Sorry, I am going to be a little long-winded here but I really would like someone's input on what I am seeing. I have sent the following information to Aaron at IH but I had hope that others might add their opinion. The main problem is that I am having problems getting reliable cuts in a reasonable fashion. E.g. I move to cut a couple of thous and find I had cut 7 or so. It seems that a lot of this error comes from the deflection of the quill and head. What I mean by "reasonable" is that I think I should be able to move the quill either by loosening the quill lock, moving, and tightening or moving the quill with the lock snug without undue force and maintain position. However, it seems to maintain position I must keep the quill really tight and/or move in the Z by raising the head. However, the gib needs to be so tight in the head that it sticks and falls while moving down making it dangerous to work pieces. Please read on to see what I did and what I saw.

Following the instructions for tramming on IH's website I mounted an 8" ground square on the table and place the point of a dial indicator against the blade. I ran the head up and down indicating the error of the head's movement in the X and Y dimensions. I also did the same with the quill and found a lot of runout in the quill unless I tightened the quill up to nearly the point that it cannot be moved with the microdial along. With my hands alone I can deflect the quill .006 in some cases. I can't imagine how much an endmill would deflect the quill. As it is now unless the quill is tightened before indicating the edge of a piece and left tight there is a lot of error present. (I included the raw measurements below.) The head appears to be fairly square with the table. With some squaring of the head I can improve the situation some as seen in the tight column. I guess what I am concerned most about is that I need to keep the quill so tight to maintain position and how easily the quill and head deflect with just hand pressure.

Also I have a problem with the head movement. In order to maintain position of the head I had to seriously tighten the gib. In fact I had to shorten the gib screw in order to have the travel necessary before bottoming the screw out in the threads. After doing this the head has less deflection (the numbers here measured after doing this), the head was more square with the table, and tightening the gib locking screws did not deflect the head. However, the head sticks while moving up and down. It is not such a big deal that it sticks moving up - a little extra muscle is all it takes, but moving the head down is a little different. The head sticks but I am still cranking and then all of a sudden the head drops what seems like a lot. I was able to take some of this out by inserting about 3/8" of spacers under the cap holding the z-axis on top of the column but there is still some head falling happening. I don't know if more spacers will help. I added spacers after noticing that when change directions cranking the Z-axis handle would move the entire head leadscrew up. The bevel gear on the leadscrew did no move and the key kept the screw and gear connected. The head would ultimately fall the distance the lead screw was able to move in the bevel gear. I have lapped the Z-axis.

To summarize my questions:

1) Is it reasonable to be able to deflect the head .005 or the quill with a person's hands alone?

2) Is it reasonable to have to tighten the quill to maintain position so much that it is barely movable?

3) What can be done to keep the head tight but not fall while moving down as it does now? Does this mill need to have the Z-axis modification that IH shows on their web site where the head is more rigidly attached to the Z-axis nut? Is a ballscrew needed?

Raw data:

The measurements read off the indicator while moving it along the square with various tension on the quill and the deflection at the top and bottom of the 4.5" travel created by pushing with only my hand:

Loose Mild Tight
Y-axis Travel .005 .003 .002
Deflec Bot .006 .001 .0005
Deflac Top .006 .001 .0005

X-axis Travel .007 .004 .001
Deflec Bot .004 .003 .0005
Deflec Top .002 .002 .001

An additional .005 deflection in the X comes from simply pressing on the head, which drops to .003 after seriously tightening the screws and this is after tightening the gibs.

7.5" travel of the head has runout of .001 in the X-axis and .000 in the Y-axis.

Hmm... why is after posting I always seem to find a thread that covers my question? It looks like my numbers are right in line with what others have seen regarding the quill quality. Guess I will lock the quill down and work on removing the sticky points in the column. This weekend I will pull the head off and see if it needs the modification that Aaron has on his website and rig up a counter balance system. I plan to go cheap and use pulleys and weights. Do you think this will cleanup the sticking and play in the movement of the head?

Thanks.

