Inductive probe on 24v SKR 1.4

Hi all,

So I've been planning the conversion of an old commercial closed-source machine to open source in order to use it with Marlin and custom parts. Including an inductive sensor (since it has a cast iron heatbed, this would work nice).

The thing is that I am finding a lot of good info on 12v sensors with a voltage divider using a diode. And I did this with lots of success on two of my other printers as well. The thing is, this commercial machine rebuild will run on 24v and this means I'll have to find a way to do it in another way. Now, reading about it I came across this user's comment:
[mertarauh.com]


Will this indeed be a good solution? And besides that I remember the SKR 1.3 has 10K pullup resistors on the endstop input that cannot be disabled. Does anyone know if the SKR 1.4 has the same problem or can they be disabled in the firmware this time around?

I couldn't find a definitive answer on this through google but did find two topics on thingiverse and stackexchange so figured it might be interesting to hear what forum members have to say about this topic.. I do believe the diode option is the cleanest and it has worked for me in the past. So if anyone can confirm the option discussed in the screenshot is a safe one... How would I achieve that?

Thanks for all your wonderful knowledge!

---

http://www.marinusdebeer.nl/

Replacing the 1N1848 that is listed in the blog post that has a 0.7V or so voltage drop with a 1N5818 Schottky with a ~0.35V drop may make it work. I'm sort of surprised that the sensor is only dropping the voltage to 0.7V - it should be much closer to 0.

Don't try to use the more common 1N5817 Schottky - it's reverse voltage rating of 24V is too close for comfort.

Edited 1 time(s). Last edit at 07/07/2020 07:45PM by MMcLure.

The SKR 1.4 has the same hardwired pull-ups on the end stop inputs, but does not have one on the probe input. You can see the schematic at [github.com] for more details.

Quote
MMcLure
Replacing the 1N1848 that is listed in the blog post that has a 0.7V or so voltage drop with a 1N5818 Schottky with a ~0.35V drop may make it work. I'm sort of surprised that the sensor is only dropping the voltage to 0.7V - it should be much closer to 0.

Don't try to use the more common 1N5817 Schottky - it's reverse voltage rating of 24V is too close for comfort.

Thanks for your advice!! Do you think this one could work as well??
1N5819 1A 40V SCHOTTKY DIODE DO-41

This is the list that I have currently in my possession:

1N4148 switching signal Diode DO-35
1N4007 1A 1000V Rectifier Diode DO-41
1N5819 1A 40V SCHOTTKY DIODE DO-41
1N5399 1.5A 1000V 4.8W Rectifier Diode DO-15
FR107 Fast Recovery Diodes FR107 1A 1000V DO-41
FR207 2A 1000V Fast Recovery Diodes DO-41
1N5408 Diode Rectifier 3.0 amp 1000V DO-201AD
1N5822 40V 3A SCHOTTKY DIODE DO-201AD

Edited 1 time(s). Last edit at 07/11/2020 04:20PM by Ohmarinus.

---

http://www.marinusdebeer.nl/

Okay, so a little update, I decided to use this schematic:

Source: [www.thingiverse.com]

But with a 10K resistor on the U1 pin instead of a 470 Ohm resistor.

Now the voltage is successfully divided between the optocoupler and the inductive probe. It pulls both 5v and 3.3v down to zero volt.
The optocoupler I used is a PC817 which I got for free from the local electronics shop because I know the guy and they probably only cost 1 cent anyway. (I happen to have ordered a bag of them from Aliexpress as well because I think these optocouplers are quite interesting to play with and also isolate the other probes that I am planning on using with my other 3D printers).

I have not tried out the diode option because this optocoupler solution, however it being overkill, is really elegant. I love it. And I can turn it into a schematic so I can make it plug&play.

---

http://www.marinusdebeer.nl/

... hmmm ... if the VO level ist +24V, then be sure, there is an internal resistor from the OUT1 pin to GND! - if it's an "floating" input circuitry, then with "OFF" transistor it will be pulled to 24V throug the 10k resistor!

If "floating", then add a 2k7 resistor from OUT1 to GND to define a safe signal level ...

Edited 1 time(s). Last edit at 07/11/2020 06:01PM by VDX.

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Quote
VDX
... hmmm ... if the VO level ist +24V, then be sure, there is an internal resistor from the OUT1 pin to GND! - if it's an "floating" input circuitry, then with "OFF" transistor it will be pulled to 24V throug the 10k resistor!

If "floating", then add a 2k7 resistor from OUT1 to GND to define a safe signal level ...

No worries, the right side of the coupler only has the voltage that is supplied from the board itself! It works with either 5 or 3.3v

The 24v is only on the left side of the schematic.

---

http://www.marinusdebeer.nl/

Quote
VDX
... hmmm ... if the VO level ist +24V, then be sure, there is an internal resistor from the OUT1 pin to GND! - if it's an "floating" input circuitry, then with "OFF" transistor it will be pulled to 24V throug the 10k resistor!

