Rotary engraving questions

[Joe Walmer]

I used to own the signwarehouse pinnacle mercury rotary engraver, but never plugged it in and sold it with our business for someone else to try. I see how everyone engraves left to right and the rotary just moves down to do the next line like regular engraving. I was wondering if anyone has figured a way to just spin the bottle pretty fast and then just engrave as you slowly move down the bottle. This seems like if it worked would be 10x faster and 100x less stress on the machine from going back and forth. one of the reasons I think my laser still works mechanically after 13 years with all original motors, belts, and pulleys etc. is we only slowly vector cut.

I realize most laser software etc is not designed this way but are there any limitations in theory for not doing it?

My main reason for asking is if it was say bowling pin shape if you did it this was you could have a z axis programmed with that outline and it would stay focused the whole time. You could also do concave long things all in one pass on flat engraving as long as you can program the height if its a simple shape over a long enough gradual change it wouldn't be hard to do it almost at full speed. Engraving the entire back of a chair for example would now be slowly possible also but that is all aside from the rotary question I know We can do those things already.

[Scott Shepherd]

My guess is that the workholding on the rotary attachments has never been a strong suit. If were rotating it over and over, then it would have to be perfect in it's positioning. When doing line by line, it never has to come back to that place again, so the quality is only the next line, leaving little room for variance. But once you start rotating mass around, over and over, with things are held in with friction, the chance for error gets exponentially higher. You'd essentially have to line it up perfectly with every rotation and I think you'd have issues with that under a friction drive system.

Just my guess.

[Jeff Body]

I also think the circumference of the item would place a huge factor in this idea.

[Joe Walmer]

I knew it would be harder but I think its doable at least on a commercial level and I bet they do it like that somewhere already. With servos and accurate encoders.

If you had to go old school you could use a strobe light timing system and a line on the shaft of the motor and if you had a know exact circumference which you could fiddle with to find all you would need to do in theory is have a program that chops a picture up and puts it all in a long straight line with info on when to fire and when to jump down one space every rotation.

Another way to zero in would be say if you had a 1 inch tube have two dots firing at the end of the theoretical end of the line and beginning of the next line. Then you have a file that moves those dots closer and further away and where they actually converge on the bottle would give you the actual width of that bottle. I imagine the cut file would look like 2 sine waves starting in the top left and ending in the bottom left and another on the right side. Then you look on the corel file where the points on the corel file match the aligned ones on the bottle and measure the 2 points and resize the corel work area to the new size.

So basically you can use a gcode cutter to make the file one long line and code when to fire and not fire the laser with all the 3rd and now 5th order acceleration on curves etc in tinyG etc. this would have to be easier than having an engraver go fast then stop engraving at the end of a line then do deceleration and re acceleration to engrave right back where it left off. I mean the kids make rotary egg engravers that almost do this already.

Just dreaming while I save up for new parts lol but in theory i think it would be pretty sound and even able to align using strobes the old school way as long as your stock was a constant size it would be worth it imagine the speed increase.

[Kev Williams]

This is the cylinder attachment for my LS900--
314041

obviously not in the machine at the moment ;) --
I believe it's the same unit that's supplied for the IS400 Volumes
and other Gravograph CNC engraving machines. It has different jaws,
and other sizes/shapes of cones & such.
Someone could possibly make a few bucks building similar units
that would work on all these other machines..

I honestly couldn't work with one of those 'rotators'. This unit provides
repeatablity. My LS900's driver also allows for automatic table height
adjustment based on color. Some odd shapes are going to require a 2nd
pass due to focusing, the table adjustment sure helps with that. I engrave
acrylic goose calls that have 3 different engravings in 3 different
places where the diameters are fairly different from each other.
The biggest dia. engraves first (as black), then the table raises to the
2nd focus point, then onto the third. Without the auto-raising table,
the changes would have to be done manually. This works out great.

For angled stuff like shot glasses, I drilled a couple of holes at either end of the
dovetail, then thru the machine table. I can now raise the right end of the
dovetail up to 4", then place a wood wedge under it, then using some long
machine screws I tighten the unit to the table, nice & snug. Now I can make the
edge I'm engraving nice & plumb to the X rail. If I'm doing tall glasses with
a big angle, going from top to bottom will result in the engraving towards
the bottom getting condensed due to the diameter change. The fix is to
expand the bottom of the text or image in opposite proportion to the angle.

Doing this is a piece o' cake with my old Casmate-
I just import everything, draw a quickie rendition of the glass
(or whatever) I'm engraving for the angles. Then flip the top/bottom,
and make it fit the engraving. With Casmate's '6-side' distortion, I just
drag the bounding box to fit the upside-down 'glass'. Send it back to
Corel, then run it. Doing this, I can easily draw a perfectly square box
on an angled drinking glass.