Carvey - Self-zeroing clamp question

Talk about all things CNC
groguard
Posts: 71
Joined: Wed Mar 19, 2014 4:01 am

Re: Carvey - Self-zeroing clamp question

Post by groguard » Tue Oct 21, 2014 8:03 pm

cvoinescu wrote:You don't even need G43.1. If you know the thickness of your touchplate, hold-down bracket, and so on, all you need is the probe command and G10 L20 P1 Z<thickness> (or G92 Z<thickness>, if not using work coordinate systems).
groguard wrote:Assuming you are using an all metal body router/spindle you could technically attach it to a metal portion of the spindle and achieve that.
I suggested the same thing a while back, and I was told that the resistance between the tool and the body of the spindle was in the hundreds of ohms range -- which is not very good. I guess that can happen if the bearings are seated in plastic holders, or if they're ceramic, or not preloaded much. This was a Chinese 300 W DC spindle (the one that Inventables call "Quiet Cut").
I would say if there was some way to have it touch the shaft at the top and being able to rotate freely, that would solve that. Maybe a spring loaded arm of some sort.
Shapeoko 2 #3920. Nema 23's. Belt drive Z axis, ACME rod upgrade,
1m rails on X and Y, X upgraded to 60x40 extrusion with openrail, Mid span supports,
DW660 with Super-Pid, IC's DW660 dust shoe.
Build Log

Chrisprols
Posts: 38
Joined: Thu Mar 27, 2014 7:48 am

Re: Carvey - Self-zeroing clamp question

Post by Chrisprols » Wed Oct 22, 2014 7:46 am

Hi all !


I haven't installed grbl 0.9 just yet and maybe there is already material available to explain all of this ... if it's the case please let us all know and share the link ;)

I've read on https://github.com/grbl/grbl this paragraph :
Probing: The G38.2 straight probe and G43.1/49 tool offset g-code commands are now supported. A simple probe switch must be connected to the Uno analog pin 5 (normally-open to ground). Grbl will report the probe position back to the user when the probing cycle detects a pin state change.
It was previously mentionned by ThatOtherGuy435 that :
GRBL 0.9 should be able to handle that through G38.2 (Probe)
And mentionned by cvoinescu that :
all you need is the probe command and G10 L20 P1 Z<thickness> (or G92 Z<thickness>, if not using work coordinate systems)
Imagine I have an aligator clip on my mill bit connected to the arduino pin #5 and the ground of my arduino connected to a metalic plate of known thickness ... Then what is the procedure to zero on the z-axis ? (I mean step by step if you have a bit of time)

Thanks for the help !!
Shapeoko 2 #4852 with Bosch Colt
Location : Rouen, FRANCE

cvoinescu
Posts: 4442
Joined: Thu Jul 19, 2012 6:50 pm
Location: Camberley, UK
Contact:

Re: Carvey - Self-zeroing clamp question

Post by cvoinescu » Wed Oct 22, 2014 11:57 am

One piece of advice: you want them connected the other way round, the metal plate (insulated from the work) to A5, and the tool to ground. This is so that you collect the minimum of interference and reduce the chance of ESD destroying your Adruino.

So, measure the thickness of your touchplate. Say it's 3.33 mm. Place the touchplate on the surface you want to be Z = 0. Jog the machine to be a little above the touchplate, say 5 mm.

G91 ( incremental distance mode )
G38.2 Z-10 ( straight probe, 10 mm down )
G90 ( back to absolute distance mode )
G10 L20 P1 Z3.33 ( set G54 Z offset so that current position is 3.33 )

That sets the top of the touchplate to Z = 3.33, that is, the bottom of the touchplate to Z = 0, in the G54 work coordinate system (the default one). Replace P1 with P2, P3 etc. to set G55, G56 etc. Alternately, you can do G92 Z3.33, which will set the non-persistent G92 offset instead.
Proud owner of ShapeOko #709, eShapeOko #0, and of store.amberspyglass.co.uk

Chrisprols
Posts: 38
Joined: Thu Mar 27, 2014 7:48 am

Re: Carvey - Self-zeroing clamp question

Post by Chrisprols » Wed Oct 22, 2014 12:25 pm

... Thanks cvoinescu !!

and if I understand well, this will be automated in Carvey.
As M. Kaplan said they are going to open source everything ... plus Carvey is running on grbl + a piece of their firmware, we can hope to be able to implement it on a shapeoko !
In the mean while : use cvoinescu's commands :)
Shapeoko 2 #4852 with Bosch Colt
Location : Rouen, FRANCE

ThatOtherGuy435
Posts: 108
Joined: Sat Feb 16, 2013 2:08 pm

Re: Carvey - Self-zeroing clamp question

Post by ThatOtherGuy435 » Wed Oct 22, 2014 12:36 pm

I suspect they implement it as a macro in their control software (ie, GrblPanel, UGS, etc) rather than modifying the GRBL firmware.
ShapeOko1 #??? - 1000mm Dual-drive Y, Double X, S2 gantry motor plates, ACME Z-Axis.

cvoinescu
Posts: 4442
Joined: Thu Jul 19, 2012 6:50 pm
Location: Camberley, UK
Contact:

Re: Carvey - Self-zeroing clamp question

Post by cvoinescu » Wed Oct 22, 2014 2:46 pm

You're welcome.

