Duet 3 Scanning Z probe

fcwilt

Thanks much.

I have signed up to get emails.com

Frederick

Notepad

This post is deleted!

deckingman

@dc42 @droftarts @T3P3Tony. On the basis that you would like feedback from the wider community, and a range of different machines, if you want to loan me one, I'd be happy to graft it onto my machine and run some tests. It's currently fitted with a Bondtech LGX Ace / Slice engineering Mosquito and I use the nozzle itself as a Z endstop via a hinged mount and two brass plates that form a switch. That arrangement is extremely repeatable and AFAIK, not at all temperature sensitive so would be a good comparison to see if your scanning probe can be used as a Z probe. I can also print edge to edge (400mm X 400mm) without mesh compensation so having a height map might shut down the naysayers who keep telling me that isn't possible (despite multiple videos showing otherwise). Anyway, the offer is on the table.....

dc42

@deckingman thanks, we would certainly value your input. Do you print direct on a metallic bed? You will need to install the latest 3.5.0-rc1+ firmware; is that OK?

deckingman

@dc42 said in Duet 3 Scanning Z probe:

@deckingman thanks, we would certainly value your input. Do you print direct on a metallic bed? You will need to install the latest 3.5.0-rc1+ firmware; is that OK?

Ref the build plate - not at the moment but that will change. I'm currently printing on 6mm thick glass but intend to try some sort of PEI sheet. Glass has worked well for me but I tire of using 3DLac and times move on. As I understand it, your probe will measure the topmost surface of a metallic sheet, so anything on top of that must follow any contours of the underlying metal sheet for the readings to be valid. By "metallic", do you mean ferrous metal or will it work with aluminium? I'm thinking along the lines of sticking PEI directly onto the aluminium plate but no purchase decision has yet been made so I could use steel backed PEI or some such.

Ref the firmware, I don't mind trying a release candidate as long as I can revert back to the stable version. I envisage making a new aluminium rear carriage plate which will accommodate the new tool board and probe as a well as a part cooling solution that I'm working on. So by the time I've done all that, the firmware might be stable in any case.

dc42

@deckingman the tests we have done have been on flexible magnetic steel plates, but aluminium should work too. As you run your machine in standalone mode you can revert easily to firmware 3.4.6 but of course you will lose the scanning probe functionality (which I know you don't actually need anyway). It would probably be OK to leave DWC at version 3.5.x.

Let us know when you have made a decision on the bed surface.

jens55

@deckingman said in Duet 3 Scanning Z probe:
As I understand it, your probe will measure the topmost surface of a metallic sheet, so anything on top of that must follow any contours of the underlying metal sheet for the readings to be valid.

.... and this is exactly why I have issues with this kind of probe. In early days I installed a sheet of PEI on a glass plate for my printing surface and found out that at higher bed temperatures (IIRC at around approx 90C and higher) the glue sheet lets go and the PEI starts to curl up at the edges. In addition to that, getting an approximation of the surface by scanning the metal underneath does not catch PEI surface irregularities or, more importantly, the occasional air bubble between PEI and glass.
I assume, but have not tested, that both of these issues do not apply to PEI that is directly deposited on spring steel.
The other reason I am still printing on glass is cost - a PEI coated spring steel system can be relatively expensive and is subject to nicks, scrapes or damage from a printhead crash. Setting wrong parameters and getting the print stuck too well on the PEI and you can peel the PEI off the steel substrate.
Float glass is cheap and readily available when you start getting divots on it from too strong adhesion.

R4ffers

@dc42 are there any drawings available please? I'd like to look at making a mount.

deckingman

@jens55 said in Duet 3 Scanning Z probe:

@deckingman said in Duet 3 Scanning Z probe:
As I understand it, your probe will measure the topmost surface of a metallic sheet, so anything on top of that must follow any contours of the underlying metal sheet for the readings to be valid.

