Best practices for Infrared Height Sensor

markz · 21 Oct 2020, 12:26

I just finished my build of a Blv Cube and want to get the height sensor working as well as possible. This is my first try using a height sensor and I'm curious what others do to get reliable results.

My bed is a not-flat aluminum 3mm (CR10) bed and on it there's a silicon baking sheet and then a borosilicate glass bed (Creality). The glass bed has a taped PEI sheet on one side and I'd like to put a Garolite sheet on the other side but haven't yet (I have the G10 sheet and tape).

The printer has 3 Z-axis motors/leadscrews so I could level via the motors.

The glass bed has a black dotted surface on one side (the side with the PEI sheet).

Things I've found...

After using the PEI for a while I've depressed parts of it - you can see the dips visually in a reflect-light - and those seem to affect the infrared sensor. If I do a G29 mesh scan on the PEI sheet I see all sorts of minor anomalies - I assume caused by the PEI refraction or physical imperfections or something. I tested to see if the anomalies were related to the dotted bed and didn't find a correlation.

If I flip the bed to the glass side and do a mesh scan it's almost perfectly flat.

My current work-flow is to now-and-then flip the bed over (to the glass), do a mesh scan and ensure the bed is level, then clear the height-map and flip the bed back to the PEI sheet. Once it's set to PEI I do a single G30 in the center before each print.

Concerns:

I'm worried that if I put the garolite sheet on the other side, over time I'll have the same effects as the PEI and I won't have a flat surface for reference.
I would rather not have to flip the bed over to do a height map.
I'd like to use the 3 z-axis leveling but that seems to require the glass surface for consistency.

Possible Solutions

I'm thinking about putting 4 "dots" on the bed corners like Lulzbot and having those produce a reference for the bed leveling. The bed is larger than I need currently.

So...

Does this make sense? Is there a material for the "dots" that would be flat and be a reliable infrared reference point on the glass bed? Am I missing obvious anything?

One other question. I see @dc42 comments on-line about mounting the infrared sensor near the hot-end for accuracy but when I looked up the processor on the sensor it has a very limited temperature range so I don't see how this doesn't fry the board.

Thanks.

Mark

Veti · 21 Oct 2020, 12:31

see
https://miscsolutions.wordpress.com/mini-height-sensor-board/

you can sand the top surface with very fine grit sandpaper until it has a dull matt appearance

Veti · 21 Oct 2020, 12:32

btw there is now a special blv bed

https://www.aliexpress.com/item/33012420368.html?

that surface should work with the ir sensor

markz · 22 Oct 2020, 01:24

@Veti Thanks, I may switch when I'm tired of this setup.

Caveman · 22 Oct 2020, 08:38

that blv bed is garbage, Please save your money and buy a better surface. And a bltouch. I got that bed and it is now in the garbage bin. Bed surface that you link too bends when heating up, and that surface is absolutly not working with the ir sensor.

Veti · 22 Oct 2020, 08:50

@Caveman
good to know.

what are you using now?

markz · 22 Oct 2020, 23:54

What seems like a teaching moment...

I reworked my IR sensor carrier to move the sensor closer to the hot-end and while I was at it I made the sensor board edges more orthogonal to the bed. This seems to produce far more consistent results and noticeably more range than the last less-careful mounting. (The first iteration had oval holes since I wasn't sure where it would end up)

dc42 · 23 Oct 2020, 07:57

@markz said in Best practices for Infrared Height Sensor:

I'm thinking about putting 4 "dots" on the bed corners like Lulzbot and having those produce a reference for the bed leveling. The bed is larger than I need currently.
So...
Does this make sense? Is there a material for the "dots" that would be flat and be a reliable infrared reference point on the glass bed? Am I missing obvious anything?

If you don't need to do mesh bed compensation, then having targets at the bed corners for the IR sensor should work well. Almost any matt finish opaque material will work, for example self-adhesive labels, or masking tape.

One other question. I see @dc42 comments on-line about mounting the infrared sensor near the hot-end for accuracy but when I looked up the processor on the sensor it has a very limited temperature range so I don't see how this doesn't fry the board.

The IR sensor can be mounted with the back of the board quite close to the heater block (e.g. 5mm away) and it won't get too hot, especially if there is a silicone sock over the heater block.

markz · 23 Oct 2020, 12:11

@dc42 Thanks for the info.

markz · 31 Oct 2020, 20:17

Postscript:

I tried pasting four small squares of 25lb paper glued down with Elmer's glue-all (easy to remove with water) at the corners. They worked great... very reliable and consistent.

The problem was that they also read 0.8mm because they rested on top of the transparent PEI sheet -- one day I'd forget something, do a G30 at a non-paper spot, and the nozzle would then launch 0.8mm into the glass bed. It's possible I could put them under the PEI but at the moment - not.

Instead I cleaned those 4 square corner areas on my PEI sheet. They had never been used for printing and I avoid them so they're flat and transparent - nearly as good as a paper square for a 4-point calibration. I determined this by running a mesh, flipping the glass plate over to solid glass and running another then comparing the results.