I'm doing fine with mine if I just lock the quill. I did measure a lot of slop in the quill. My Z-Axis on the column doesn't stick. I have lubed it carefully and adjusted the gibbs. I understand some others have seen sticking problems. I also know Aaron's modification is HIGHLY recommended.

I can't imagine why counterweights wouldn't be an awesome idea in any event. Do take care where you attach them as half the battle is that the head cantilevers. You want to not only relieve the weight, but also the cantilevering.

Interestingly, Aaron on his site suggests that the gibbs on these machines be very seldom adjusted (4 times a year with the seasons), that they be adjusted as tight as can be and still get smooth travel, and that they never be locked, as locking will change your position, especially on the Z-axis.

I had been locking mine to reduce rigidity, but will try it Aaron's way when I get back to playing with the mill again soon.

Best,

BW

Bob, I assume you meant you were locking your gibs to INCREASE regidity ;)

Unless I really tightened the gibs screw there was a lot of movement in the head (at least .005 in one axis) when I locked the gib. Surely if locking the gib is moving the head then maching forces would as well.

This weekend I plan to pull the head off and get a better idea of what is actually going on. The current idea is, after checking if the head modification would help and checking the acme screw in general, to hang some weight off the bottom of the head slide. Since I am using my machine in a manual mode I am not worried about motors driving the extra weight. Since I am using the acme screws there is quite a bit of backlash and since I had to tighten the gib down pretty snug to take up the head deflection the head sticks in places. It would seem that the addition of maybe 50 lbs hung off the bottom of the slide would help to overcome the sticking and keep the head firmly down against the acme screw and lock the gib when I want the Z to stay put. It would seem that unless the jib is locked that without a ballscrew and motor keeping it in place the head could wander down. Anyway, if the extra weight does not do the trick then on to trying a counterbalancing system.

Regarding the quill - I going to leave it up and locked down snug - my HF drill press is actually better. For z-axis measurement I guess I will mount a caliper on the column. I don't want to spend the cash on a DRO for this mill until I have more experience with it to know if it's justified.

I can understand now why some call the import mill/drills "mill kits" :tired: I am sure it will be nice after all the kinks are worked out however.

hanging the weight from the bottom of your z slide will make it worse, hang the weight behind the column using cables and pullys up and over. mounts for cables will have to be to hollow portion of box not entering oil chamber, and as far to front as possible, and definately get the z saddle modification done by aaron. better concider at least a little bit of lapping too to insure a smooth slide.

Bob, I assume you meant you were locking your gibs to INCREASE regidity ;)


I guess you could view it as increasing rigidity, and that was why I started out doing it. In the end, I realized it was going to flop around to much to accurately align to the parts until I locked. So I run it down to depth, lock it, remeasure Z, relock, yada, yada, until it is in the right place. It works, and you get used to how to tickle it into position pretty quickly. A Z-Axis DRO is hugely useful for this. I made up a real cheap set of DRO's and have found it well worth it:

http://www.thewarfields.com/MTMillDRO.htm

Cruiser is right. You need to counterweight and reduce the weight, not add more to it. You also need to use pulleys or offsets if using gas struts so you take the weight off at the center of mass of the head (center along spindle axis is probably close enough, even if done at top or bottom). You want to really counteract that cantilevering. 275lbs is a lot of lead hanging out there, so I am more a fan of the gas spring approach:

http://www.thewarfields.com/img/Toys/MachineTools/OthersProjects/IHGasStrut.jpg

Thomas Powell worked out that design, details on his site:

http://imageevent.com/tppjr/rf45dovetailmill/zaxiscounterweight

I'd be real tempted to try to attach the cables a little more towards the front of the mill myself. A clever man might even hide some of this gimmicry inside the voluminous column--gas springs are small.

Another fella on the 'Zone just stuck a gas spring on either side and attached one end to top of column and the other to bottom side of head. That was real clean, and worked well. It's one disadvantage is that because the gas springs are angled, their force changes as you move the head up and down. He says its not a problem.

