What's new
What's new

Haas Mini Mill, how to realign X axis for squareness error?

Joined
Aug 30, 2022
Hi guys, I have a slight squareness error on my Haas Mini Mill, identified by a ballbar test and confirmed with a square. The X is 40 microns per meter out of square with Y. I was thinking of loosening up the X mounting to the Y blocks and twisting the axis slightly to bump it into perfect alignment, but I'm unsure of whether this amount of error would require me to loosen the ballnut on Y to allow the adjustment, or how difficult this process is in general. Has anyone done this before and is it a PITA? Is this error even worth correcting?
BTW, machine is fully levelled!
Thanks
 
6 tenths over the travel of the machine. I don't know that it was designed to do much better than this. It is a weldment base garage machine.
 
There is an error map for position compensation. I just had Productivity Quality do a laser and ball-bar calibration on my CM-1, and they dialed in it a few microns.
 
This guy puts a ballbar on a Haas, and is now having anxiety. This is like my racing days where the first thing you do is throw the speedo out the window. You don't want to know! Just go make parts man! At the end of the day, you can likely improve the problem, but it will still be a Haas.
 
This guy puts a ballbar on a Haas, and is now having anxiety. This is like my racing days where the first thing you do is throw the speedo out the window. You don't want to know! Just go make parts man! At the end of the day, you can likely improve the problem, but it will still be a Haas.
As long as it's within spec I'm happy - the 40 mics over a meter is really nothing considering the machining area is 400MM, but of course less would be better 😅 Just wanted to get a general consensus as even though it's a Haas, it's a lot of money! I've seen sub 20 micron results from Okuma and similar machines... but you're talking way more money for those, so not sure what to expect with the Haas.
 
As long as it's within spec I'm happy - the 40 mics over a meter is really nothing considering the machining area is 400MM, but of course less would be better 😅 Just wanted to get a general consensus as even though it's a Haas, it's a lot of money! I've seen sub 20 micron results from Okuma and similar machines... but you're talking way more money for those, so not sure what to expect with the Haas.
The "spec" is whatever Haas feels like they can BS through. If you call them out, it will never be their machine, it will be the floor, or possibly the air quality causing it.

I know I won't lose any sleep tonight over those 40 mics. I have just learned to stay well away from those rigs as they always bring problems and BS that I don't care to deal with anymore.
 
Do you need more accuracy? If so, hire a calibration company to laser and ballbar it. Probably $1500 and 6 hours down time; I negotiated one time in with the purchase of the machine, and just got around to it a couple weeks ago. There wasn't much error, but they cut it about in half. If it won't make a significant difference for your parts then don't sweat it. I did mine because I'm making mold blocks for medical devices, and if nothing else my clients will feel better knowing I'm making them as tight as I reasonably can.
 
Do you need more accuracy? If so, hire a calibration company to laser and ballbar it. Probably $1500 and 6 hours down time; I negotiated one time in with the purchase of the machine, and just got around to it a couple weeks ago. There wasn't much error, but they cut it about in half. If it won't make a significant difference for your parts then don't sweat it. I did mine because I'm making mold blocks for medical devices, and if nothing else my clients will feel better knowing I'm making them as tight as I reasonably can.
Honestly… no… 16 mics across the travel range is plenty good for me. It’s more of a curiosity and, well, an OCD thing. The other haas machines (VF2 etc) were the same or worse, so I assume it’s perfectly normal. Honestly I’m quite pleased with it in isolation, it’s only that I’ve seen better ballbar test results from other machines. That said — much more expensive machines! I think I found a thread on a user with several Fanuc machines which the same squareness error as me after a ballbar. I think all of them were 0.04mm per meter or more.
 
The ISO 230-1 requirement for XY squareness of any CNC machine up to 2 meter travel is 20 microns per 500 mm.
The top of the line Mazak's Variaxis 730 machine Certificate of Accuracy requirement is 10 microns per 500 mm. So let's be realistic. One of course can make effort to achieve better result. Unfortunately, the machine like the one in subject will not keep this alignment quality for more then one working hour.

Stefan
Cogito Ergo Sum
 
The ISO 230-1 requirement for XY squareness of any CNC machine up to 2 meter travel is 20 microns per 500 mm.
The top of the line Mazak's Variaxis 730 machine Certificate of Accuracy requirement is 10 microns per 500 mm. So let's be realistic. One of course can make effort to achieve better result. Unfortunately, the machine like the one in subject will not keep this alignment quality for more then one working hour.

Stefan
Cogito Ergo Sum
Hi Stefan, thanks for the info. So my machine is within the ISO spec, which is good. What makes you say that it will lose alignment? As I understand it, Haas run lots of test files on the machine prior to shipping it out, so it's already performed cuts.
 
I know this is basic but you did not mention if the floor and footing for the machine is up to par and if the machine is level. You might want to check this if you have not done so already.

We had nice machines at an older site but one really old conversational mill was kept for machinists to use. It was a 40 taper mill but the XY travels were huge. Anyway, we discovered the machine was massively out of square due to an extreme leveling issue. The error range was .0015"-.0025" over 40" but a tech was able to bring it back in.
 
I know this is basic but you did not mention if the floor and footing for the machine is up to par and if the machine is level. You might want to check this if you have not done so already.

We had nice machines at an older site but one really old conversational mill was kept for machinists to use. It was a 40 taper mill but the XY travels were huge. Anyway, we discovered the machine was massively out of square due to an extreme leveling issue. The error range was .0015"-.0025" over 40" but a tech was able to bring it back in.
Hi Areo, thanks for your reply. The machine seems to be level to within 0.05MM per meter according to my level. The Haas tech just set the level up and did say that he'd be able to tweak the alignment with the level if it was a larger machine.
 
The OP’s question really highlights one of the differences between a budget machine and premium machines. For Haas it seems that a new machine right at the top of ISO allowed geometry error is fine, ship it.

When I was at Makino years ago for FSE training I got to work with a couple guys setting the column geometry on some HMCs. The company geometric accuracy requirement was half the ISO allowed error. When we reached that accuracy, I was surprised when the column was pulled again for further work. On asking why we were going to scrape a bit more, the assembler said he always worked to half the company spec. That way his work would always be within the company spec for a long time and within ISO spec for even longer.
 
Also make sure your square isn't misleading you. My square is off by 3 tenths over 12 inches.
Thanks, yes, I measured with the square flipped over and replicated the same squareness error within 2 microns or so.
Question guys, I ran my very accurate level (0.02MM per 100CM) along the y axis and it appears there is a very slightly twist in the rails to the tune of 0.03MM per 100CM along the entire Y axis, is it worth adjusting the levelling feet to compensate? (if this will have any effect on the casting since it's a steel base machine) For reference it appears the front left foot is low. For the squareness, the X appears to be further forward on the right, and further back on the left. Would a lower left front foot cause this?
 
Last edited:








 
Back
Top