Rotary testing for engraving and cutting

[Tech_Marco]

I just made two video for a simple test using a single axis "Clamp type" rotary to demonstrate cutting and engraving on AWC608 DSP.
I was told by the programmer that on this kind of rotary, the cutting ratio may not be accurate. In other words, the size of the actual cutting work will be a little off from the image size. But engraving would be good. I will do more testing to confirm this issue

Cutting:
http://youtu.be/7uGX-sWkXg8
http://www.youtube.com/watch?v=7uGX-sWkXg8

Engraving:
http://youtu.be/BEHtxHmwqIs
http://www.youtube.com/watch?v=BEHtxHmwqIs


The difficult to use a rotary is to get correct "um" value. If the motor driver resolution setup and the gear ratio is the same as Y, then the 'um' should be the same. But in my case it was not. The motors on the machine were 3-phase (1.2 degree) and the motor drivers were 3-phase type. The 'um' for both X & Y was 6.3. I tried that value and I could make the rotary to work. Seems OK but found that somehow the speed and the "Y" (curve line) was much much off. I ended up used 31.9 for the "um" Then the result looked much better in the size scale. I made a cut for a 60mm x 60mm box and it ended up 59.87mm which I considered was good for the test.

To measure the "Y", I have to use a small wire and measure length in curve shape. After calibrated, I made a sample tests and you may see it on the youtube video.

Will do more tests tomorrow. Time to sleep (1:43am) . Good night, Buenas Noches,晚安!


Marco

[Greolt]

The "Rotate Engrave" function found under "Advance Functions" does work for raster engraving (back and forth like an inkjet printer) as shown on your second youtube.

All you need to do is enter the correct values and enable it. See pic. below.


However unfortunately that function does not work for vector cutting or vector engraving (as shown in first youtube) It only works with raster engraving.

So what we must do for getting vector work to come out correct size is enter the appropriate step value in the Y axis step field.

The calculation is as follows.

D*pi/step per revolution. This gives the step value in mm. Multiply the answer by 1000 will give it in micrometers (um) as needed by the AWC608

D = the job diameter

pi = 3.14159

steps per revolution = the number of steps to rotate the rotary adaptor one full 360 degree revolution. You need to work out what this is on your machine. On mine it is 20,000

So if I had a job of 80mm diameter the calculation would be;

80 multiplied by 3.14159 divided by 20,000 equals 0.01256636 As I said that answer in in mm. So convert it to um. 0.01256636 multiplied by 1000 equals 12.56636

12.56636 is the correct value for an 80mm diameter job, in my machine, to enter in the Y axis step field. You need to know the steps per revolution for your rotary.

[Tech_Marco]

Greolt, the "Advance option" seems not showing any difference when I did my tests. I talked to the programmer last night and he said it should make difference if it was enabled. I think what he mean was that we keep the Y's "um" value unchanged when we disconnected the Y motor for a rotary. Then, we enabled the "Advance option" for the rotary engraving. Over that menu, we entered a new 'um' for that rotary which could share the motor driver with Y or the 3rd driver that dedicated for the rotary only.

I'm still a little confuse about the driver resolution may be you can give me some lecture. Take a 1.8 degree motor as an example. It should turn a full revolution by 200 pulse/step : 200x1.8degree= 360 degree

So, if we set the motor driver to "5000", what my interpretation is that it the default pulse is going to be divided by 5000, so it would take the 200*5000 = 1,000,000 pulse to get the motor one complete revolution, am I right?

I found there was no problem to get same scale on both engraving cutting on my test as well. What I did was to get a new "um" for the rotary on the "Y" setting. I knew that when doing a simple box like 100mm* 100mm, the scale of engraving and cutting got no difference. Confirmed that with the programmer. He said that if there was complicated image then the scale of cutting could be an issue. Well, I imported some graphics (raster), then added a few characters (vector) and doing engraving/cutting, the scale seems all right.

By the way, the programmer said that the cutting scale issue only happened to the "clamp" type rotary. For the traction type, the one that employed two friction bars or wheels has not scale issue when doing the cutting or engraving.


Marco

[Tech_Marco]

Ok, I just talked to the programmer again and got things cleared up like 90% if not 100%

Per Li (programmer), the steps per rotate(pulse) is actually the resolution it used on the motor driver. If "3200" is selected, then put down 3200. There is no need to times the motor step per revolution. So, in my last case, I would have to use "5000" on that field

The 'um' could be found by using the um calculator on the Y axis setup. Once the 'um' of the rotary if found, there is no need to change afterward but use it for different cylinder shape object but just need to enter the new object's diameter

Now I got my question cleared up and I'll do more test tomorrow to confirm it.