Best,

BW

Regarding adding weight - I knew I would hear from someone about doing that :) What seems to happen now is once the backlask of the screw (and related slop) was been taken up the screw gives the head a little push and down the head comes. It would seem that a slight pull at the bottom of the slide, so close to the column, would be enough to keep the head down against the screw and not add to the cantilevering but instead overcome the friction. This will not work? Lapping has already been done though perhaps not enough? I can't really say that lapping helped much more than apparences and I think others have mentioned that biggest benefits of lapping are to CNC conversions while mine is manual.

Bob, I was looking at exactly those scales on eBay and I am glad that you show how to mount them. That helps a lot to know they mount welll. Also, I have seen the counterbalance pictures you referred to. The gas struts on the side of the head is a nice looking system but I was surprised over the expensive of gas struts. I will give it some more thought.

Thanks

I think that the little push, followed by slip, happens because of all that cantilevered weight. Think about how it tries to cock the head in the slideways. Aaron actually uses an even longer slideway so it has more leverage against that in his latest incarnation. If you add weight, you are just causing the forces to be even less equally distributed over the slide.

You want to adjust the gibs carefully so there is no room for "cocking" (i.e. make it hard for the head to tilt down in front), generously lube everything, and potentially counterweight to further reduce the cantilevering.

Mine never slips in little jumps, and after I put way lube and lubed the screws, I noticed if I didn't keep the gibs tight, it would slowly and smoothly slide down with vibration. So, from this, I know smooth travels are possible, at least in one case! LOL

Call Aaron too. I've seen other reports of this from time to time and given similar advice. He almost certainly knows even more tricks.

Best,

BW

I don't think I'm telling tales out of school to say that I very recently had a short conversation with Aaron re the head slide mod and he advised against it if the machine was to stay manual.

The reasoning is that if you solidly mount that area and there is any slop elsewhere at all with the Acme lead screw you'll bind up. CNC folk are changing over to ball screws so alignment isn't a problem, therefore the head needs to be as rigid as possible.

Wildcat, I'm having the same symptoms as you and, nearly as we speak, I'm lapping the column as step one to the grip/slip problem.

However, if we study the design closely we must arrive at the undeniable conclusion.... we are screwed! (Shoulda bought the Bridgeport (chair))

I say this with no rancor at all. It's just the facts.

But part of the fun is trying to think through these situations:idea:
For now I'm just trying to get it smooth (even if the head is a bear to move with the gib tightened) and then work within the limits and capabilities of the machine.
(Yet all the while plotting and pondering a fix and praying the plagirists prayer that someone beats me to it!;) )

More ideas then!!

Bob, Cruiser, and QSIMDO, thanks for the suggestions and warnings - I will certainly pay attention. When I got home tonight Aaron had left a message and hopefully we will get in touch tomorrow. If I remember, I will ask him about perhaps using his mechanicals to convert the Z-axis to motor drive if that would do the trick. Well, got to go open up some boxes of goodies from Enco. :banana:

Anyone have any suggestions for a gas strut that has a longer travel than 12"? This picutre that Bob posted looks like a really nice solution (compared with free weight) but limiting counterbalanced travel to 12" is the not great especially when yo consider 4-5" of that travel is taken up if using a vise or rotary table but still want to be able to come down to the table to use fixtures. I assume that a person could simply tie two struts together end to end. I would guess the two 150lbs struts together is still a 150lbs but just twice as long.

If ya really want to use it then i'd suggest mounting it to the gear box persay normally and build a bracket with multiple mount locations so that you can set it to were your work is going to be.........ps ... Don't mount it the same as the pic above, mount it more to front in the cavity area behind spindle, this will allow you to tighten gibs a bit more and reduce some of the cantilevering, still allowing the head to be supported with your strut ! OH YA, I think i've finally figured out what Aaron meant by "pretty darned tight" pertaining to the Z saddle gib, mine is now tight nuf that i can't get a .003 feeler to catch gap at the top, and it travels full steam with no complaining at all from drive system of Aaron's servo config. If ya can swing the change then go for the gusto and go full cnc, The IH cnc system is strong.