Their control software is Easel. But yes, that's my suspicion too. It seems to be GRBL-based, and it wouldn't make a lot of sense to fork GRBL, unless really necessary.
Proud owner of ShapeOko #709, eShapeOko #0, and of store.amberspyglass.co.uk

edwardrford
Posts: 1250
Joined: Mon Apr 09, 2012 5:40 pm
Location: Dixon, IL
Contact:

Re: Carvey - Self-zeroing clamp question

Post by edwardrford » Wed Oct 22, 2014 8:32 pm

Rock vs Hard Place:
I run the risk of sounding like 'sour grapes' for any comment I have about Carvey. But, I want to both chip in with my opinion and help sort out fact from fiction. So, with that said, this isn't a sour grapes situation. I wish Inventables luck with their new machine.

Open Source:
Chrisprols wrote:... As M. Kaplan said they are going to open source everything ...
I couldn't find anywhere stating they are going to 'open source everything'. The only thing I saw was a reference to releasing the firmware (which is already open source? in the MAKE video Zach said they use GRBL) and the electronics controller (i.e. the driver board which they call 'Carvin' according to the kickstarter comments). It will be interesting to see if they innovated on anything, or just DFM'ed an existing solution.

In the kickstarter comments they mentioned that the design isn't conducive to open sourcing. Which, on the one hand I agree with, but on the other hand I think misses the point of open source all together.

Clamp:
The clamp in discussion is actually very similar to something I designed for the Chicago public library in June of 2013. There wasn't a touch off in that design though, it was just the corner jig. The way I had it setup required different thickness clamps (or lips) for different size materials, but the general principle is the same.

Here are some pictures of the setup notice that it's 3D printed! They also have one made from white UHMW to accomodate 1/4" material.
clamp_through_door.jpg
clamp_through_door.jpg (109.7 KiB) Viewed 1490 times
milled_through_clamp_closeup.jpg
milled_through_clamp_closeup.jpg (106.42 KiB) Viewed 1490 times

-Edward
Shapeoko 1 #0 - a couple of upgrades.
Shapeoko 2 #0 - a couple of upgrades.
Shapeoko 3 #2 - Stock

crazeegeek
Posts: 118
Joined: Thu Mar 13, 2014 9:37 pm

Re: Carvey - Self-zeroing clamp question

Post by crazeegeek » Wed Oct 22, 2014 8:42 pm

Based on some of the information they have put out so far, I can not see how its a macro based setup in the external software(such as easel).

They have stated so far that:
"Carvey uses a motion controller called Carvin designed by Inventables".
"The machine controller and firmware are open source. The source files will be released before we ship any machines".
"we have developed Carvey so you are able to use any other CAD/CAM package with the machine".
"Zeros itself when machine first starts". see video
Last edited by crazeegeek on Wed Oct 22, 2014 9:33 pm, edited 1 time in total.

ALuomala
Posts: 136
Joined: Mon Feb 17, 2014 7:30 pm
Location: Alberta, Canada

Re: Carvey - Self-zeroing clamp question

Post by ALuomala » Wed Oct 22, 2014 8:46 pm

cvoinescu wrote:One piece of advice: you want them connected the other way round, the metal plate (insulated from the work) to A5, and the tool to ground. This is so that you collect the minimum of interference and reduce the chance of ESD destroying your Adruino.

So, measure the thickness of your touchplate. Say it's 3.33 mm. Place the touchplate on the surface you want to be Z = 0. Jog the machine to be a little above the touchplate, say 5 mm.

G91 ( incremental distance mode )
G38.2 Z-10 ( straight probe, 10 mm down )
G90 ( back to absolute distance mode )
G10 L20 P1 Z3.33 ( set G54 Z offset so that current position is 3.33 )

That sets the top of the touchplate to Z = 3.33, that is, the bottom of the touchplate to Z = 0, in the G54 work coordinate system (the default one). Replace P1 with P2, P3 etc. to set G55, G56 etc. Alternately, you can do G92 Z3.33, which will set the non-persistent G92 offset instead.
Catalin, thanks a lot for this info (it's exactly what I was hoping for)!!!

On a related note, is it possible to put all of those lines into the macros that the beta version (1.0.8 nightly build) of UGS allows?? Meaning, can it all be in one macro, or does it need to be separated, line by line (i.e UGS needs to complete each command before moving onto the next)?? I'm away from my machine ATM, so I can't test it out.

Another question (since I have never used a probe before), the distance implementing the G38.2 Z -10 (in the above example) command: is the Z-10 the distance it will move in total, or in 10mm "bursts" until hits the plate?
ShapeOko2, serial ??
DW660 spindle;
Upgrades: 900mm X axis and 1000mm Y axis; ACME Z Axis; modified MDF wasteboard with t-slot
Primary usage: wood crafting (signs, plaques, and ultimately a CNC-made electric guitar)

crazeegeek
Posts: 118
Joined: Thu Mar 13, 2014 9:37 pm

Re: Carvey - Self-zeroing clamp question

Post by crazeegeek » Wed Oct 22, 2014 8:57 pm

edwardrford wrote:Rock vs Hard Place:
I run the risk of sounding like 'sour grapes' for any comment I have about Carvey.
I really have to stop walking away then coming back to finish a post. Seems every time I do the conversation changes. :lol:

Anyway, your post doesn't sound like sour grapes to me and seems to hit the nail on the head about what parts they are going to open source. It is interesting about the clamp design too. While it is simple in design, people tend to draw off of things they have seen, and I wouldn't doubt at all that it has occurred in this case.

All of this together appears that the clamp/touch setup, is just digital tool presetter and custom firmware on top of grbl.This firmware could be built into the controller or even just the startup script options that grbl supports.



Now maybe we can find a deal on a small presetter and see what can be cooked up around .

Post Reply