Hello, The construction that I want to describe here is a modification of the existing solution, and the topic serves to convey the idea and provoke discussion about this solution.
I have a CNC milling machine of my own design, in which the drive is transmitted using trapezoidal screws (a simple, popular and relatively cheap solution). In order to remove the slack on the screw, I originally used 2 nuts countered and twisted in the mounting.
This system worked very well and ... very briefly. Although at the beginning I was satisfied with this solution, convinced that clearance would not be necessary too often, I did not foresee quick access to the nuts. Despite the fact that the milling machine is easily unscrewed, clearance of the slack is troublesome, and after making some small details the measured slack was 0.1mm !!!
On the network I often found ways to automatically clear slack by putting a strong spring between the nuts. But it doesn't make sense! Why does everyone duplicate it, it doesn't work, the first thought is that in order for the erasing to work, the spring should push the nuts with a force greater than the force during machining.
Winter came and again the motivation to play CNC came again, so I quickly came up with a way to clear slack. A very quick way, the first that came to my mind. I searched the internet, but I didn't find a similar solution anywhere. Maybe my idea has a big drawback that I don't notice now. Hence, I decided to provoke a discussion on this topic.
The idea is to fasten one of the nuts firmly and force the other nut to rotate through the spring. By turning, the nut counteracts both nuts between the walls on the screw. In this way, the clearance is automatically erased and the nuts are rigidly fixed. Below are some photos and a film from the clearance measurement (0.01 mm).
Very cool solution. I suggest supplementing with a small latch (e.g. solved in the same way as in a ratchet wrench) cooperating with teeth cut on a movable nut, which will allow the use of a weak spring and prevent the nut from retracting with a larger friction coefficient in the bolt-nut system.
The latch here would have nothing to do, just an unnecessary complication. One of the nuts is fixed, the other is rotated by a spring. The spring is very weak here, it only serves to pretension the nut, further tensioning takes care of friction. As long as the nut-friction friction is greater than the nut-housing friction, the nut tightens. When turning the other way, the screw slightly releases the movable nut and pushes the stationary nut, which will not unscrew the movable nut.
Very creative. But can't there be such a situation that the movable nut firmly locks the trapezoidal screw so that the engine will lose steps. I don't know the construction of your machine. I look through the lens of my milling machine. I am also fighting against deleting slack on trapezes. For now, I did the way you described, i.e. counterattack. And with stronger countering, the engines lost steps, and reducing the feed speed is not a satisfactory solution for me. I have good access so regulation is not a problem. However, I ordered ball screws. Alli prices are quite affordable. I don't know how with endurance. We will see. I mainly mill PCBs.
But it doesn't make sense! Why is everyone duplicating it
Because there is no risk of jamming such a system. There is such a risk in the solution you present.
Well, but the fact that there is no risk of jamming does not mean that you need to complicate the system in a way that does not work. Because if it doesn't work, why is it installed?
As for my method, it has already been established that there is actually a risk of the nuts getting stuck if the screw is driven into a less worn place. But as long as the jam does not occur (if it occurs at all), the system cancels the play correctly. As for the occurrence of a deadlock, it must be verified in reality. I see this risk but in 50% chance. In addition, the wear of the nuts will certainly proceed much faster than the wear of the screw, perhaps so fast that the replacement of the nuts will be necessary sooner than the difference in screw wear will be large enough to seal the system. And when replacing the nuts, you can replace the screw, which in my system costs PLN 3.
Hello Finally the ali ball screws arrived and I will tell you that no more trapezes. Less resistance, higher speed and no looseness yet. We'll see how it will be in the long run but for now it is ok. The prices are already very attractive, I also recommend. They make tailor-made ends or not, if you prefer.