Direct Connect Vs. Belt

[Cartierusm]

I all ready finished my cnc router and it's working. I designed it myself. My questions is I see A LOT of people on here building there own machines using belts instead of direct connection. What's the main difference?

I have no previous experience with these machines so I went for the direct connect method. Using a coupling I hooked my stepper motors directly up to the 1/2" acme screw. SO far everything seems to be working good, it doesn't seem to be straining. I have a 48" x 30" table with 1/2"-10 2 start acme screw and anti backlash nuts with Keling 425 oz. motors running at 24V.

[CWBAILEY]

Using toothed belts is an inexpensive method of getting a 3:1 or 4:1 reduction.
This can be for two reasons one is to increase the tourque of the stepping motor without going for larger stepping motors. The second is to increase the accuracy eg using a 1/4 lead screw and a 200 steps per rev stepper each step will move the nut 0.00125 if it was geared 4:1 the accuracy per step is 0.0003.
Both have their down sides with geaing down the number for steps your computer has to generate for rapid movements is much higher and so the computer can loose steps or you get a much reduced rapid traverse.

[kram242]

I would like to know the answer to that as well.
Mark

[yoopertool]

The reason I went with belts is because I needed to create a gear ratio. I also wanted to be able to modify that ratio in the future, so belts were the best choice. Im sure there are other reasons out there, but this is mine.

[kram242]

Is it because the gears allow a smaller stepper motor to become more powerful.
For example a 50 oz stepper geared could now become the equal of a 100oz?
Mark

[djmurray]

I used a combination of direct AND belt drive on one axis. The reason is that I'm using TWO 1/2 inch 10TPI ACME threads (one on each side of the carriage) to drive that axis without canting the carriage (approx. 26" long). One thread is powered directly by the stepper, while the other thread is powered by a drive belt, connected to the direct-powered thread by a 1:1 pulley setup.

As a result, I can easily adjust the carriage to ensure EXACTLY 90 degrees from the other axis by turning only one of the threaded rods until I achieve alignment, then tightening the setscrew on that rod's drive pulley. Now, when the router moves across that axis, there is no chance of axis misalignment due to cutting forces exerted on the carriage.

Does that make sense?

[geebee1912]

Backlash. If your coupling is a good one there will be no difference between the movement of the motor shaft and the screw. With belts it will depend on the belt tension how much backlash you will get. With proper adjustment belts are fine they just don't have the same precision.

[ibwood2]

One thing you need to keep in mind with stepper motors and reduction to increase torque. A stepper motor torque decreases with an increase in RPM. adding reduction gains you little increase in driving torque because the motor must spin faster to reach the same output speed and will not have any more torque than a direct connection. Unless you have a very low target speed, there would be no real benefit to a reduction drive.

[thkoutsidthebox]

I originally used belt drive to drive my two Y axis screws using one stepper. When I changed to two steppers for twice the torque I drove each Y axis screw directly.

[ger21]

Bottom line is that the screws need to match the motors torque curve to maximize the performance from your motors. It's easier imo to buy screws with the correct turns per inch and direct connect rather than use belts and pulleys to make a screw act like a different screw.

[Cartierusm]

Thanks for the responses. I figured it was a combination of what everyone said, but as I've never built or dealt with stepper motors before I wanted to be sure I wasn't missing something.

I figure for my size table 425 oz. motors should be fine. For some reason I have some backlash in my Y axis even though i'm using anti-backlash nuts .I've been trying to come up with a way to measure this so I can adjust for it. I finally figure out how. I might post this on it's own so other people can benefit as this is post is not titled as such.

I will place a dial caliper on my table, for those that don't have one you can buy them pretty cheaply at auto stors (for checking brakes) or Harbor Freight, jog over until the dial moves on the dial caliper. It doesn't matter what the gauge actually says as we are just using it's sensitive nature to see when the table moves. Then I will reverse jog with the STEP JOG button set to my resolution. In Mach 3 you can set a step jog, meaning when you hit an arrow key it will jog per step of your motor to a distance you determine in the Jog Dialog box. I will set it to .0001 which is 1 inch divided by my total number of steps per inch, 10,000. I then zero out my Y axis, then I will jog one step at a time until the dial caliper moves. Then you look at your DRO and see how far off it is and then put that value into Mach 3. I'm sure a lot of you all ready know this and I didn't want to ask questions to something I knew I could figure out and since I've been asking a lot of questions lately I figure I would contribute.

[sansbury]

Originally Posted by ibwood2 One thing you need to keep in mind with stepper motors and reduction to increase torque. A stepper motor torque decreases with an increase in RPM. adding reduction gains you little increase in driving torque because the motor must spin faster to reach the same output speed and will not have any more torque than a direct connection. Unless you have a very low target speed, there would be no real benefit to a reduction drive.

Not so fast, there. You're right that this is not a completely free lunch but there's definitely room to work with. I'm looking at the torque curve for one of my steppers. At 1070 PPS it puts out a little over 60Ncm of torque. To cut the torque by half, you have to go almost six times as fast, which leaves plenty of room to gain advantage from a reduction drive. Bear in mind too that the screw's thread pitch is a reduction drive in itself.

Originally Posted by Cartierusm time until the dial caliper moves. Then you look at your DRO and see how far off it is and then put that value into Mach 3.

My experience is that software compensation usually makes the problem worse, because whatever backlash you get fiddling with a dial gauge, may not be the same as what you get when cutting. How much total backlash are you seeing?