What's new
What's new

Haas TL-1 Accuracy issues

CBlair

Diamond
Joined
Sep 23, 2002
Location
Lawrenceville GA USA
I have a TL-1 and although I am generally pleased with the machine I have been chasing a accuracy problem all day. Most of the time I only run parts in lots of 4-6 ea. Today I am trying to cut 40 small shafts. 1" CRS with a 20mm bearing journal, .0004 total tolerance on diameter. I have programed for the middle of the tolerance band and I am roughing with one tool and using another to cut one finish pass to size. However I keep chasing my dimention all over the place from part to part.

This is a very simple part, held tight to the chuck, consistant tool pressure and yet my parts are comeing out with +/- .0006 from nominal. I try and adjust the tool .0001 to get back to the middle and then it will be .0004 in the opposite direction on then next part and then without making any changes back to the other side of the tolerance band.

Does anyone know what the actuall tolerance capability of this machine is supposed to be? I would have to believe that it should be expected to run within +/- .0004 at least.

Charles
 
It sounds like when you are changing to the finishing tool it is not repeating. Have you tried to rgh and finish with the same tool. I know on the manual lathe I run it doesn't take much to throw the tools off. A small chip or a little piece of dirt. Especially when you are trying to hold tenths. I hope this helps.
 
I do not have one of those lathes, but if I am understanding you correctly, it sounds as if you do not have a turret, and I would have to agree with B. Witte.
You will never obtain a tolerance that tight without the aid of an indexing tool turret.
Try using the same tool as stated above for roughing and finishing.
Keep the program as it is(but make the tool offset # the same). Take a measurement of the part prior to the finish pass, and then offset to make up the difference in the actual cut, and the programmed diameter of the final roughing pass. That method should allow you to hold .0001.
Good Luck.
 
I do not have a turret, the TL-1 does not come with one. I am useing an Aloris toolpost and it is in good condition. I wouldnt think that it would move as much as I am seeing? Possible but I am not sure. I have used the same tool for roughing and finishing but I still have to do the same thing because of tool wear. Not as much on CRS as in toolsteel but there is wear none the less. I am trying to eliminate having to measure each and every part every time to maintian tolerance. That is why you use a different tool for finishing, so that it cuts consistantly with less wear.

Of course it may not be possible to do this with this lathe but I would still think I should have better results than I am having. Anyone who actually has one of these lathes want to chime in?

Charles
 
I do not have a turret, the TL-1 does not come with one. I am useing an Aloris toolpost and it is in good condition. I wouldnt think that it would move as much as I am seeing? Possible but I am not sure. {snip}
Anyone who actually has one of these lathes want to chime in?

Charles
I do not have one, but I have looked very intently at buying one. I do know that th tl-1 does not "come with" a turret standard, but it is in fact an option. They do offer one.
Yes...that is why you use a seperate tool for a finish pass...to remove the wear factor....but...
I don't care if your Aloris tool post was gold plated and polished to a shine, it is impossible to hold the kind of tolerance you are wanting when you remove and replace the tool from it's carriage.
Even with a turret, you could not expect that type of repeatability if you were to remove and install your tool between each part. Dunno...maybe I am wrong...but that is why I decided against that type of lathe for production of that sort. You give up way to much on swing capacity when installing the turret
see thread post #s 9&10
and you might pm the guy from post #12 he seems to be able to do so.
 
We use a process called "PUT & TAKE". What we do for simple OD turning is to 1st turn a diameter on all parts to chuck on. Then rough turn all parts....then finish turn all parts. This works best on parts held in collets or freshly turned soft jaws.
Carl
 
Carl,

that is an interesting idea, I dont think it would help make things any faster but it would eleminate the toolchanging as a problem. I was more interested in finding out how accurate the machines are supposed to be. I have had good luck with runing 4-6 parts and adjusting for final size on each one. That just isnt possible when running part quantities over 20, just too much time. I have heard too many people complain about the Haas machines not holding tolerance but this is the first time I have seen this for myself. Of course the toolpost is a dissadvantage but I wouldnt have thought it would be that bad. I have had good luck with these on a manual lathe, dont see how they can change that much just because they are on a CNC machine.

Charles
 
I have to disagree with the Aloris Toolpost comments.
On a manual lathe, I get total repeatablity with a "real" Aloris and Aloris tool holders, and easily less than .001
For other tool post manufactures, I would be suspecious

Heat and dirt would be the two culprits I would look for after confirming that the chuck is securely fastened..
I also would run a routine multiple times with two seperate dial indicators against a previousely machined smooth surface and check repeatability for both operations
Then, and only then would I change out the tool post
Rich
 
+1 what Rich Carlstadt said.

Try attaching a .0001 indicator to the crossslide, then program it to make several rapid moves then feed up to touch the chuck with the indicator. see if the indicator shows the same thing every time.

