Hello guys, I have a 25x36 half home built half old dyna cnc machine. I was just cutting some cabinet sides for one drawer bases. The short x axis is deflecting(maybe) about 1.5mm in the middle of the cut then returning to 0. They are non supported rails but i have just put an extra bearing to stabilize the z. The cuts before and after the install are the same.
Could this be just the 5mm compression bit deflecting or a router issue?
Anyone think i could just program an arc on those two sides to compensate?
Speeds are 20 ipm at about 8000 rpm.
We are dealing again with axis confusion. when you say X and then call it the short axis do you actually mean the gantry? Further wich of the four possible directions are the rails deflecting. That being said the cutter will always generate reaction forces, how strong and the direction of those forces varies considerably. Just moving from climb cutting to conventional can significsntly impact deflection, even the direction of deflection.
I have no idea what you mean by adding a bearing to the Z to stabilze it. Adding bearing to an axus saddle will do nothing for you if the linear bearing rods are deflecting. You make no mention of your linear bearing rod size, that is a huge factor in deflection as is the unsupported span length.
One can argue all day about the value of unsupported linear bearing rods but im completely confident in saying that they are almost universally undersized in the router implementations ive seen in the DIY world. One quick way to prove to yourself is this is an issue on your machine is to apply 50 and then 100 pounds of force to the center of the rails and then measure the resultant deflection. If the deflection is in that 1.5mm range then you will need to build a new machine if you need tighter results.
As far as trying to mitigate the issue via software if you have the time and money any thing is possible. Ive actually seen some pretty advanced tool path corrections in the machining industry. The gotcha is that such correctionsare done on machines that are already rock steady. I doubt that the machine you are talking about is rigid and repeatable enough to even consider tool path compensation.
Your only options here are techniques that lower the loads on the spindle to reduce deflection. Some of those will drastically impact run time such as a much lower depth of cut combined with multiple passes.
I don't want to be the bearer of bad news here but it can be very very difficult to get good results from unsupported linear bearing rods. Until you can upgrade to fully supported rails you will likely need to run the machine tenderly.
Note all the above is mostly focused on deflection related issues of unsupported rods. Yiu might very well have other issues. I just focused on deflection because the description fit.
Part of the problem is they are too close together but I am working with what I have. The aluminum plate is straight according to my straight edge.
How straight is T slot....If I would just put v bearings and bolt on a length of c channel on the front of the Y carriage.
It is hard to tell from the pictures but here's what I see. That is if these pictures need rotated. Can't tell about the orientation.
You will have a torsion load on the carriage bushing in the lover rod and pulling on the upper rod. It looks like there may be a roller bearing trying to take up that load that may not be working.
How thick is that plate on, what I am assuming, is the mount for the router head? That plate looks thin and could be deflecting.
The flaw in the design is the guide rods are floating. How much trouble would it be to use linear guides that are bolted to the gantry? No matter how stiff you make that gantry the load going in to the carriage will not be taken up without some additional guides. Check that bearing riding on the opposite side for any clearance of slop. Is the surface it ride on clean.
Move the carriage to the center of the gantry, get your dial indicator on the rods, and just put a little weight on the spindle and see how much those rods deflect. Have you ran your dial indicator on the carriage against the gantry? My money is on deflection from the rods but without testing you can't be sure.