.... and this is exactly why I have issues with this kind of probe. In early days I installed a sheet of PEI on a glass plate for my printing surface and found out that at higher bed temperatures (IIRC at around approx 90C and higher) the glue sheet lets go and the PEI starts to curl up at the edges. In addition to that, getting an approximation of the surface by scanning the metal underneath does not catch PEI surface irregularities or, more importantly, the occasional air bubble between PEI and glass.
I assume, but have not tested, that both of these issues do not apply to PEI that is directly deposited on spring steel.
The other reason I am still printing on glass is cost - a PEI coated spring steel system can be relatively expensive and is subject to nicks, scrapes or damage from a printhead crash. Setting wrong parameters and getting the print stuck too well on the PEI and you can peel the PEI off the steel substrate.
Float glass is cheap and readily available when you start getting divots on it from too strong adhesion.

I tend to agree - but with caveats. I have to say that I have no problems using 6mm float glass and I still have the original 3 sheets that I bought several years ago. I do however use 3DLac so some other form of removable print surface might be useful. As you say, powder coated PEI would be preferable to adhesive sheet. But regardless of that, a scanning probe might be useful for tramming the XY gantry with regard to the bed and also for checking the flatness of the (underlying) plate. Any high spots could be scraped or lapped out. That is the approach that I personally would want to take, rather than using software compensation with the inevitable lead screw wear that would result. But that's just me. I've done these measurements before with a DTi but it's a tedious process. So a scanning probe might still be a useful tool, even if the print surface itself is rigid.

T3P3Tony

@R4ffers https://docs.duet3d.com/en/Duet3D_hardware/Duet_3_family/Duet_3_Scanning_Z_Probe#dimensions

R4ffers

@T3P3Tony thankyou

T3P3Tony

@R4ffers the STEP file will go up at some point in the next week or so as well.

gtj0

Hmmm. From the video it looks like the probe has to be mounted "flat" in reference to the bed surface which makes sense but would take up a bit of space. Not that 25mm is a lot but it's "something" . Now if it were a bit bigger, and the center of the coil were open, the hole might be big enough to stick a nozzle through.

Also any limits on the length of the flat FFC cable? My tool platform is very light and I'd prefer not to place the control board on it.

zapta

@T3P3Tony, does the sensor require 'magnetic clearance' around or above it?

Also, is it affected if using individual embedded magnets to hold the removeable bed sheet? (vs adhesive magmatic sheet which has a more uniform static magnetic force).

Notepad

@zapta said in Duet 3 Scanning Z probe:

Also, is it affected if using individual embedded magnets to hold the removeable bed sheet? (vs adhesive magmatic sheet which has a more uniform static magnetic force).

Yes, I believe so.

dc42

@gtj0 you can use different size coils, you just need to adjust the number of turns to get the inductance to be around 16uH. Larger coils can work at greater distances to the bed, while smaller coils provide greater XY resolution.

The FFC cable can be longer than the one we supply, as long as it is secured so that its capacitance to ground doesn't change as the head moves. However the board should be mounted on the tool head. It's very light.

@zapta we found that when using small coil we were able to detect the magnets under the bed, but not when using the larger coils the we are shipping. Preferably, choose the mesh to avoid the magnets. The sense coil should not be mounted very close to large metal parts on the hot end.

droftarts

@gtj0 On the stand at Formnext we also had a development coil board with a hole through the middle. The idea with that one is that you could mount it on the end of a BLTouch, and the BLTouch could probe through the hole in the centre of the coil. Not sure if it has actually been tested, but could then be packaged with the BLTouch.

Ian

R4ffers

Just thinking out loud here, but what if you had a coil mounted upside down next to the bed. Use that one to scan the nozzle. Mount one on the toolhead for scanning the bed. That way it wouldn't matter if anything was stuck to the nozzle. Would it be accurate enough to set Z height that way?

T3P3Tony

@R4ffers I mentioned this in the blog post AFAIR - potentially useful, especially for multi axis printing where the bed autocalibration solution we have right now in Open5X relies on electrical contact. I have not had a chance to do experimentation yet. I may well require a different coil size and possibly more than one coil.

o_lampe

@T3P3Tony said in Duet 3 Scanning Z probe:

I may well require a different coil size and possibly more than one coil.

I thought the same, because otherwise the adjustment-macro would only know the nozzle is off, but not in which direction.
Fullsize metal detectors often have two or three overlapping coils. Wouldn't be a big issue to make a PCB with three layers on one side, I guess.