If "floating", then add a 2k7 resistor from OUT1 to GND to define a safe signal level ...

Indeed, it is very simple. I was looking at the SKR 1.4 schematic as I intend eventually to buy one.

Most probably the output of this probe is an open collector NPN and in that case it needs close to nothing, just implement the same protection circuitry found on the X, Y endstop as the probe doen't have anything, just direct to the CPU. In case Z endstop is used (It is what I do) just put protection diode as it is missing (strange ?). Or use the ExDET input and configure firmware.

Yet the best is not tho use this archaic method and have the detection done by the nozzle touching the bed (piezo, strain gauge whatever...).

---

"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)

Quote
MKSA

Quote
VDX
... hmmm ... if the VO level ist +24V, then be sure, there is an internal resistor from the OUT1 pin to GND! - if it's an "floating" input circuitry, then with "OFF" transistor it will be pulled to 24V throug the 10k resistor!

If "floating", then add a 2k7 resistor from OUT1 to GND to define a safe signal level ...

Indeed, it is very simple. I was looking at the SKR 1.4 schematic as I intend eventually to buy one.

Most probably the output of this probe is an open collector NPN and in that case it needs close to nothing, just implement the same protection circuitry found on the X, Y endstop as the probe doen't have anything, just direct to the CPU. In case Z endstop is used (It is what I do) just put protection diode as it is missing (strange ?). Or use the ExDET input and configure firmware.

Yet the best is not tho use this archaic method and have the detection done by the nozzle touching the bed (piezo, strain gauge whatever...).

Honestly in the build I'm doing the inductive probe is perfect for it's application. However, I did order a precision piezo kit last week and hope to receive it soon so I can at least install that on the Delta. And I bought an imitation piezo kit from Aliexpress to compare the kits and their quality etc. I might make a through-hole piezo disc and design a hotend mount for the current project. But first I want to use the inductive probe because it's what I know and I like it so far on my other printer

The probe I'm going to use for this project is a PNP NO probe. Are there any drawbacks by using a PNP NO probe??

---

http://www.marinusdebeer.nl/

Quote
Ohmarinus


Honestly in the build I'm doing the inductive probe is perfect for it's application. However, I did order a precision piezo kit last week and hope to receive it soon so I can at least install that on the Delta. And I bought an imitation piezo kit from Aliexpress to compare the kits and their quality etc. I might make a through-hole piezo disc and design a hotend mount for the current project. But first I want to use the inductive probe because it's what I know and I like it so far on my other printer

The probe I'm going to use for this project is a PNP NO probe. Are there any drawbacks by using a PNP NO probe??

Too bad it is a PNP !
A NPN pulls the output to 0V which is easier to interface with low voltage electronic.A PNP pulls it to the probe power supply, here 24 V !!! So you need to clamp it to 3.3V. Just a 6.8K resistor in serie and connected to a protected input like E0DET or E1DET to produce about 3V as input V. Then define this as probe pin. To use the probe pin requires to protect it as E0DET.

Edited 1 time(s). Last edit at 07/12/2020 12:13PM by MKSA.

---

"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)

Quote
MKSA

Quote
Ohmarinus


Honestly in the build I'm doing the inductive probe is perfect for it's application. However, I did order a precision piezo kit last week and hope to receive it soon so I can at least install that on the Delta. And I bought an imitation piezo kit from Aliexpress to compare the kits and their quality etc. I might make a through-hole piezo disc and design a hotend mount for the current project. But first I want to use the inductive probe because it's what I know and I like it so far on my other printer

The probe I'm going to use for this project is a PNP NO probe. Are there any drawbacks by using a PNP NO probe??

Too bad it is a PNP !
A NPN pulls the output to 0V which is easier to interface with low voltage electronic.A PNP pulls it to the probe power supply, here 24 V !!! So you need to clamp it to 3.3V. Just a 6.8K resistor in serie and connected to a protected input like E0DET or E1DET to produce about 3V as input V. Then define this as probe pin. To use the probe pin requires to protect it as E0DET.

Strange, as I used this probe on an SKR 1.3 previously without any issues. The only difference is going from 12v to 24v PSU. I will play around a bit. As I also have an 18mm thick probe which is NPN, but this one it way too beefy to use on the printer hahaha I guess I'll then use the 18mm thick one until I receive the new NPN 12mm thick probe that I ordered. The circuit I built now works perfectly with the 18mm thick probe. Eventually I want to get a Pinda-sized probe. Nice and thin.

Edit, decided to order this one as a future replacement. Looks like it's perfect!
trianglelab M8 inductive proximity sensor DC5V 3-wire 2mm for 3D printer Z probe auto bed leveling CR10 ENDER3

Edited 2 time(s). Last edit at 07/13/2020 11:56AM by Ohmarinus.

---

http://www.marinusdebeer.nl/