Adding precicion to the IH mill

[M4RK_23]

I was wondering it the Ih mill could me made more accurate. EX use ground ball screws and double ball nuts. I would like something with the precision of the tormach but with a larger envelope. If not possible I guess I'll end up with the tormach. Thanks -Mark

[pstockley]

I don't see why not. I am of the opinion that backlash is the most important thing to minimize first. To do this you need a good set of ballscrew duplex bearings (not cheap). A good set of rolled ballscews e.g. hiwin (< 0.001 over 12") would be ok if you performed some compensation mapping. The only problem is you need some very accurate way to measure the requested/actual movement. Usually done with a laser. Alternatively you could use a really good set of glass DRO scales. Otherwise you could just go for ground ballscrews. I am guessing these would set you back $1500 to $2000.

The other thing I would want to do is increase the top end speed to at least 4500rpm and fit a VFD for variable speed. This will probably require changing the gear head over to belt drive. I think IH has a kit coming out for this.

Best bet is to talk to IH, I think it can be done but the question is how much is it going to cost and how much work will you have to do.

[wizard]

I haven't seen a Toramch, nor have I tried to check it for precision. I do wonder how you have comed to the conclusion that the machine is precise?

I have seen the IH mill at the model engineering shows. It is a nice machine --- FOR THE MONEY! I do beleive it is misleading to believe that all you need for precision is a nice set of ball screws.

Taking in to account experience I've had servicing machinery (mostly high precision lathes) I'd have to point out that fit up of your mechanical components is extremely important. This universally requires manual work to scrap the ways in and at the same time remove machining imperfections. The better the fit up the better repeatability, plus well fitted ways and gibbs can help control vibrations and movements due to tool reactions. I suspect that much could be done with the IH machine with respect to the fit up.

The second issue is of course leadscrews. Of course the better the quality the beter the results, but there are issues outside of leadscrew quality that impacts machine performance and can be an issue on cheap machines. One consideration is that the leadscrew when installed must be parallel with the way both horizontally and vertically. If not stress is introduced which impacts repeatability. Related to theis is the leadscrew bearings, the better these are the better the mounting arraingement that has to be implemented. If the bearing cell when installed forces a torque on the bearings you will have positioning problems again. So retro fitting a leadscrew requires care if you expect precision.

Thus far we already have two independant issues that affect accuracy that need to be addressed when the conversion is done. There is a lot of potential in this machine so I'd take that as something to consider.

Now that doesn't even take into account the controls. There are many options here so getting into a discussion probably is not of much help. A well designed stepper system might get the precision you want, though a servo system might be smarter.

I geuss what I wanted to say is that precision requires effort. Tormach hopefully puts in that effort for you. A CNC conversion of an IH mill requires more effort on your part. Still from what I've seen everything is there to get what you want.

I geuss the only other thing to note is that precision is nice but what really makes a machine a production machine is repeatability. If you can reproduce the same part a 1000 times all good and in tolerance, that is much better than producing one part that is precisely on target one cycle and way out the next. Repeatability and precision are two different things and each has value, for production work you can't dismiss repeatbility.

Thanks
dave

[Mortgaged]

Dave you are 100% correct on the scraping being the most effective precision upgrade. Precision ground ballscrews won't do anything to compensate for the fit and condition of the ways and quill, table flatness, z-axis perpendicularity, etc.

If the Tormach and IH use the same ballscrews and encoders you can expect to get the same locating repeatability with the same programming. How precise that is will depend on the type of ballscrew, encoder resolution and machine programming (look ahead, decel, dwell, etc.). Again, with the same level of precision, the results should be nearly identical between the two machines. Generally, the most accurate industrial machines use glass scales as the feedback loop to the controller and don't rely on an encoder due to rotational errors.

[pstockley]

I don't disagree with the comments above. However, I was assuming that you have to generally live with the inherent accuracy of the machine as delivered. Its true you can fix just about anything with remachining and/or scraping. However, its not realistic to expect that anyone but a skilled professional can make the necessary corrections. Its one thing to scrape a surface flat, a totally different matter to correct dovetails. I've done enough scraping to know that you can get yourself in real trouble very easily.

I agree that repeatability is in some ways more important than accuracy. There is nothing that kills repeatability more on a CNC mill than backlash. Due to cutting forces and slight variations in toolpath the results can be very unpredicatable. Plus if you use Mach, backlash compensation doesn't work in CV mode which makes it pretty useless. On a manual mill you can work around backlash pretty easily, not so on a CNC mill.

Regarding the Tormach, it comes with a signed inspection certificate
http://www.tormach.com/document_libr...C_addendum.pdf

My specs were significantly better than the minimum requirements above. So while its not even comparable to a low end VMC, it isn't too bad.

[wizard]

Originally Posted by pstockley I don't disagree with the comments above. However, I was assuming that you have to generally live with the inherent accuracy of the machine as delivered.