Cruiser, sorry didn't quite follow your suggestion. Are you suggesting to use cables as in the picture above and then have multiple mounting locations for the end of the strut?

what i meant was to mount the gas strut to box in hollow area behind spindle and then mount the top of the gas strut to a bracket that you can change or better yet a slider that you can pin when your in your work zone, and unpin for changing beyond stroke of the strut personally tho i'd look into other avenues, gas struts and counter ballances are bandades with the problem still there to be a weekness when you need rigidity !

Cruiser, OK - that makes sense. Unfortunately, I have already made and mounted my brackets at the top and will give this a go before trying something else if nothing else to see how well it works regardless of the range of travel. I am curious if you have other suggestions. What I am doing may be a bandaid but if the problem is the weight of the head and distance from the column what other viable options exist? If the slide was larger or the head was lighter or smaller or if the weight was distributed around the column at the head's CG things would be better... With your CNC system you don't see any of the sticking as the head moves? Hmm... should find someone in Iowa that has one all setup and check it out.

Ok - what I did to reduce this problem was the following. I drilled and taped three holes at the top of the retainer on the Z-axis. Bolts inserted through these holes pull the top up and remove the backlash.

http://www.cnczone.com/gallery/data/500/medium/CRW_3656.jpg

Also, I make a couple catilever for pullies and a joining block to create a counterbalance. The two bolts in the pictures attach at the top of the column with cables coming around to attach in the middle of the head on either side. (Sorry, I have since CNCed the mill and do not have a picture with the counterbalance on the mill). Personally, I am not convinced there is much benefit to counterbalancing. The additional holes are just for looks - this metal does not add much to the total strength according to FEA.

http://www.cnczone.com/gallery/data/500/medium/CRW_3658.jpg

This is all quite disapointing to hear this now when I had my mind set on getting an IH mill. I had always heard that the round column mills were notorious for head movement and all that.

Has anyone here done the same test with a round column mill? I'm curious if they will also move a similar amount when pressure is applied to the head.

welcome 1990notch!

Just an observation and thankful that I don't have one of these mills.....this looks like a place where some massive linear rails rather then counterweights would aleviate all the problems...

Thanks for the welcome!

I haven't had access to a mill since the days I worked daily with a Bridgeport some 25 years ago. I don't have room or the money for a Bridgeport and won't for at least 10 years or until I win the lottery, which will never happen. I was really hoping that I could get a small mill that could at least give repeatable accuracy to .002 or better. With a quill that moves 5-15 thou, and a head/column that moves 4 thou with hand pressure, I realize it's just not possible, at least with this mill in unmodified form.

I like the linear rail idea. What's the approximate cost of something like that?

1990notch - since I started this thread I should provide an update. I have recently CNCed the mill and have discontinued the use of the quill as it is mostly redundent and not very rigid. It is still possible to deflect the head with my hands but not to the same extent because now the gibs are tighter since they are now motor driven instead of muscle driven. I am extremely pleased with mill as a CNC mill. If I was purchasing another mill for manual operation I would give the Grizzly G3102 serious consideration after having seen one recently at Grizzly's showroom. Regarding your comment on .002 repeatable accuracy I can only describe a test I did recently. I zeroed a test indicator against a block, rapid out a ways, and then rapid back to the same position and the test indicator was dead on. This was actually a little scary to watch the point of the test indicator zipping towards a solid block like that.

Wildcat - Thanks for the quick response. I hope you don't mind me asking more questions about it, I was planning to buy one by this weekend but want to make sure I'm making the right choice.

When you checked the head for deflection, did you also check to see if the column was flexing at all? Maybe using a dial indicator with the mag base on the table could pick up some column movement. I'm curious as to where the deflection is mostly coming from. It could be a little in different places all adding up to what you see at the spindle. Or is it all in the way the Z dovetails are just a little loose?

I'm glad the table repeatedly comes back to the same location without any cutting going on, but what happens when you're cutting material and pressure is being exerted on the head by the cut? Thanks again in advance.

