Need Help with Zero

[nambass]

I need help / advise on 2 general topics

I am using MachSTDMill Pro.

Topic 1 – Zero
I am using Mach 3 on my DIY CNC router.
I am using a vacuum table (also DIY) and on top of it a spoilboard.
I mainly cut wood / cabinets and toys.
What I found is that when I zero on the top right corner of the wood piece, I not always get it 100% and sometimes I have onionskin and sometimes not. This is frustrating and is casing problems.
Attached are some pics of my setup. I have a touchoff plate of 1.6mm that I use to zero the top of my wood. I also have a gauge block where I always zero my bits.
On this photos you will see the grey vacuum table top and on top of it is the spoilboard.
What I was thinking. The top of my table is a constant, and for most of the time the top of my spoilboard is also a constant, unless I re-surface my spoilboard.

So this constant will help me to have an accurate Z-Axis calibration and all I need to do is then to zero my X- and X-Axis.

Topic 2 – Onionskin
If I then zero on the lower part of my wood and I define the wood to be the exact thickness. My mashine will not leave an onionskin. In Aspire , is there a common way to setup unionskin? Or should I just manually reduce the cut depth to leave onionskin?

[DaveCVI]

Hi,
re topic 1:
Am I correct to think that all your tool length offsets are = 0 and that you are resetting the work coordinate Z zero level each time you change a tool (this seems to be a common practice with wood router users)?

What you described sounds reasonable but I did not see a question in part 1 so I'm not sure what you are asking..?

re topic 2:
I've never used the vectric aspire software so I'm not much use re advise on that. Perhaps one of the other users here can chime in.

Dave

[nambass]

Hi
Well, yes, I use the master tool and then with the manual tool change, I need to measure each tool once changed. An this works well. But the measuring of the tool is not my problem. I use a fixed position gauge block to measure my tools. and then I have a separate metal piece I put on top of each new wood piece, to measure the top of the work piece. But this is where the problem comes in. Even if I set the jog mode to step I not always get the top of the wood 100%. So to try and eliminate this problem, I was thinking of using the fixed height of my spoilboard of table surface. In my cam software, I can use the bottom corners of my wood as reference points. Therefore, if I use the surface of my table as reference, this is a constant and will not change, I then need to add an offset equal to the thickness of the spoilboard to know the top surface of the spoilboard. then I will be much more accurate in the measurements of my wood piece thickness.

I am trying to get to a same kind of solution that I assume an auto Tool change setup will provide me.

Hope this makes more sence

[DaveCVI]

Hi,
AS MSM has many options, to understand the details of what you are trying to do, I need some more info:
1) You said you are using a master tool. So I assume you have the "Master T Mode" button on the tooling page turned on - correct? (that's not quite as silly a question as is may sound, I've had users doing a master tool approach w/o using MSM's built in support).

2) Please tell me what tool change options you are using:
Tooling page, TC Options & Master Tool Mode panel
TC Auto TCP?
TC Auto TLO?
TLO= PTL?
Skip RH Tools?

3) on the setting common page, have you selected
TLO G-Blk is TP?

TCP TP Installed?

Dave

[nambass]

1) Ref All on Load screen
2) Zero X,Y,Z Coordinates to make sure all DRO settings are 0.0000
3) MDI = M6T200
4) Insert Master Tool (Tool 200)
5) Touch off on touch off position/plate – fixed position.
6) Place work piece on table
7) Place plate on workpiece and touch off with Z-Axis = Set Z-Zero position
8) Move to correct corner and set X-Zero and Y-Zero
9) Load g-code
10) Run G-code

[DaveCVI]

Hi,
Thanks for the additional information - now I think I'm following along with you re what options etc are in use.
It seems that the problem is that the depth of cut is varying a tad - which creates the "onion skin" that you refer to.
OK, that leads me to think about how some physical error could creep into the system.

I'm thinking off the top my my head so bear with me a bit...
Some possibilities that come to mind:
1) machine rigidity
How rigid is the gantry? Could the Z level of the bit be varying as the spindle moves in X,Y? Would this amount happen to be about the thickness of the onion skin?