Well you have to live with a machine that meets the specs that where quoted in the sales documentation. That doesn't imply that all of the potential is realized in the machine.

Its true you can fix just about anything with remachining and/or scraping. However, its not realistic to expect that anyone but a skilled professional can make the necessary corrections. Its one thing to scrape a surface flat, a totally different matter to correct dovetails. I've done enough scraping to know that you can get yourself in real trouble very easily.

I have to disagre to an extent, each side of a dovetail is a flat surface. Scraping wise there is nothing difficult about dovetails. Keeping the geometry right in your head can be an issue for some, but a pencil and paper often help here.

On the other hand scraping does require that you have the associated tooling and time. If you don't there are machine tool builders that do have both. Of courses once you think about hiring somebody for such work on a new machine you need to think about spending a little more $$$ for the initial purchase.

I agree that repeatability is in some ways more important than accuracy. There is nothing that kills repeatability more on a CNC mill than backlash.

I have to disagree here. Having spent some time on machines with precision ball screws with one nut and preload accomplished by tight fit of the balls, there is much beyond the lead screw/nut that can impact repeatbility. Of course a nut that chucks back and forth on the leadscrew is no good either but much of my repair work addressing repeatability issues involved fixing things other than the leadscrew.

Due to cutting forces and slight variations in toolpath the results can be very unpredicatable. Plus if you use Mach, backlash compensation doesn't work in CV mode which makes it pretty useless. On a manual mill you can work around backlash pretty easily, not so on a CNC mill.

To some extent you can work aroung backlash on a CNC machine. Backlash never goes completely away.

Regarding the Tormach, it comes with a signed inspection certificate http://www.tormach.com/document_libr...C_addendum.pdf My specs were significantly better than the minimum requirements above. So while its not even comparable to a low end VMC, it isn't too bad.

Unfortunately Tormachs web site doesn't do much to promote the machine. At least with respect to photography. From the standpoint of machine tools it is often best to validate that the machine does what you want. The vendors goal is to make sales and hopefully happy customers, that doesn't mena though that they always understand what you want 100% of the time.

Thanks
Dave

[pstockley]

That doesn't imply that all of the potential is realized in the machine.

In theory true, but at some point it becomes impractical due to either the time, cost or skill involved.

I have to disagre to an extent, each side of a dovetail is a flat surface.

That may be true but if you have one that is out in two planes it isn't trivial to fix. I just don't want to give the impression to any novices that it is easy to fix all your problems by scraping.

I have to disagree here. Having spent some time on machines with precision ball screws with one nut and preload accomplished by tight fit of the balls, there is much beyond the lead screw/nut that can impact repeatbility.

Agreed there are a lot of factors. However as stated you are assuming decent ballscrews. A lot of people buy really cheap rolled ballscrews with no or incorrect preload adjustment. End result is accuracy only good enough for woodworking! Plus as I stated you cannot really use software backlash compensation unlike on industrial controls.

Unfortunately Tormachs web site doesn't do much to promote the machine. At least with respect to photography. From the standpoint of machine tools it is often best to validate that the machine does what you want. The vendors goal is to make sales and hopefully happy customers, that doesn't mena though that they always understand what you want 100% of the time.

Don't really understand what a photo is going to tell you. The tormach site from my point of view has more indepth info for a low end tool than any other I have seen.

[wizard]

Originally Posted by pstockley In theory true, but at some point it becomes impractical due to either the time, cost or skill involved.

Absolutely true. The issue with the imported chinese stuff is that impracticality is a long ways off. That is due to the rehter lax workmanship in many of the lower cost tools. Often the castings are good enough to support the tweaking can lead to better performance. Often there is a bigger pottential for performance increase in these imported machines then one might get with a new industrial machine.
[/quote]


That may be true but if you have one that is out in two planes it isn't trivial to fix. I just don't want to give the impression to any novices that it is easy to fix all your problems by scraping.
[/quote]
Well it may require a bit of effort to fix a badly done dovetail, but it is not impossible even for a beginer. The issue is if you have a dovetail that is out of whack in two planes as you suggest then there is no sense whatever to putting ball scfrews in the machine. Ball screws can not over come a machine that is not right mechanically in its bearing surfaces.

Agreed there are a lot of factors. However as stated you are assuming decent ballscrews. A lot of people buy really cheap rolled ballscrews with no or incorrect preload adjustment. End result is accuracy only good enough for woodworking! Plus as I stated you cannot really use software backlash compensation unlike on industrial controls.

Last I knew there are not many CNC controls that don't have some sort of backlash comp. Granted that doesn't make up for low backlash screws but I think you might be over stating the impact a little back lash in the dirve system might have. It really depends on expected usage, even with metals, backlash might mean nothing.

