Calibrating a Haas Probe and Tool Setter on a CHC Controller (Pre NGC)
ฝัง
- เผยแพร่เมื่อ 20 ส.ค. 2022
- I purchased a WIPS (Wireless Intuitive Probing System) probe and tool setter from Haas. Because my machine is a 2009 machine, and therefore has the CHC (Classic Haas Controller), there is a lot less information out there about calibrating the probe. Plus, because of the tight installation, they had to angle it at 45 degrees.
VQC Instructions: www.haascnc.com/service/troub...
Tool Probe Mount: exactengineeringshop.com/coll...
One quick tip:
When you need to measure any distance you can set 0 (zero) the operator counter by press the "ORIGIN" button.
I hope this helps.
Those probes are going to make life so much easier. Plenty of other stuff you can do with them as well. Great job calibration them !
Thanks. I'm looking forward to seeing what else I can do with it.
how much did it end up costing to add this altogether?
The probe was $5,500 and installation came to $800, which included a travel fee from the HFO.
Nice to see our tool probe mount in action. Clever program to decouple the tilt axis during leveling which is always a time consuming endeavor. BTW, we are now developing a more user friendly and tip/tilt system that eliminates the finicky grub screws and the push/pull adjusters.
Such a helpful video. Thank you so much!
As someone who just got a haas with a CHC controller, all I want to say is THANK YOU. You’re quickly becoming the “tip of the day” for classic controllers.
I was curious if your HAAS tech came out, did they download anything to your control to activate the tool setter/probe menus on your control? (I’m missing those menus after the install & I’m guessing the tech skipped a step somewhere)
Keep up the great work, super helpful and very well done!
Thanks. The techs enable the required features using some codes. Then installed a set of macros from a thumb drive.
I bumped my tool steer with a cooling bucket and the tool setter loosened upside down. Do you have a setup to align it? It is loose and doesn`t tighten straight and flat.
Really nice probe. I wondered whether you introduced errors when measuring the probe Z length with the tool setter. The probe has to retract a certain distance to open the internal circuit and send the signal. When you use the tool setter, it doesn't account for that retraction. Usually, people log the tool offset of an end mill, face a surface that is the known Z0, and then probe it to get the probe length.
I was wondering about that as well, but that's the only approach the NGC controller uses to measure the probe length. And I verified that the results are within 0.001 inches or better. So Renishaw must have thought that through.
@@JohnSL I often see the facing surface approach when checking other tutorials with a Renishaw probe. Grimsmo also made a video about it. Nevertheless, if it works for you, do not change it. In the end, the tests you run are the final judge.
Agreed. As long as I'm getting the accuracy/repeatability that I need (which I seem to be), I'm happy.
Hey John. As always great job :)
1 quick tip. Sometimes I notice the audio in this (and other videos) switching from stereo to mono. Is it just me?
Thanks for letting me know. I checked and it seems that my wireless microphone switched to mono mode. It's done that to be before, which is really irritating. I will make sure I use headphones to check the audio in the future.
When you get the 2mm probe stylus, do yourself a favor and get the really short (10mm) stem version. Either Renishaw A-5003-2932, or Q-Mark TM4-2021. I've was running the 40mm stem 2mm ball and I couldn't figure out why I was chasing dimensions all over the place. Turns out, those long stems are extremely delicate; even high-speed use can screw them up with relative ease. The 2mm ball is great! But the stubby stem is 100% necessary in my book.
Thank you for the suggestion. That makes a lot of sense. I'll have to do some testing to makes sure I'm getting good measurements.
@@JohnSL Also, the factory Haas probe calibration routine is strange. I'm the sales/apps guy for Yamazen (Brother Speedios) in the area, and we calibrate the spindle probes in Z off of the table. Touching it off the OTS inherently has errors (two switches touching? think about it). I might find an alternative length touch method.
I was wondering about that. Spring against spring seemed like it wouldn't be precise. So far it looks plenty good enough, though.
If you hadn't said so or written it in the description, I would have guessed that's a NGC. The screens look the same to me. I thought I had the CHC control on my 2006 VF1 but my screen is smaller (but still color) and the different sections of any one screen are not separated like the NGC (with the rounded corner rectangle style). What would the controller prior to yours be? I'm a new operator to CNC mills.
Edit: Just took a look at another vid with a NGC. I guess they are different, but I would still like to know what my controller is. Do you know?
CHC dates back to around 2009, which is when I machine was made. I haven't found a common description for even older controllers.
I think the "period" you are referring to after the F10 is really a required decimal point.
Yes, I guess I didn't explain that we'll. The decimal point is required to makes it a floating point number instead of an integer. Apparently, it treats integers as "machinists" units, which in this case would be "tenths." That means it converted 10 to 0.001 ipm. Pretty slow.
YAY
explain all . 5 AXIS TERMS
G68.2 P0
G54.4
G53.1
G43.4
G43.5
G254 FOR HAAS
G234 FOR HAAS
Interesting but way out of my league.
And this is why closed systems (like HAAS) suck - you have a tool that's crippled by default (no macros) unless you pay, then you pay a bunch of money for them to come do work for you that's not very accurate that you can do better yourself, and on top of that they didn't even install the macros properly. Now you're stuck waiting for them to come fix it because they want to charge you again and only allow access to their walled garden to their own people. Open / OpenSource systems are so much better. LinuxCNC or DIE!!!
As it turns out, the macros were fine--it was my own doing. I had the door fully open for filming, and the IR signals were getting blocked when the head came down. When I closed the door and tried it again, it worked each and every time. It seems the IR signals need to bounce off the inside of the door to be reliable. And they did send me the macros within business hours of me asking--no additional charge. Additionally, the accuracy of what they'd done probably would have been fine, but I wanted to learn how to do it myself and go do the best I could.
closed system suck? You're a hobbyist, in the real world - everything is "closed".