1990notch - If your going to convert the mill over to cnc you won't use the quill, like wildcat said, and you won't need the counterweight. People use the counterweights when they are using the machine (or a clone) manually or have a weak z motor. Aaron's kit has a strong z axis motor to move the head so its not an issue. I'm not sure if cantilevering is even an issue. You'll probably plunge the head down with more than 600lbs of force and I haven't heard anyone having issues with the head tilting up, so the 200lb head hanging down probably won't give any tilting either.

And your machine might act differently because of the newer design. The new machines have longer z-slide with give a larger surface area, so if the deflection is at the z-slide it could possibly be less. I would check for you but I don't have mine up yet.

welcome 1990notch!

Just an observation and thankful that I don't have one of these mills.....this looks like a place where some massive linear rails rather then counterweights would aleviate all the problems...

Neither a pure counterweight nor a pure linear rail solution will alleviate all the problems. Have a look at this Hurco VM-1:

http://www.thewarfields.com/img/Toys/MachineTools/OthersProjects/VM1framemonster.jpg

Notice it runs on massive linear rails in Z. But what's this at the top of the head? What are those roller chains going up and over? Why bless my stars! It's a counterweighting system!

No quilll there either. They're not necessary for a square column CNC mill. The knee mill crowd uses them on CNC because moving the knee requires such strong motors most prefer to go the quill route. I will be looking at some way of locking my quill mechanically when I've CNC'd to eliminate any possibility of motion there.

I will also say the head of my mill does not deflect with my hand. That's an issue one would have to fix. Surface finish alone would have to be atrocious and I would think climb milling all but impossible.

As I've said somewhere else, there is a Tormach mill for sale right now on eBay (search Tormach) with a tapered Z dovetail. That is an ideal machine for someone to purchase, mill the column to take a couple of linear slides, and be up and running with a fine new machine. FWIW, if IH had a mill on eBay super cheap with the same defect, I would buy it and perform the same surgery. I am surprised we have not seen more folks convert these machines to the linear rails, and more than a little curious what that might do for performance. Having gone to ballscrews and linear slides, there's not a lot left to "fix" except perhaps for more rigidity. My machine seems fine rigidity-wise when I look at surface finishes (particularly with a power feed) and depths of cut. I guess a higher speed belt drive spindle would also be on my list--there's always something you could tweak.

Best,

BW

Thanks for the replies. I'm planning to use whatever mill I get manually. Though I do change my mind sometimes. The thing I was most concerned with was the statements I've read of the head deflecting .004 or more with only 25 lbs of force applied in a side load. Glad to hear that's maybe an isolated case. I'm guessing some folks have the gibs on the column loose enough to make it easy to crank upwards and that is probably the reason the head is moving. If I get the IH, I'll probably add a counterweight system and see if I can figure out why the quill moves off center and see if I can fix that.

Welcome to the never ending search for perfection in a DIY machine.

The machine pictured in post 23 is using linear guides that are probably machine tool grade with preloaded balls. From my experience with nearly identical units (we bought 2 for to make our cam measuring machine) they run between $2500 and $3500 EACH. There is no slop and they definitely repeat. When you're measuring cams to 0.0001", you don't get away cheap.

When doing comparable tuning on our Bridgeport (we tried to use it cut cam masters, we learned essentially the same lesson re: moving the quill and/or table when doing precision cutting - you simply DON'T. The chances of moving up or down and holding X/Y positioning in a gibbs or sliding column quill are two - slim and none.

You can minimize it to maybe thou or two but as soon as you loosen the gibb, BOING, things move and you don't know where.

You can minimize it but if you really want accuracy, plan your cuts well and get rid if any hysterisis that you can but be respectful of the fact that you can't get rid of it all - not without replacing acme's with ball screws and gibbs with expensive preloaded machine tool grade linear ways.

I just checked an RF40 round cloumn. I can deflect the head +-0.002" by a not to agressive push/pull on the head. I didn't check the quill but I now from previous investigations its quite sloppy. These are things that are not easily fixed. But then everything will deflect when you apply a force, it's all a question of what you can live with.