Don't really understand what a photo is going to tell you. The tormach site from my point of view has more indepth info for a low end tool than any other I have seen.

It is one thing for somebody to write down a bunch of specs it is another thing to see how something is put together. Plus workmanship is often a visible indicator of what a machine can do over the long run. It is sort of liek buying a used machine, the ownder may claim that it is holding up fine in produciton, but upon viewing you might realize that the concept of maintenance is un heard of in this plant.

Maybe it is me, but there is nothing like seeing the machine in person. It is oen of the reasons that I'm somewhat positive about the IH mill. If you can't see it in person then the next best thing is a set of high resolution pictures. So the Tormach spec sheets are nice, but that is only part of the equation.

Thanks
dave

[MikeAber]

Perfection vs ability, skill, and cost.

Mike

[phil burman]

The Tormach comes with a comprehensive test certificate. What does IH commit to? If you can tell the performance of a machine just by looking at it, either in person or even more incredible in a hi res photo, then you are a better man than most on this forum.

Best Regards
Phil

Originally Posted by wizard Maybe it is me, but there is nothing like seeing the machine in person. It is oen of the reasons that I'm somewhat positive about the IH mill. If you can't see it in person then the next best thing is a set of high resolution pictures. So the Tormach spec sheets are nice, but that is only part of the equation. Thanks dave

[wizard]

Yes they do but that tells you nothing about dynamic performance. Many a CNC machine can slew back and forth and get excellent repeatability while doing nothing. More important is how the unit holds up while actually doing real work. Will the ways be stiff enough to maintain repeatbility and is there enough mass to provide for good surface finish?

A person can gain a certain sense about how well something is assembled and designed by looking at pictures. It is the next best thing to being there. This is certainly not a substitute for measured parameters but such parameters don't mean alot to me unless I'm there observing the measurements. That may be pretty course but I spent a few years working in a machine tool rebuild operation and have seen many a questionable measurement technique.

Now I'm not saying that the IH machine is the best thing since slice bread. What I'm saying is that I've seen the unit online and at NAMES and can see that it is a nicely done CNC conversion. Is it an optimal solution, nope but improvements have been discussed here and on IH's web site.

So until I can see the Tormach machine first hand or get a better impression from some photos I can't really support it. That may very well change if I start to hear alot of comments from people using the machine with extensive machine tool experience and good metrology methods. As it is now though I have little to go on with respect to the Tormach machine.

Thanks
Dave

Originally Posted by phil burman The Tormach comes with a comprehensive test certificate. What does IH commit to? If you can tell the performance of a machine just by looking at it, either in person or even more incredible in a hi res photo, then you are a better man than most on this forum. Best Regards Phil

[phil burman]

Hi Dave,

I don't think you looked at the Tormach sample test certificate. It includes an actual test piece so dynamic performance is included. It also includes metrology requirements/standards. The test are very extensive. Of course it could all be one big lie, but then that's a whole different ball game.

For me, looking at a display example of a machine at a show, in a showroom or in hi res photos is not a good way to judge the quality of the machine you are going to be taking out of the box. The IH website includes extensive details on lapping the ways, which I have to admit is untypically honest compared to some sellers, but it also tell me a lot about what to expect out of the box. Why does the IH website not have any details regarding precision, other than a theoretical resolution? I can only take this to mean that the precision guaranteed to be whatever you find. If you where a commercial buyer looking for a mill what would you be most concerned about, it looks good in the showroom or it has extensively documented performance criteria and verification.

To be fair I should point out that I have ordered a Tormach 1100. I didn't have the opportunity to see either in the flesh (I live in Norway) so for me comparing the technical content on each website resulted in no contest. If your have no choice but to laydown 6 or 7,000 USD sight unseen then the Tormach is a much more convincing argument.

Regards
Phil

Originally Posted by wizard Yes they do but that tells you nothing about dynamic performance. Many a CNC machine can slew back and forth and get excellent repeatability while doing nothing. More important is how the unit holds up while actually doing real work. Will the ways be stiff enough to maintain repeatbility and is there enough mass to provide for good surface finish? A person can gain a certain sense about how well something is assembled and designed by looking at pictures. It is the next best thing to being there. This is certainly not a substitute for measured parameters but such parameters don't mean alot to me unless I'm there observing the measurements. That may be pretty course but I spent a few years working in a machine tool rebuild operation and have seen many a questionable measurement technique. Now I'm not saying that the IH machine is the best thing since slice bread. What I'm saying is that I've seen the unit online and at NAMES and can see that it is a nicely done CNC conversion. Is it an optimal solution, nope but improvements have been discussed here and on IH's web site. So until I can see the Tormach machine first hand or get a better impression from some photos I can't really support it. That may very well change if I start to hear alot of comments from people using the machine with extensive machine tool experience and good metrology methods. As it is now though I have little to go on with respect to the Tormach machine. Thanks Dave