindle for each change, the measurements won't work. (But you could have a tool setter instead, for a similar workflow?)
Assuming workholding is repeatable, and you have a power drawbar, it's probably easier to change tools (with offset) while the part is clamped, rather than swap out the parts.
But, back when I used a 1100 with manual drawbar, it was easier to keep the tool mounted, and swap the part out. So it depends on your specific workflow.
All my parts have the same Z0, on the first ops, then the part is put on a fixture and the second ops run, would that be the difference, a different Z0 for the second side?
A G54 for the first side and a G55 for the second side? If this is the case then it would eliminate having to set Z0 again for the second side?
Yes,
I use a few cam strategies for offsets to generate code and make parts.
1) Single vise using a different offset g54-g59x for each side of part. This would be same as above. Then at mill I would run code for each side using repeatable vise work stops. Each time I flip the part to new side I set new g5? x,y,z offsets for up to 6 or more sides. The mill now has learned or stored all the offsets for this complete part. Then next part I just load the vise with stock and run code. Each time the code finishes that side I flip or reposition the stock and continue code until finished part. I use pictures from cam simulation to show me how the part is flipped and where the g5? offset it located. PP even has the ability to reference the pictures during code execution and show them on the control screen to prompt you. Makes multi sided complex parts a breeze to do months later. Even after you have done many other setups and parts. If you take the time to set it up with pictures :) btw I use screen clips for all the strategies noted here at mill to aid in getting it done right, but no display in pp just a small tablet computer for pictures and reference. Simple and fast
2) Two or more vises or fixtures and a offset for each of those fixtures. Then completing one side of part at each of those fixture locations. The process for setting the offsets is a little different. I load the fixtures with stock and locate all the g5? x,y,z fixture offsets then run the program and it performs all the operations at each offset. You end up with one side completed on 2 or more parts.
3) Two or more vises or fixtures and a g5? offset for each of those fixtures. Then completing a different side of part at each of those fixture offset locations, thus ending up with a finished part. "picked this strategy up on this forum" The process for teaching or setting the offsets is much like number 1. You load and run only first fixture code. Then I load 2nd fixture with part from first fixture and load new material in first fixture and run code for both fixture offsets. Repeat this for the number of sides/ fixtures/offsets that make up the part program and you end up with finished part on last fixture in series.
Then if part run is done I run code for the last couple offsets/locations only to complete the last few sides of the last few parts.
4) Custom made fixture that uses 1 offset like g54 with many operations repeated in a pattern on the fixture at some predefined distance. This is the most simple way to complete ONE side of as many parts as you can fit on a fixture and hold in place. My cam software makes this very straight forward process and by far the best way to mass produce. If the part has more then 1 side then a custom fixture would need to be made for each side of part but all run in batches and only one offset like g54 would be required no matter how many parts.
5) Same as Number 4 with multiple custom fixtures. Each fixture has a g5? offset with a repeated pattern of operations on each fixture/offset. This is decent way to do mixed batches of different parts or more then one side of large batches of parts without changing fixtures or offsets.
I setup templates in cam software for each of the above strategies ready to go, saves a little work. Still the first 3 strategies can take a great deal of time to setup and get right. Number 4 and 5 are easy to setup cam wise but it takes planning and other programs to build custom fixtures and of course small parts that fit in soft jaws or clamped on fixture plates .
And my opinion for keen is a ATC would only complement any of the above code/offset/fixture strategies. And as TS mentioned I think you are more then qualified to get a ATC working correctly and get your moneys worth :)
Thanks for all the replies, I think I have it now, I just have never used it.
I will experiment with this so that I understand it. I make mostly small parts in multiples, machining all the ops on one side, then place the stick of parts on a fixture and run a separate program to machine side B.
thanks again..........