View Full Version : Machine Knee Error?
I have a new Jet JVM-836 (Knee Mill) and have been trying to setup my new rotary table. I am using a co-ax indicator to align table under spindle. Once the co-ax is removed and the knee is positioned back up towards spindle, I have been seeing a major error in the alignment. Mainly the "X" axis being off. At first I thought it may be the co-ax, so I put my Indicol and Compac indicator on the spindle and checked the table. Alignment was to within what the co-ax had shown. Then it occured to me that the error was in the knee. Question is, how much error should I see on knee travel? Or how much is common. I don't really know how to accurately check the error but just using a parallel,vise and indicol I am seeing about .020 in 6 inches or less. This is quite disturbing, but being my first machine I am not sure what to expect... Any information is greatly appreciated. Thanks!
06-16-2004, 08:30 AM
Am I reading you right,that your dialing in the RT to the spindle and then moving the the knee up and getting an offset?
If so how much is the error your seeing?
Could it be the knee locks are pulling you over,or have you planed the mills spindle in first,if its not planed in to the mill table then the co-ax will center it localy,but when you move either the quill or the knee it will go off center.
One thing you might check is your co-ax indicator,many will stick and give you a false reading if they don't see a drop of oil now and again.
[This message has been edited by wierdscience (edited 06-16-2004).]
06-16-2004, 08:54 AM
Check that your head is trammed in vertically in the X and Y.
Quick sketch [ exagerated ] to show what can happen.
Sorry for the crap sketch but it is dinner time http://bbs.homeshopmachinist.net//biggrin.gif
Weird is also on the money with the knee locks.
That's way too much error. Make sure the milling head is square to the table. Also make sure the knee gib is adjusted properly.
I'd check the alignment by putting an accurate angle plate on the mill table, an indicator in the quill, and cranking the knee up and down so the indicator moves along the angle plate.
I am seeing, from the best I can tell right now is about .020 in 6" of knee movement. I have not been locking the knee gibs. I did plane the spindle head with my Indicol, but only with the knee in one position. What worries me is that the knee is moving this much. Thanks!
I did tram or plane the table in on the "X", but if the "Y" is off I am not sure what I would do. With the Jet, there is no adjustment on the "Y". I cannot pivot the head up or down, only left to right. Thanks!
Make sure the casting the ram goes through, that sits on top of the main column, is seated properly on the column. If that's off, due to whatever (a loose bolt?), it would give you the "Y" error. If need be, that can be shimmed, although that's not a very satisfactory solution.
If alignment really is off as much as you're saying it is, I think it falls into the "excessive" range (by a lot) and Jet ought to replace the machine.
[This message has been edited by SGW (edited 06-16-2004).]
06-16-2004, 11:03 AM
If the error is off in the x axis only, then you almost had to misread your indicator, which is very easy to do when it goes out of sight behind the spindle.
If the table was sitting on a 45* angle,you could still tram the spindle to the table by swinging the head over to the same angle. This pretty much eliminates any problems with the knee. Once trammed in in any knee position should still be trammed in any other knee position.
Any error in the y axis, with no adjustment in that direction is a manufacturing fault. Knee clamps or a loose gib could be the cause of a small error in the y plane.
Thats my story and I'm sticking to it,unless somebody changes my mind.
Jim W. http://bbs.homeshopmachinist.net//confused.gif
06-16-2004, 01:43 PM
Look at the mill testing diagrams at -
These help me, especially when I go wrong, get 'spooked' and lose confidence in myself or the machine.
Try some of the knee tests.
You probably forgot something basic. I do that all the time.
You have all of the gib screws tightened up?
For what it's worth...I just checked the specification sheet that came with my Jet JVM-840 (you didn't get one with the -836?) and it gives the maximum permitted error on knee alignment for the -840 as 0.02mm/300mm, which works out to be about a thousandth of an inch per foot. I expect the tolerance spec for the -836 is similar -- certainly not 0.02" in 6"!
I'm with Dean; I find it's awfully easy to fool myself when I'm trying to measure stuff like this.
Thanks for all the help. I have been standing cheap parallels up in the vise and taking the knee readings. I don't think the parallels were ground(precision) on the end. I am going to get a precision ground angle plate tomorrow so that I can check this. I do have to say that Jet has the best customer service I have ever dealt with! She was ready to send out someone to look at the mill today. I told her that I was going to do some more test before I started to jump to more conclusions. Again, A++ people to deal with!
Ummmm...yeah. I'd be highly suspicious of the reading you got off a parallel standing up in a vise.
06-17-2004, 10:11 PM
Mount an indicator on the head or any fixed portion of the machine and a precision square on the table, indicate the beam of the square and start cranking the knee. Do this 4 times with the square at 0, 90, 180 & 270 degreees to account for error in the square. You will soon find out where the problem is.
Of course, there's always the possibility that the table surface is not at right angles to the knee travel, too.