Regards
Phil

Has anyone here done the same test with a round column mill? I'm curious if they will also move a similar amount when pressure is applied to the head.

Phil - Thanks for checking your round column mill. Good to know. I'm tempted to get the round column HF (green) mill since it's cheap at $800, close by, has scraped ways and the belt drive has some good features in that no oil to leak and a 2500 RPM max spindle speed.

On the quill: When I used to use a Bridgeport, the only time I used the quill was when drilling, spot facing or plunging with an end mill or once in a while with a boring head. Other than that, I allways locked the quill and moved the table up or down as needed for making a deeper cut, etc. So I intended to use a square column in a similar manner, locking the quill and moving the head up or down as needed.

I just checked an RF40 round cloumn. I can deflect the head +-0.002" by a not to agressive push/pull on the head. I didn't check the quill but I now from previous investigations its quite sloppy. These are things that are not easily fixed. But then everything will deflect when you apply a force, it's all a question of what you can live with.

Regards
Phil

1990notch - I used the HF mill you mention prior to purchasing the IH. I liked everything about about it except the round column. It did not seem to make a difference how careful I was in setting the head height initially there always came a point where I had to raise the head and on a round column that was very painful. Given the tolerances that you desiring you may also find a round column mill frustrating. I never measured it but I would be surprised if the round column I used was better than the IH. I know very well I can take much heavier cuts successfully with the IH than I was able to with the HF round column. HF has a square column for a little more or give the mini-knee mill I mentioned a few post up a look - it is very nice looking but I must admit that I have never used it.

Bob Warfield....apparently that Hurco was a stepper driven model....because the new one is servo driven and they (Hurco) mention that counterbalancing is not needed with the servo drive.

http://www.hurco.com/NR/rdonlyres/554F9243-FF9C-427B-B6AE-E2581FF715BD/0/VM2Framemonster.jpg

This will be my notes of my IH mill referring to the cantilevering and tollerance of the head "Z". I took a piece of tool steel (a damaged lathe tool holder) and using a 7/8" end mill, and took a prg'md path around the stock of nearly 1.25" depth and close to .15" d.o.c. and started with the material outside of cutter from column, i moderated the feed and it was all and good, then it came around the horn starting to buck and dance a bit then it went around again so matr'l is between cutte and collumn and the cutting force was enough to pick up the weight of the head and the cutter tried to walk on the material. I found out what i wanted to know, it was noisy and it did a real dance, but it did it ! after this i backed off to a more reasonable cut and it was smooth enough to be acceptable, then when i got to finishing, it was very docile during cut and when finished it was exactly on size and looked very good ! The part I was making was a key set to stake my kurt to the table, so there was a lot of matr'l to come off of the 1.5" tool shank i was cutting up. In the shop i work, even the Big Mills have issues, Am I impressed with what the IH mill will do ? YOU BETTER BELIEVE IT ! this is a lite machine yet powerful and full of capacity, did i say powerful ? I stalled a cutter in alluminum that i had clamped down solid ! and I moved the matr'l under the clamps with an end mill, I had to chuckle on that one, it caught me off gaurd but it didn't strain nor loose its position and the two clamps i had were tight as H**L before and after. I am assured that I purchased the Mill I wanted, it has great potential ! I won't be bothering with counterbalancing nor any other fancy, altho, I do have a potential idea of machining a box slide rail system to add onto the collumn and build a new saddle to accomodate it which would add rigidity to the collumn and eliminate any cantilevering etc. but that will be next yr before i possibly get to it, and it may not be called for exept maybe some roughing rigidity. . . MY personal opinion !

viper...the counterweight vm1 was also full closed-loop servo driven, as all our machines are...it was first generation vm1, we now use a larger capacity servo motor (higher output) to drive the headstock,and no counterweight system.in fact none of our machines use counterweights any longer,even our largest machinery...many of the z axis servomotors incorporate a brake.

Cruiser - Thanks for that post. I had a tip that the price would be dropping to $1500 on the IH mill today. I was trying to get a sense if I would be running into trouble down the road and what to expect. Your post helped ease my concerns alot. I ordered one today.