HAAS VF1 retrofit

I don't mean to rain on your parade, but you really haven't gained anything . The only thing you don't have to do now is move the head up and down, you have effectively lost 4" of tool of table clearance . The cam just swaps tools, Matsuura has them also, you still have to count the tools, open and close the drawbar , get the pocket up and down ,in addition now you have the hassle of a random tool changer , which it it gets out or sequence, will be a mother to resequence. You can't save your tool locations with out some programming, although the tool table might if random tool changer is set to 1. Good luck, but I think you just made your job a bit more complicated.

The new tool changer arrived Thursday, it is smaller than I thought.
I don't think I am going to loose any clearance I can mount it a little higher then
the current carousel is, remember the spindle has to move up for the tool change.

The whole moving the Z axis is the main reason I decided to do this.

Now I can do the whole tool change in ladder logic. I am really weak in programming.

I am guessing when Tool change is set to random the program calls
the pocket number instead of the tool number ??

The tool changer is made by Gifu and I am sending a request to see if they
have a ladder program written for it.

It has about 6 proximity switches on it and two motors with brakes.

I understand g code a little , this is a hobby interest that I want to grow into a
full time job someday , unfortunately my main job gets in the way and i tend to head
straight home after work instead of going over to my shop to work on the machine.
And my shop is on the other side of town.

So I havent been able to spend a whole lot of time on it lately. I am only about 4-5 hours of wiring
away from having it working ,with exception of the the tool changer.

The learning curve is sort of intimidating. Recently I almost caved in and bought Camsoft
I built a machine with my old boss about 10 years ago and it was a piece of cake with Camsoft.

Coding something for a C component is probably out of the question, ladder logic doesnt look
too bad, I am trying to find a good book.

Subroutines will work but Id rather not have to modify the way the tool changer works
I think I would have to install a encoder and treat the tool magazine as a axis .

I found a few ladder programs that will work for the indexing of the magazine with just
a zero switch and a pocket switch.

I just have to spend more time on it and I am working on getting a new place to live with a
workshop/garage , all my leases are coming up in November. So I'm running out of time.

If I can get this thing working, install the twin arm changer and get it working it will be a huge
accomplishment.

I wish someone with experience with LinuxCNC lived close by, would make things so much easier.

On the machine is a geneva mechanism carousel.

The twin arm magazine works very similar , it is a roller cam, one rotation moves the magazine one tool,

The arm is a dual action cam, the side action moves the arm up and down, the circumference is
machined to do the rotation of the arm.

On the new changer it has a rotating drum that is timed to trigger proximity switches which
tell the controller when the arm is home, when to release the tool , when to clamp the tool and when
tool change is completed. the pocket tilt is operated by a air cylinder.

When I said boss, I meant my employer , but the machine was a Bridgeport too. I just visited
him a few weeks ago and its all apart again, upgrading to AC servos.

It may be good for me to put a shout out to user in Florida and see who responds.

I live in central Florida . Programming in Camsoft uses a similar to Basic lauguage , yes it has a graphical user interface, but it's not ladder. I seriously doubt you are just slam that tool changer on the machine and get it working. First off the tc has to mounted perfectly parallel to the. Spindle and square to the table, then the arm has to be exactly in the middle of the spindle taper, or it will drop tools,smash one side of the taper and or hammer the drive keys on the spindle. Moving it up could make the whole thing worse , as if you are lucky, the casting is drilled and reamed for both toolchangers , the chances of you getting that mounted straight are not real good, then you will either have to shim and or scrape to get alignment. You still gotta count tools. Cam operated tc are not new, Brown and Sharpe had them in the 1970s....

Well I found a new workshop to rent, it is big enough to fit all of my stuff in one place.
So the next few weekends I will be moving again.
Then hopefully I can get back to working on the machine again.

I have read the thread, because I have to integrate a toolchanger in my machine. At the moment I try to make the mesa cards run. Well after that In need the ATC. But its much more complicated than the simple Pickup type.
Not really, because I don't need to shift the Z-Axis while changing. But I have much more positions I must call in the right direction. My Machine have 2 independent tool arms and the arms were at a 90° rotate head.
So I have much outputs and nearly double of it inputs, because nearly each position is proofed with 1 proximity switch on each end. Is it possible to make a M-funktion of each partfunction. The original had it, what makes it easy, if an error is happened during the changing, I can simple get back or advance with it.
Because the toolchain is an independent numeric axis I prefer to turn it to the position of the released tool while the changing is running. That seems me better than the chaotic toolchanging, what sspeed have with his new Doublearm changer. I know LCC can use both, but if I have to change a damaged tool during work, its easier to access the right tool, if I don't must first search a list, whats the actual position of my desired tool.

I have to say I agree with JR1050, Camsoft isn't going to make this particularly much easier. It's a lot more expensive as well (in the thousands of $). They even charge you an annual support fee. The technical support is good, but not that much better then what you get here for LinuxCNC for free! That is really saying most about the support here though, because they are both quite good. But only LinucCNC provide it for free!

You are also bound to particular hardware platform, Galil controllers, that are quite ok, but about 10 fold the price of a standard Mesa setup. And to be honest the Mesa setup is more configurable and expandable. Camsoft won't work with Mesa hardware (which is hobby use only in their words.... I asked them if they would consider supporting it...)

And the GUI looks like it's made on an Amiga in the 1990's!!!! At least the one you get with the package. The only other alternative is to make your own. Probably almost as much work to do that with their software as with LinuxCNC. But with LinuxCNC you get access to Gmoccapy and other advanced GUI's ready made. There are no user made GUI's available for Camsoft.

On a purely technical note, the complicated task of making an ATC work is one of the more challenging things to do when retrofitting a machine like this. This will be complicated no matter what control system you choose. I know since I'm trying to do the same thing with a Maho MH600E from 1990.

I haven't decided what approach I'm going to go for yet as I'm working on other parts of it for now (PID tuning and mechanical restoration to get my axes moving). But given all the other functions I need to control on my machine, lubrication amongst them I figure I might as well go for ladder logic. I don't have the programming skills to do it in C (I did learn it once about 20 years ago in colleague, and don't fancy revisiting it.... if I don't have to at least).
As I see it the safety functions when things fail and what happens when it fails is the important part to get right. And I don't feel confident in fault finding and debugging in C.

Ladder programming is kind of more hands on, and you can follow what happens in simulations. And I did attend a course that had a fair amount of ladder programming in it at work not that long ago. BTW, the teacher on this course told us that as a rule of thumb 90% of the code will be for the safety and error handling, the rest will be the normal program. And the amount of work involved with the safety system will probably take at least 90% of the time spent testing as well...... this was coming from a man who had been in to PLC programming professionally for many years, so I take his word for it.

I don't think this will apply for ladder logic only though..... the need for safety and fault tolerance is the same no matter what programming language you choose. And debugging safety routines in C isn't something I think I will feel comfortable with.

Is there examples of things like this done in G-code for us to look at somewhere? It would be interesting to see and compare with ladder.

sspeed33317, can you tell us where you found the ladder logic to do the indexing with only zero count and pocket sensor? My ATC is built like that, so it would be of value to me to see that. As I guess others also might be interested.

Anders