X5SA Z-probe now drastically shorter trigger distance.

DevoCut

This sensor was in the upgrade pack for my x5s

https://www.aliexpress.com/item/TRONXY-3D-printer-accessories-Auto-leveling-sensor-with-mount-bracket-for-PRINTER-3D-DIY-kit-part/32725703925.html?spm=2114.search0104.3.2.30b95669sZO8KM&ws_ab_test=searchweb0_0,searchweb201602_6_10065_10130_10068_10890_10547_319_10546_317_10548_10545_10696_453_10084_454_10083_10618_10307_537_536_10059_10884_10887_321_322_10103,searchweb201603_61,ppcSwitch_0&algo_expid=3e800667-3592-44ad-a241-c699cc54a953-0&algo_pvid=3e800667-3592-44ad-a241-c699cc54a953

It worked great on the orginal board that came with the upgrade, the Aluminum head bed + plastic sheet + buildtak like surface and I still got probably 2mm of trigger distance on top of all of that. I'm trying to convert the printer over to the Duet 2 wifi (on firmware 2.02(RTOS)) I read someone else had suggested to wire the power and ground to an available always on fan port, then wire the signal into pin 1 on the z prob. But once I got it working, the trigger distance is maybe 1-1.5mm from the heat bed so I can't fit the plastic build plate, or the new glass plate I just bought. Could this be an issue that the ground is too far (electrically speaking) from the signal wire and that affects it's trigger distance? Or is there another mode I should be using?

; Z-Probe
M574 Z1 S2 ; Set endstops controlled by probe
M558 P5 H5 F120 T6000 I1 ; Set Z probe type to modulated and the dive height + speeds
G31 P500 X0 Y0 Z2.5 ; Set Z probe trigger value, offset and trigger height
M557 X15:195 Y15:195 S20 ; Define mesh grid

On the sensor it shows up as a reading of 0 or 1000

dc42

The only reason I can think of for the trigger distance changing, assuming you haven't changed anything mechanically, is if the supply voltage to the probe has changed. What supply voltage were you using before, and what are you using now? If you are powering the sensor from an always-on fan output, the supply voltage now will be either VIN or 5V depending on the position of the V_FAN jumper.

DevoCut

@dc42

Good point, I know the sensor does 6-36v, I don't know what the older proprietary board was outputting, but it would be easy to probe that and see what it is, thought I imagine it be 12v because the whole board is 12v. But I can double check that fan jumper and see if it is set to 5v. I didn't change it and figured it would come from the factory set to VIN. But that gives me a few things to try out. I'll check it out tonight and report back.

DevoCut

@dc42

So it does seem that the sensor is running off 12v, (probed the wires in the connector it's self) do inductive sensors just go bad like that? or is there another possible solution?

DevoCut

For fun I hooked up an old LJ18A3-8-Z/BX 8mm sensor I had lying around in the same configuration, and while I don't expect 8mm on an aluminum heatbed, I only get around 3.6mm activation distance, which is just barely enough for the plastic sheet, but the glass is closer to 3.8mm so that's not possible. Is this normal? I'm going off the LED activation lights at this point just to try to narrow down any software issues. I'm just surprised because the tronxy sensor seemed to work just fine on the proprietary 32bit board, and I've seen other on this forum mention they're switch Duet 2 install worked with this sensor.

Do I just spend the money for a bltouch at this point? I'm kinda getting frustrated with my lack of knowledge on what's happening.

Veti

your bed is made out of aluminum. that is part of the reason for the reduced trigger distance.

you can give this a try. it can be build very cheaply. and its compatible with the bltouch settings
https://www.thingiverse.com/thing:3303618
i found it to be very accurate (sd of 0.001) and not prone to interference like the bltouch due to the optical sensor.

DevoCut

@veti I know the trigger distance will be shorter than spec do to the aluminum, What I'm saying is the only thing that changed was from using a tronxy 32bit board, to a duet board, and I'm getting much shorter trigger distance.I didn't swap the build plate or do anything different.

timcurtis67

@devocut said in X5SA Z-probe now drastically shorter trigger distance.:

Do I just spend the money for a bltouch at this point? I'm kinda getting frustrated with my lack of knowledge on what's happening.

I set one up on one of my printers and it works great. I ordered another one to put on my second printer. I think they work really well. I get very consistent touches.

dc42

@devocut said in X5SA Z-probe now drastically shorter trigger distance.:

@veti I know the trigger distance will be shorter than spec do to the aluminum, What I'm saying is the only thing that changed was from using a tronxy 32bit board, to a duet board, and I'm getting much shorter trigger distance.I didn't swap the build plate or do anything different.

Are you definitely probing at the same point with the Duet as you were using the other board? If you have entered probe XY offsets in the G31 command in config.g, the firmware on the Duet will be taking those into account.

Assuming that the probe produces a digital output that corresponds to the LED being on or off, I can't think of an explanation for a difference in trigger height other than using a different voltage, probing at a substantially different temperature (inductive probes are a little sensitive to temperature), or probing over a different point. It's well known that using inductive probes with glass-over-aluminium beds is problematic.

DevoCut

@dc42 The XY offset probably isn't the same, but because I was trying different probs I'm basically resting the prob on the bed it's self. And with the Plastic sheet over the heatbed it doesn't respond on matter where I try on the platform. You might be right heat, I'll try preheating the bed, I can't honestly remember if the old controller did a preheat on the bed before it did it's mesh measurement. I don't think it did but I'm not sure so I'll test. I know typically the glass over aluminium is problematic, don't quote me on this, but I think this probe is a different style. I think in it's circuit it's alternating the current flow to produce a magnetic field so less ferrous materials react stronger? I maybe don't have the science right but I've seen a lot of these cheaper printers start using this smaller style sensor and get better results than the old 8mm sized inductors. Reliable enough that they all ship the printers with plastic or glass on top of they're beds and people seem to have good results using said sensor.

I already bought a bltouch and I'll go forward with that since it seems to be better documented and more common configuration for this controller. But for the sake of education I may put together a video if I gain any knowledge from testing both boards.

DevoCut

Just to update the status in case someone comes across this post. I'm installing the bl touch now, which should work just fine, but to prove I'm not crazy I hooked the x5sa board up one more time just to make sure it wasn't my imagination, you can see a video of the results here:

https://youtu.be/x-KrowYSX28

You'll notice the duet board does respond to metal just fine, but even if I run the sensor into the plastic + aluminum bed, it won't trigger. Then I switch back to the x5sa board, and you'll see there's somewhat acceptable clearance in the same situation. Both tests had an unheated bed, running off 12v, on the same xy of the build plate.

I have no idea what's going on, but I'm tired of trying to chase down the blackmagic going on here, so I'm installing the bltouch and setting that up per the guidelines in the duet wiki. If anyone has any idea what's going on though I'd be edger to learn.

dc42

Is this with the bed heater on or off? Do you know whether the sensor is inductive or capacitive?