I am sorry guys I did not explain myself well. I do not think the column is bowing/flexing as you have described it. Let me try again. Below is my theoretical concept that I was suggesting.
Yes, I know about James, YouTube, and his mill. Unfortunately he did not describe exactly what they were, just that he was disappointed. He also traded it in so he did not fix the problem. He also was not trying to use it as a CNC and the knee mill he got is not very good for a CNC conversion. I think the replacement mill that he got is the same one that David Best purchased and David told me that even that mill he had to work on it a lot to get it to work right. He may have even replace the entire head!
All of these bench Mill columns seem to be made in a U shaped cross section. This is so that the head can be attached to the lead screw through the opening of the U. At the top and bottom of the column the U is closed at the front to form a box shaped cross section. The distance between these end boxes determines the travel distance of the head. The ball nut is mounted to the head saddle mount and reaches through the U opening to attach to the turning lead screw. So the saddle attaches (hangs from) to the ways on the column with the column ways (55 degree corners) being at the opening of the U. The head saddle is about a foot long and the gib is in between the saddle ways and the column way. As the head hangs on these ways it tend to torque about the x-axis. That is it apply pressure against the column at the bottom, but pulls on the ways at the top. Due to the angular shape of the 55 degree ways this pulling tends to press the ends of the U together. This torque, for a heavy head, is quite substantial and I believe can deflect the width of the U at the ways. So when the head is in the vertical middle point the column the U bends together, but when the head is at the top or the bottom of the column where the box shape is right at the edge of the U shape part the box starts to play a stronger role to stiffen up the U portion, so the U bends very little if any. Think of the ways as a board supported at the ends but with noting under it in the middle... and you were going to walk across the board. It would deflect downward as you approached the middle. It is just a spring as the board bends. So what does this mean?
It says that the distance between the two sides of the column ways varies as the head goes up and down! In the middle position when the head is far from the ends the distance between the column ways is the smallest. Hence the gib is looser there than it would be when the head is at the top or bottom of the column! It would not take much of a flexing of the U shape to reach a few thousands and a few thousands is a lot for a gib. You can adjust the gib for the middle or for the ends, but it is different. And in this theory it is a function of the head position. Probably somewhat linear. I purchased the parts to make a special micrometer to fit across the column ways but have yet to make the rest of the parts to put it together. If I ever get around to this I will measure the column U separation (distance between the two column 55 degree ways) as a function of the head position. I want to do this anyway to see if the two way sides are parallel. I will of course use this same tool to measure my x and y ways as well. In concept if a had a large distance micrometer I could do the measurement, but there is not much room to insert one behind the motor etc. I am essentially going to build one that is very accurate, but which is not so big. I purchased an open end 1 micron sensitve digital micrometer for this. One inserts a couple of cylinders into the ways to press against in this sort of measurement.
If I do find this U flexing effect is coming into play there is not a lot one can do to the column ways to fix it. But here the possible solutions I have come up with: 1) In concept one could machine the ways at the top and bottom to be narrower, i.e. to be equal to the flexed U distance, but this is a dynamic system and would be different for different Milling heads etc. In concept I could do this with a lot of difficult "scraping" while measuring the flexure. That is scrape down into the 55 degree ways while the head is mounted so as to allow measurements. But is almost impossible to manufacture, even if the manufacturer were precise at manufacturing. NOT! 2) Figure out some internal bracing inside the U to hold the ways from flexing. Possible, but very difficult given that the ball nut must travel inside the U. Maybe reinforce the shape from the outside would work, but it would be in the way of all of the external moving parts ... limit switch travel, DRO, etc. 3) Rebuild the entire column to be stronger. Maybe a new saddle while doing it. It would be better if the manufacturer had done this to start with! 4) Reduce the head weight so that the torquing effect is reduced. Or, counter balance the head weight in some manor. The best approach I have come up with for this is to mount some pulleys on the top of the column, run a couple of cables over the pulleys, attached the ends of the cable to the head and to a weight that is hanging down the back of the column. As the head goes up and down the weights move in the opposite direction to help lift the head. The cables need to be mounted to the head near its center of gravity to prevent the torquing effect. So probably out past the motor and the ideal approach is for the cable to travel vertically so that the lift does not change angle as the head is lifted. In my mills case this weight needs to approach 275#s to fully counter balance the head weight. Instead of weights some sort of spring contraption might be used. I had though of either air springs that pull, but I have not found any yet that have as long a retraction distance as my mills travel. Or maybe a preloaded torsional spring like the ones used on some garage doors. 5) The last solution I come up with is to get a better mill! But which mill model is it that does not have its own problems. Hopefully just fewer and less severe problems.