Could the material be a problem? I have a hard time getting a good finish on CRS - I assume it's 1018 or similar? Could the surface finish be changing your size? Have you run these parts before on other machines and had no problem? What insert material and geometry/chipbreaker are you using to finish? What speeds and feeds?
 
Just taking a 1mm finish pass, only removing about 3mm of material anyway. Spoke with Haas today and they said the machine should keep about .0004 positioning. With a couple of tenths play in the tool post I can see .0006 difference, wish I could get it more repeatable than that but it just may not be possible with this machine.

Charles
 
I would have to agree to check things out with an indicator, assuming the machine is ok, you might try to get both tools in the machine at the same time and avoid any switching issues. It works well to get one upsidedown on the backside.
 
I've had pretty good luck with the Dorian Quadra dex tool post if it's kept clean, .0005 wasn't too tough. You know when some crap gets in it though because it starts to vary also. This tool post would eliminate the possible changes in the wedge pressure on your Aloris.
I agree with Mudflap about the mat'l making it tough to hold that tolerance if it's 1018.
 
Charles,
I think your question is pretty well answered thru all the posts. Repeatability of the machine is .0004, repeatability of the tool post is probly a few tenths at best and then give a tenth or so for the material finish and you are out of tolerance or at the very least way out of the mean.

I run a similar job in my SL-10 and found changing tools was a waste. It took more time to setup a second tool, have the turret move to a safe position, index, then get back into the material then just running using the same tool to rough with an additional pass. Using a good quality CNMG 431 insert I run 40-60 pcs before I need to index insert in a similiar application, 1"daimeter turned for a 20mm x 1-3/8" long Bearing fit. I rough to .02 over then finish with the same tool with a slightly lower feed.
 
Thank you SIM, I have had good success with one tool when doing only a few parts. This was a simple job and I wanted to use it to try a different technique. I have a job coming up with A2 steel and I thought this would be a good way to see if using two different tools would do what I wanted. I dont think this machine is really up to what I want it to do.

I dont expect micron precision but the machine has very few hours on it and I really felt that it should do better. I am sure all manufacturers are a little conservative when they make accuracy claims but I really felt that the machine would be better than it is.

Thats what I get for being cheap...:)


Charles
 
Charles:

Own a Romi M17 CNC lathe. Similar to the TL1...Had no end of similar trouble with tight tollerences when changing tools. I had the lathe fitted with an Aloris CXA post.
Went with a MultiFix 40 position post that a friend loaned me....Problem solved.....Holds .0002 all day with tool changes. The Aloris just would not repeat well enough.
Additional notes, i fitted the Multifix to the cross slide with a shop made riser, used fitted center bolt and 2 locating dowels. Holds the post rigid enough to allow drilling off the post with 1" drills...No twist or shifts..
Cheers Ross
 
Thank you Ross, I remember you mentioning that before when you bought your machine. I almost bought a Multifix but I already had the Aloris and I got a great deal on holders. I dont think I can afford to change out all the tooling for now. I will try and find a better way until I can afford a different machine. I didnt buy this thing expecting super accuracy, I just happen to have several jobs come my way where it was important. Most of what I use the machine for is +/- .005 so it is fine for 99 % of what I need right now.

Post more photos of your work, you get to do the kind of jobs we mortals dream about at night...

Charles
 
i think what you have to remember is that you bought a lathe that was 28,000 new (maybe 40,000 fully equipped). i have a 1984 Hardinge CHNC chucker that will hold .0002 all day long (and cost more in 1984 than the HAAS does now). we have two new TL-1's at school, and they just weren't made to be rigid, or hold tenths. they have their applications, they just aren't small precision machines.
 
I understand you are set in the way you wish to do the job, been there done that.

A few years of additional experience has taught me to just get the job done.


A quick story... when I got my 1st CNC lathe I got a job that looked pretty much like a large Ball Hitch. I could not wait to see my tool cut the finish contour in One Pass. So I set everything up programmed the part, ran the simulator and the roughing looked perfect but I would get an alarm on the finish. I spent the better part of my weekend working on that program with no luck. Well I could do it in two sections, but I wanted to do it with one contour pass. So monday morning I call the machine manufacturer, sent him the program and ask what I am doing wrong, he answers "nothing the program is good" he ran it on a new machine. I tell him mine alarms out and he goes and checks it on an older machine and tells me "it alarms out" it's a program glitch and I should run it in two sections.
 
Sim,

No one is set in any way, the job is done, new work is in progress and next time it comes up I will try something else. Just looking for how other people who use this same machine have delt with this issue and judgeing by the responses...no one on this forum uses one of these machines.

All the ideas presented here have value and I have listened to them all. Time and experiance with this machine will offer me a realistic view of its limitations. For now it is all I have and will have to suffice as is. Thank you for taking the time to share your experiances, I do appreciate it.

Now for cutting through 5" of stainless steel castings on a 24 yr old machining center, and not pulling my hair out in the process. :)

Charles
 








 
Back
Top