Capacitive Auto-Leveling on TAZ 5

How is the optoisolator then wired into the Rambo board?

Vcc - > +
Out-1(3 in your case) - > S (signal)?

I wired it that way but printer doesn’t see it as triggered when the sensor is active.

Check the initial wiring diagram, do a little research on how opto-isolators work, you need to hock it up the the endstop of the z-axis.

Is there a git repository for this firmware? These are great additions and should be tracked! I created a repository here for you: https://github.com/keithjjones/MarlinEnhancementsForTaz Let me know your Github user name and I can give you write access. Thanks for the hard work!

Ok - so I just literally moments ago finished installing the hardware on my TAZ5 for capacitive auto leveling. Haven’t even looked at the firmware yet, thats next.

I’ve been reading this thread, and while I admit I havent read every post (a good number seem to be “my specific issue is this help” - which may not even apply to me, so, skim over unless I run into a problem) I do feel I’ve gotten the gist of things. I have a few questions, just to make sure my understanding is correct on a few points:

There are two types of (auto) leveling - autolevel, and mesh level.
A) Autolevel assumes your plane is flat (unwarped glass) and using a matrix of reads determines the tilt of your assumed-flat bed. This is available in the custom/new/must-self-install-it firmware. Correct?
B) Mesh leveling appears to read a grid of your build surface, and develops an up and down wavy mesh of your warped-you-should-probably-just-replace-me bed. This is NOT supported in the Marlin firmware, even the custom built one, correct?

Want to be sure I understand the options before I start working on anything. I havent yet wrote a Cura plugin, but intend to soon, as I have a few custom things I’d like to accomplish. If the mesh works the way I think, and its not in the firmware, there may exist the possibility to write a Cura plugin that accomplishes mesh leveling by post-processing the G code. In fact it shouldn’t be too hard at all. But, we’ll see how my bed autolevels and performs first before undertaking a possibly unnecessary project.

Also, one more question. The values X_PROBE_OFFSET_FROM_EXTRUDER and Y_PROBE_OFFSET_FROM_EXTRUDER - it does not specify where to measure from. The extruder has a tip, thats easy.

From the tip to what? The closest point of the probe to the tip? The farthest point? The center? If it uses these to make safety adjustments, and you specify the center, you could well run into a corner bracket. If it uses it to make precision adjustments, and you specify the farthest point, it may not be accurate.

How do you measure those?

In case anyone finds them useful, I posted my modifications to TAZ 6 which includes auto leveling:

http://www.thingiverse.com/thing:1499551

Would you please identify what In-3 and Out-3 are?

Is the isolator in place of the resister?

My E2K-X4ME1 looks like this:

NPN
no

Is this correct? for my setup? 1.Brown = Out, 2.Black = In. and 3.Blue=GND Then all I have to do is supply the V’s?

does this look correct? If it were to go on a ramps 1.4

editied: attached different wiring pattern

I used the diagram posted here as a guide:

https://forum.lulzbot.com/t/so-no-more-ao-101-uh/124/1

I think blue is -, brown is +, and black is signal. Don’t quote me until you check it, but that is what I remember. I did a Google to find the common color codes for the wires before I soldered everything together. It worked the first time for me, which was surprising.

Thank you, I’m printing them now.

Hi everyone…i am not using a fancy capacitive sensor. Instead of an optoisolator, I had a spare switching regulator. That seems to be fast enough at converting the high output of the sensor back down to 5v. I’m using a 12v line for the sensor - partly because I already had it running out to the extruder for a servo for my extruder. I’m still waiting on my nice clean not-a-bunch-of-wires-in-a-proto-board PCB from oshpark, though.

I actually tried using a capacitive sensor while building my Taz from scratch, but could never get it to take acurate enough readings, but im guessing its the cheap chinese sensor i bought… ended up going with a servo mounted micro switch.

I’m actually moving away from the capacative sensor. The reason I did it was to get an auto-level on the bed at any point, but, and I cannot speak for the $100+ sensors out there, the sensor I bought is affected by heat. If it lingers near the heated bed too long, it’s reading is inaccurate. So if I print something right after a long print, the next reading isn’t correct.

I’m in the process of building a sled which will hold a microswitch. Also, since I often switch out print heads which are/may be of different heights, I’m building an auto-level for the auto-level. Basically, after I switch extruders, the extruder will pick up the sled, then will home to a microswitch mounted to the rear left of the print bed, and lower until the head itself activates the switch. Once depressed, a stepper on the sled will then lower the leveling microswitch until it touches another known point (another part of the mount that holds the microswitch for the head). Given the height differential between the first switch and the plate the second switch activates on is a known constant, the whole thing can then autolevel on the bed itself with near perfect accuracy.

Then the microswitch can autolevel on the bed, and drop off the sled, and I can print autoleveled. Side benefit - it will save me another 80 grams or so of weight off the extruder.

Hey Ho

just a general comment on the capacitive sensor you can find on the market:

I have to use them in my day job in almost all available variants and over the years found the following:
a) Even if you buy 10 manufactured in the very same batch they are not reacting on the same distance.
b) The detection distance is very much depending on the actual dielectric value of your target. The dielectric value of the target is changing with temperature.
c) The detection distance changes with changes in the supply voltage

All the above is very much independent from the price / manufacturer of the sensor.

I was forced by clients to use capacitive sensors for positioning and learned it is possible but extremely challenging.
In many cases we had to “optimize” the sensor with a different sensing technology (I did not want to say we replaced it :wink: )

To cut it short:
The bed and the temperature has to be stable during the leveling. Ideally the both temperatures should be known and compensated.
If then the supply voltage is kept always the same the sensor should work each time. Not the easiest solution at the end :wink:

All the Best
Frank

I scrapped my capacative auto-leveler. The biggest reason I wanted it was to make leveling a non-issue. Bed not perfect? Who cares. Auto-leveler.

But if I switch from PLA to ABS and the bed is a different temp, my leveling is off. If I do a long print, then do another right afterwards, the probe has heated up, and my leveling is off.

Just switched to an auto-adjusting sled with a microswitch today. Even auto-accounts for changes in head height (swapping out tool heads). So much nicer. Wish I was aware of the heat issues before putting all the work into it.

If you have a nice design, it would be awesome if you shared.

The issue with that is the fact I had to extensively modify the Marlin firmware to make it work. To make the auto-adjustment work I had to add a stepper. Since the Rambo board only has 5 stepper controllers, I had to hijack the second extruder stepper controller. Without using a very custom Marlin firmware, what I did won’t work for anyone else. :confused:

Just saw this video for the Prusa MK2 calibration. It would be great if someone could make this gcode work for the TAZ

https://www.youtube.com/watch?v=q_LqgABOICQ

That live adjustment is interesting, but I’m not the toolhead is using capacitance as the probe… looks mechanical. Well, it looks to be inductive

And not sure how you would write that into the EEPROM. It would be nice if we had an M212 option.

The issue with the capacitance probe seems to be the variance of detection based on different factors.


Maybe an “off-bed” FSR type probing could work… the “at temp” nozzle wouldn’t harm the PEI. Or metal probe points off-bed for inductive probing.

As a side-note of interest, they adopted PEI print bed also. And the 4 zone heat bed is sexy.

I got the BLTouch a while ago and haven’t hooked it up yet but I can’t wait to get it going. I want to get 1 for each toolhead I have, maybe we can set up a group buy to get them cheaper? Google Workspace Updates: New community features for Google Chat and an update on Currents