As they say, "Man plans and God laughs!"
Good night.
Dave L.
Yes, I know about James, YouTube, and his mill. Unfortunately he did not describe exactly what they were, just that he was disappointed. He also traded it in so he did not fix the problem. He also was not trying to use it as a CNC and the knee mill he got is not very good for a CNC conversion. I think the replacement mill that he got is the same one that David Best purchased and David told me that even that mill he had to work on it a lot to get it to work right. He may have even replace the entire head!
All of these bench Mill columns seem to be made in a U shaped cross section. This is so that the head can be attached to the lead screw through the opening of the U. At the top and bottom of the column the U is closed at the front to form a box shaped cross section. The distance between these end boxes determines the travel distance of the head. The ball nut is mounted to the head saddle mount and reaches through the U opening to attach to the turning lead screw. So the saddle attaches (hangs from) to the ways on the column with the column ways (55 degree corners) being at the opening of the U. The head saddle is about a foot long and the gib is in between the saddle ways and the column way. As the head hangs on these ways it tend to torque about the x-axis. That is it apply pressure against the column at the bottom, but pulls on the ways at the top. Due to the angular shape of the 55 degree ways this pulling tends to press the ends of the U together. This torque, for a heavy head, is quite substantial and I believe can deflect the width of the U at the ways. So when the head is in the vertical middle point the column the U bends together, but when the head is at the top or the bottom of the column where the box shape is right at the edge of the U shape part the box starts to play a stronger role to stiffen up the U portion, so the U bends very little if any. Think of the ways as a board supported at the ends but with noting under it in the middle... and you were going to walk across the board. It would deflect downward as you approached the middle. It is just a spring as the board bends. So what does this mean?
It says that the distance between the two sides of the column ways varies as the head goes up and down! In the middle position when the head is far from the ends the distance between the column ways is the smallest. Hence the gib is looser there than it would be when the head is at the top or bottom of the column! It would not take much of a flexing of the U shape to reach a few thousands and a few thousands is a lot for a gib. You can adjust the gib for the middle or for the ends, but it is different. And in this theory it is a function of the head position. Probably somewhat linear. I purchased the parts to make a special micrometer to fit across the column ways but have yet to make the rest of the parts to put it together. If I ever get around to this I will measure the column U separation (distance between the two column 55 degree ways) as a function of the head position. I want to do this anyway to see if the two way sides are parallel. I will of course use this same tool to measure my x and y ways as well. In concept if a had a large distance micrometer I could do the measurement, but there is not much room to insert one behind the motor etc. I am essentially going to build one that is very accurate, but which is not so big. I purchased an open end 1 micron sensitve digital micrometer for this. One inserts a couple of cylinders into the ways to press against in this sort of measurement.
If I do find this U flexing effect is coming into play there is not a lot one can do to the column ways to fix it. But here the possible solutions I have come up with: 1) In concept one could machine the ways at the top and bottom to be narrower, i.e. to be equal to the flexed U distance, but this is a dynamic system and would be different for different Milling heads etc. In concept I could do this with a lot of difficult "scraping" while measuring the flexure. That is scrape down into the 55 degree ways while the head is mounted so as to allow measurements. But is almost impossible to manufacture, even if the manufacturer were precise at manufacturing. NOT! 2) Figure out some internal bracing inside the U to hold the ways from flexing. Possible, but very difficult given that the ball nut must travel inside the U. Maybe reinforce the shape from the outside would work, but it would be in the way of all of the external moving parts ... limit switch travel, DRO, etc. 3) Rebuild the entire column to be stronger. Maybe a new saddle while doing it. It would be better if the manufacturer had done this to start with! 4) Reduce the head weight so that the torquing effect is reduced. Or, counter balance the head weight in some manor. The best approach I have come up with for this is to mount some pulleys on the top of the column, run a couple of cables over the pulleys, attached the ends of the cable to the head and to a weight that is hanging down the back of the column. As the head goes up and down the weights move in the opposite direction to help lift the head. The cables need to be mounted to the head near its center of gravity to prevent the torquing effect. So probably out past the motor and the ideal approach is for the cable to travel vertically so that the lift does not change angle as the head is lifted. In my mills case this weight needs to approach 275#s to fully counter balance the head weight. Instead of weights some sort of spring contraption might be used. I had though of either air springs that pull, but I have not found any yet that have as long a retraction distance as my mills travel. Or maybe a preloaded torsional spring like the ones used on some garage doors. 5) The last solution I come up with is to get a better mill! But which mill model is it that does not have its own problems. Hopefully just fewer and less severe problems.
As they say, "Man plans and God laughs!"
Good night.
Dave L.
Last edited:
