[lapuser]

Haas is telling me that their new tool/part probe will only work on newer control boards due to lack of a math coprocessor on the older boards. My question is - is it possible to use my computer though dnc to run the probe on my mill? Has anyone here any experience with setting up a part and tool probe on a mill like this I would sure love to know how. Haas did offer to sell me a new mill {HA HA HA] if I really liked that option. Bought a 94 vf3 with 2900 spindle hrs on it for $25,000.00. It looked so unused it was amazing and so there is no price they could offer me on a new mill just to get that option. They sell a probe package for $5200 that I would love to use if I can run it dnc off of my computer. Anyone out there with ideas?

[gar]

060109-2047 EST USA

lapuser:

Several questions:

1. What is this "new tool/part probe"? I would assume it would be two different devices, although you could design one for both purposes by changing the probe.

2. Why is HAAS requiring a math coprocessor? What really requires this?

3. What is your definition of DNC and what purpose do you want it to serve in probing?

4. We have a 93 VF2 and it was purchased with the tool probe (note tool only). We have never really used this probe because it was more nuisance than value.

5 We bought a HL20 with tool setter. I did not consider the mechanical pivot to have good repeatability. In other words too far from 0.000,1" repeatability. Later it was destroyed by the turret and no good reason to fix it.

6. I see more value for a tool setter on a lathe than on a mill, except a really good rugged unit such as a laser beam on a mill can work very well.

.

[lapuser]

1 It is a toolsetter on the table and a Renishaw parts probe for the spindle. 2. The claim by Haas is that a math coproccessor is needed for the macros on the probing routine. This is only on newer boards-I believe from 99 on that also run the brushless motors. They claim that you can't use the older boards--although they did {ha ha} tell me I could get a new mill with those functions. I guess that it is doable but they will not help keep an older machine running if they can force someone into buying a new one. 3. DNC as in the computer serial port to rs232 on the mill direct communication. My computer is always hooked up to the mill so if I can figure out how to use this link I have lots of math coproccessor. A friend has a new mill that has this tool probe, part probe. He is very happy with it. He does parts set up with it and parts verification also. It is the lights out capability along with part probing that interest me most as there is nothing more risky than loading a big part and leaving it for the evening. Sometimes when you get back you have broken cutter garbage and no part. I would like to set my tools up without using shim stock for z level and an indicator for x-y zero too

[gar]

060110-1213 EST USA

lapuser:

If I understand your response, then the new HAAS system has two probe mechanisms. One is the tool setter, and the other is the part probe. These are not a common mechanism with a change of probes for the two functions.

We never check cutter diameter on the machine. We only set tool length on the machine. We set tool length with a home made 1" gage block from a reference surface, such as the left rear vise jaw. The home made block is so that it is large enough to hold easily. The tool is brought down below the top surface of the gage block with the block to the side of the tool. Then the block is brought against the tool and the tool is raised until the block slides under. Change step to 0.000,1 or 0.001 and advance tool in small increments and slide block in and out to test tool position. This might be useful to you.

In any event the tool setter or part probe are generally simply an on-off switch. There are some with a proportional signal.

I see no need for a math coprocessor to process information on where the machine is when a probe makes contact. You simply move the machine until contact is detected, back off, come back in slowly, stop on contact, read the machine position. Then output the machine position with DPRNT.

Go to the next point to be checked and repeat.

All of this should be possible on your VF3. If you want to calculate the actual contact point on the probe ball then do this in your computer where you have a math coprocessor.

If your definition of DNC is simply RS232 communication, then you need DNC.

If your definition of DNC is drip feed, then you probably do not need to operate in DNC mode.

To do probing you need MACROS.

A spindle part probe would be used for both part checking and setting G54 or whatever for setting your X-Y reference.

.

[lapuser]

Yes the haas has two functions as you describe. The block idea is good and I will try it soon. The real attraction for me in a good probe routine and capability is to detect tool breakage in a large part run before it becomes a major problem, especaily when the machine would be left unattended. I have an xyz indicator and I have used the mill as a cmm with this before but I will tell you it takes forever to do complex geometry and it is never as good as the routines generated by software for that purpose. I dont know if the sales guy is lying about the coproccessor or whether it is really required to do the probing macro-software routines. Yes I have real DNC. Spindle probes also will do reverse engineering and will do parts verification. The workpiece setup while the most common usage is just the tip of the iceberg for what these probes can do. Question for you Gar, do you have any real experience with what I am thinking of doing, or have an idea of where to send me?

[gar]

060113-1212 EST USA

lapuser:

Now that you have brought up broken tool detection that adds a new perspective.

How minute of a tool failure do you want? If it is very gross, like 1/2" or more is lost, then a very simple detector can be used. In fact you could do much better.

Make a block with an Oilite bearing, a spring loaded hardened plunger, and optical or mechanical switch. Mount this somewhere on the machine out of the way. You can probably get reliable 0.010" repeatability or better fairly easily. If the switch is mechanical and is switching low voltage, below 120 V, then do not use silver cadmium oxide contacts. These are typically used in 120 V quality limit switches, and relays. Pure silver not too bad, gold better, sealed (reed switch) or solid state even better.

Learn how to write your own program to do this broken tool detection.

There is a lot more to broken tool detection than "is it broken at the end of a tool use". How often do you need to check the tool during the cutting operation? If you have a large tool with substantial spindle load, then load power is going to be the best technique for a gross failure and quick detection, and this is probably very important.

For actual part measurement you probably want to use commerical equipment with their CNC software, and PC software. As I said before I see no reason that the CNC has to have a math coprocessor. If you simply tell the CNC to go somewhere, and on the way it detects probe contact and stops, then that does not require any macro math operations.

.