2) Error when setting WC Z0
OK, It's going to become apparent pretty quickly that I do mostly metal stuff rather than wood. I'm wondering how thick the onion skin is.

I've not got a good feel for how constant I'd expect wood stock to be re thickness - but I'm thinking that I'd not be surprised to see significant (in metal working terms) wood thickness variation along the board.
Could it be that the stock is not a perfectly uniform thickness?

When you put the plate in the stock, could there be any burrs that are keeping the plate from sitting "flat" on the stock? A common problem in metal working is a chip under the plate throwing things off a few thou.

I'm assuming that this is happening in jobs that are all done in a relatively short time frame (hours not days) so that changes of humidity and hence wood size are not a factor.

3) Error when measuring a tool.
I think this in unlikely given the work flow you described.
But for education sake, I'll comment on how MSM measures tools...
During a tool change (M6) sequence, MSM uses the TCP TP to measure the tool. What is measured is the physical length of the tool (PTL). PTL is defined as the difference between Machine Coord Z0 and the MC Z position when the plate triggers the input into mach. The calculation also uses the known MCZ height of the touch plate - which is stored when the TCP TP Z location is set up.
That means that the accuracy of the "known" location of the TCP TP is dependent on the repeatability of the Z home switch. If the Z home switch does not trigger at the exact same point each time it is used, the effect will be a variation ins MCZ0 - which will show up as a variation in the measured PTL of tools.

However, I don't think this is causing you trouble as the options you have selected show that you measure each tool every time a tool is mounted. So even though the MCZ0 might have variation, the Master tool is measured when it is mounted and so area ll other tools when they are mounted - so during the times between ref all home operations, things will all be relative to a constant MCZ0.

Dave

[nambass]

Thanks for the reply.

Well it believe the gantry and Z-Axis is very rigid and will not move. I am using proper ballscrews.

And measuring the top of the workpiece everytime with a piece of metal and making sure it is dust free and lies completely flat is a hassle. I still beleieve if I touchoff and set my Z-Axiz zero position to the table top or the fix hight of the spoilboard. it will be much easier - don't you think.

The other option is to define a fixed position for example G55 that is the correct Z, X and Y values I need to 100% define the correct corner and Z-value of my spoilboard. That means if the G55 then 100% define the correct spot and also reference the bottom corner of my work piece, will that not work easier?

What do you say and how should I go about making sure this G55 point is perfectly defined and also how will I use this in my Gcode?

[DaveCVI]

Hi,

Thanks for the reply.

Well it believe the gantry and Z-Axis is very rigid and will not move. I am using proper ballscrews.

If you have a dial indicator you might want to set it up and measure this. If we are talking about differences of a few thou to make the onion skin effect - well, you never know until you actually check it.

And measuring the top of the workpiece everytime with a piece of metal and making sure it is dust free and lies completely flat is a hassle.

Of course it's your choice as to trade off between effort and results - but I would not be expecting precise results with stuff between the touch plate and that stock.

I still beleieve if I touchoff and set my Z-Axiz zero position to the table top or the fix hight of the spoilboard. it will be much easier - don't you think.

You can set Z0 to any location you like - MSM does not care. If you use the table then that is Z0. Of course you then need to adjust the gcode as cuts will not be starting a Z0, they will be starting at a positive Z# (whatever the height of the stock is).

The other option is to define a fixed position for example G55 that is the correct Z, X and Y values I need to 100% define the correct corner and Z-value of my spoilboard. That means if the G55 then 100% define the correct spot and also reference the bottom corner of my work piece, will that not work easier?

What do you say and how should I go about making sure this G55 point is perfectly defined and also how will I use this in my Gcode?

G55 is just one of the available work coordinate systems that Mach supports (there are 255 of them). The default one is G54.
MSM does not care what WC system you have selected - it simply sets Z0 for whatever the current selected SC is.

Dave