3-d probing w. iosender - making a height map

[Klippert]

Hello everybody,

I'm getting started on the 3d probe. So first of all, i need to understand how to really use the 3d probe.

Here's what i'm thinking, please correct me where i'm wrong.
as the 3dprobe goes in the spindle, in a collet, i have no way of knowing how much it's sticking out of the collet, from time to time i use it. For x and y probing, this doesnt' matter, as long as the probe ball is hitting the object in the right place i'm good.

For Z, this means that i can only probe relatively, like probing the baseplate and then probing the thing i want to know the height of should give me the difference between the baseplate and the object i'm probing.

If I can probe my fixture probe with (without triggering the probe first) i could use that to actually "set" my probe height, but i guess i'm relying on the springs in the probe and the fixture to be different.

First off I'm trying to figure out how to do a height map of my baseplate. Just to get a sense of how much i'm off, and how good or bad my facing actually is.

what's the best way of doing such a probing pattern, where i'd probe every 25mm in a grid, and get a bit of a map of the baseplate?
Does IO-sender support any such probing? (or should it be set up in Fusion?)

Any help would be greatly appreciated.

[phil-barrett]

You define an area to probe and a grid size. The way I use it:

• Load the GCode program
• Make a height map with the 3D probe (use set from program limits and a reasonably small grid size)
• Change to the bit I'm going to route with
• Zero Z at the height map's 0, 0 location
• Apply the height map (apply button on the probe panel)
• Cycle Start

[terjeio]

Is there a way to set the S-height it retracts to between probing?

Yes and no - it is a bug as I use the Depth from Probing clearances for the retract. This is IMO wrong as the Depth field is disabled (as it should be), I will fix this by retracting to the initial Z position instead.

A new edge version has been uploaded with this fix.

[phil-barrett]

I ran a height map on my spoilboard before and after leveling. Was mostly curious. Before leveling, it showed about a 1.5mm deviation across the entire spoilboard. It read within .1 mm after leveling. Not perfect but acceptable for my needs.

Note that this does not necessarily mean my spoilboard is flat within .1mm - my X and Y rails could easily be off as well. If so, the milled spoilboard conformed to that.

[Klippert]

Is there a way to set the S-height it retracts to between probing?

Yes and no - it is a bug as I use the Depth from Probing clearances for the retract. This is IMO wrong as the Depth field is disabled (as it should be), I will fix this by retracting to the initial Z position instead.

A new edge version has been uploaded with this fix.

Excellent. I'll test it in the weekend. - Is the fix committed to the IOSender on github? or the new repository?

[terjeio]

Is the fix committed to the IOSender on github?

Not yet - a pre built binary can be found here.

or the new repository?

The sender is not going to be moved to the new repo, it will stay where it is and renamed to ioSender on the first production release which will be soon.

[Klippert]

Thanks - i'll try out the new version.
Ok. this is much better, if i set a reasonably safe Z height it correctly pulls up to that height before moving to next probe point.
I successfully did a heightmap of my buildplate. And nothing is ever flat ;)

I did notice a few mishaps.
My probe sometimes "locks up" - so it dosn't let go, despite being pulled up or away from what's being probed.
this causes an alarm 4. - which seems to lock up io-sender. I can click on the settings tab and the app settings, but none of the other tabs want to come to life. the top of the tab just goes slightly purple, and nothing happens. some times io-sender crashes and other times i can switch back and forth between random tabs and wait a little and then it suddenly appears (the grbl tab) where i can unlock and continue life.
Other times i just have to restart. happy path works, but if i encounter an error during probing, it's pretty certain that restarting iosender is the only way forward.

I've no idea why.

Also fusion has a probing function that i'm quite keen on, alignment of an object. i.e. probe x at one point and then move 50mm y+ and probe x again, and let me know if my piece is straight or crooked in on the baseplate/in the vice.
Is that something that would be complex to add?

[Klippert]

is there a way of getting a graphical representation (heatmap) of a heightmap?

…

On Thu, Apr 8, 2021 at 5:49 PM phil-barrett ***@***.***> wrote: I ran a height map on my spoilboard before and after leveling. Was mostly curious. Before leveling, it showed about a 1.5mm deviation across the entire spoilboard. It read within .1 mm after leveling. Not perfect but acceptable for my needs. Note that this does not necessarily mean my spoilboard is flat within .1mm - my X and Y rails could easily be off as well. If so, the milled spoilboard conformed to that. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#286 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AM3JUECUX5GW6HFJVKXMDLDTHXF7TANCNFSM42QF53RA> .

[Klippert]

It would be helpfull to me to at least know the Z-min and Z-max - is ther an easy way of getting those values?

[Klippert]

maybe a Z-avg. or even a std. deviation?

[terjeio]

My probe sometimes "locks up" - so it dosn't let go, despite being pulled up or away from what's being probed.
this causes an alarm 4. - which seems to lock up io-sender.

I'll have to check this.

I think the colors are unitless - blue is height map min, red is max.

Correct.

It would be helpfull to me to at least know the Z-min and Z-max - is ther an easy way of getting those values?

If you save the map you can find the z-values in that. It is in xml-format, if you drop it into Excel you will find the Z offsets in the point column. Or you can open the file in a text editor.

maybe a Z-avg. or even a std. deviation?

I'll add the deviation to the status line when probing is completed.

Also fusion has a probing function that i'm quite keen on, alignment of an object. i.e. probe x at one point and then move 50mm y+ and probe x again, and let me know if my piece is straight or crooked in on the baseplate/in the vice.
Is that something that would be complex to add?

This is on my todo list, another gcode transform. It will not be added to the first production release though.

[terjeio]

maybe a Z-avg. or even a std. deviation?

I'll add the deviation to the status line when probing is completed.

It is already in the status line. Or do you want something else?

[Klippert]

I didnt notice it in the status Window. Ill check. It might be good to have it as a pop-up or in the Console, so it doesn’t dissappear then one absentmindedly starts jogging or something like that.