By the way, my assumption was correct about the scale of engraving and cutting matched if the rotary is driven by "Y" without changing the diameter.


Marco

[Greolt]

Marco

The first entry in the "Rotate Engrave" is whatever is the step value for the axis being replaced by the rotary. Y axis in most cases.

Once this first entry is entered, it should under normal circumstances never need to be changed. See Pic below

When using this feature you DO NOT alter the Y axis step value in "Manufacturer Parameters". Leave it untouched.




The second entry is the number of steps required for the rotary to turn one full 360 degree rotation. You must calculate this value for YOUR machine.

Here is the calculation for MY machine as an example;

I have a 200 step stepper motor. (1.8 degree)

A ten micro step stepper driver.

A ten to one belt reduction drive.

200 x 10 x 10 = 20,000. This is the number of steps to turn MY rotary one full revolution. (yours will be different)

So 20,000 is entered in the second field. (or the number of steps for YOUR machine)

Once it is entered it should under normal circumstances never need to be changed. See Pic below




The third entry is the diameter of the job to be engraved in mm. This will need to be measured and set for every job.



Again this feature of the controller ONLY works for raster engraving (back and forth like an inkjet printer)

I DOES NOT work for vector engraving or cutting (following a line etc)

If you want to do vector work (cutting etc) then you must abandon the "Rotate Engrave" section and use the method I explained in the post above.

Greolt

EDIT; I probably need to state the obvious to avoid any readers misunderstanding. Here we are talking about a "Chuck Style" rotary attachment. NOT a "Friction Roller Style" rotary attachment.

[Tech_Marco]

Update: I talked to the motor driver manufacturer tech and he explained to me about the resolution definition

For example, if the resolution on a driver is set to "5000" it means that it take 5000 steps (pulse) for a 1.8 degree motor to complete a revolution (360 degree). The simple formula is 25*200, where 200 is based on a 1.8 degree motor taking 200 steps to finish a full revolution.

If the motor is switched to 0.9 degree, then with the same configuration on the driver, the motor could finish half way of a revolution and the speed is also cut by half.

So, back to the rotary setting: if I set my driver resolution to "5000", then I need to enter "5000" to that field but not 5000*200 (1,000,000)!

Tomorrow I will give another try on the rotary from what I learned tonight.

Stay tuned!

Marco

[Tech_Marco]

Greolt, thank you for your lecture. I'll do some test tomorrow

By the way, do you think that a reduction gear like 2:1 or 3:1 for the rotary is necessary?

The only think I know is that without the reduction gear, I have to set my driver to "10,000" in order to come up 'um' in about 21. I was told that the um better to keep 25 or smaller but bigger than 2.5 for stability issue. With help from a 2:1 reduction gear, I think I can use 5000. Per the motor driver tech, it is better not to go over 6400 to ensure smooth operation and better performance


Marco

[Greolt]

By the way, do you think that a reduction gear like 2:1 or 3:1 for the rotary is necessary?

Personally I would aim for at least 3:1 and maybe up to 10:1.

Direct driving the rotary is possible but resolution will become a problem as the diameter of the job increases.

For example if the job was 200mm in diameter then each FULL STEP of the motor would advance the material 3.14mm.

Naturally very small diameter jobs would be better. 10mm diameter would advance 0.16mm per full motor step.

My feeling is that at even with a 10:1reduction, the rotary can still turn quickly enough for most cutting jobs and the improved resolution is much better for raster engraving.

Greolt

[Greolt]

So, back to the rotary setting: if I set my driver resolution to "5000", then I need to enter "5000" to that field but not 5000*200 (1,000,000)!

I have never heard of the term "Driver Resolution"

Stepper drivers usually have the ability to "Micro Step" This means it can divide one motor step into a number of partial steps.

On most Chinese stepper drives that I have seen this is set via dip switches. With 4, 8, 16, 32, 64, etc micro steps being able to be selected.

I use US made Gecko drives which have 10 micro steps. This is not selectable on Gecko drives.

So if you had a 200 step motor (1.8 degree) being driven with a driver set to 32 micro steps then it would be 200 x 32 = 6400 steps per full motor revolution.

One motor revolution is not necessarily one rotary attachment revolution. That would need to take into consideration the reduction of the drive method (if not direct drive)

In this discussion steps per motor revolution is irrelevant. That is to say it is only relevant to ascertain steps per rotary attachment revolution. (chuck revolution)

Greolt