Need help with npn n/c proxy sensor wiring!!

Hi everyone,
i am new to this forum and really need some help to connect my proxy sensors.

I have the following breakout board:
DB25-1205 ( powered from external 5v DC psu)

https://forum.linuxcnc.org/media/kun...5DB25-1205.pdf

proxy sensors are the following:
Lj12A3-2-Z/AX NPN NC 3 Wire DC 6-36v

my plan is to use a 12v external psu to power all the 3 sensors which I intended to use for homing on my 3 axis. No limit switches will be added as I have set the soft limits. The reason I am changing to proxy as the traditional mechanical switches failed few times due to vibration when I was running a bigger tool on higher speed.

my plan was to either use:
- one sensor / input
-or all 3 sensor in series to one input.
I am using pin15 for e-stop so both option could work but i would like to make it as simple as possible.

after some research i've found if all 3 connected in series I need to use diodes and resistors which is not a problem. or if I use individual pins no need according to few related post.

if anyone could help me shed some light on this would be much appreciated, thanks.

I asume by connecting in series you mean connecting the GND for sensor 1 to the output of sensor 2, the GND for sensor 2 to the output of sensor 3

If you measure the voltage on the output when this NPN NC proxy is activated, you will measure the supply voltage. This is because most of these proxy's have an internal pullup resistance. You power these proxy's using one PSU. Just make a drawing for the 3 proxy's and their pullup resisance the way you want to connect them in series. Maybe that sheds some light.

The current of sensor 1 and 2 would flow through the output of sensor 3 and the pullup current of all 3 sensors would flow through the output of sensor 3. That whould exceed the max current of sensor 3.

Proxy's aren't the same as ordinary switches and you can't connect them in series but you can connect them parallel.

If you want to use 1 input for all proxy's, you need NO proxy's connected parallel (al the outputs connected together to the input of the board).
Beware that the input on your board should handle the 12 Volt (proxy pullup currents) of all 3 proxy's simultaneously! If the board is designed for 5 Volts, you need a resistor (optocoupler) or diode (digital input).

with this BOB

Attachment 433116

you need to use sensors with open collector NPN outputs


I have only seen N/C PNP sensors wired in series

As you have enough inputs , I would connect 1 N/C NPN sensor to 1 input



before using a 12V supply for your sensors

check they have an open collecter with no internal pullup resistors between the output and pesitive supply

simple check just power a sensor and check the output between the negative supply and output is all ways 0V


If the output goes positive when the sensor switches off

you need to add a diode between the sensor output and BOB input -

the diodes cathode going to the sensor output

this will stop any current flowing out of the sensor and into the BOB's input

Attachment 433118

one problem with normally open sensors is the control can not detect a broken sensor or wire

John

thank you guys,
I will measure the voltage tomorrow when im at the workshop :)
And as mentioned due to enough inputs I will stick to 1 sensor /input.
I am not sure about the internal pullup resistor, I will try to find some data sheet on them and also measure the voltage to be safe.

I used to have N/C connected mechanical switches 6 in series wired in to a single pin in Active High (normal state), so any failure with wiring or switch can be identified quicker. But since the upgrade I knew I will need a bit of research on the proxies as the need extra power.
Thanks again :)
will post some results how things went!

btw:
I have some 1N4001 X43 Diodes laying around
(it was used across 12v supply but i have few left from different project)
would this do the job or shall I use 1N4732 zener ? as I remember my bob takes only 5v input signal and the zener is regulates to 4.7v as far as I'm aware.

As you mentioned GND to out and so on what I was thinking at first.

If the Lj12A3-2-Z/AX NPN NC sensor has a an internal pullup resistor

connect the sensor output to the BOB input via a 1N4001 diode just to make sure its not damaged

the reverse breakdown voltage of the LED inside the BOB's opto-isolator is about 5V

with the BOB's input open circuit
if the sensor has an internal pullup resistor the cathode will be connected to +12V and anode to +5V via the
sensors pullup resistor and BOB's 330 ohm resistor

Attachment 433240

John


PS
not sure why the attachments in post 3 are not visible

thanks again John great info, really appreciated!
:)

All worked for first try was super happy, until I turned on the vfd, which started to activate the limit switch trigger in the software.
I’ve connected the shielding of the cable to earth and when the motor is not in the holder running smooth, but as soon as in the frame cuts out. Seems to me I’ve got some induced voltage on my cables or something else.... the vfd running on its own not yet hooked up with the BOB so not sure at this stage... :(

how are you powering the BOB?

Attachment 433318

if you have jumpers JN1 & JN2 installed so the same 5V supply is connected to both

Vcc & Vdd that powers the input side of the opto-islators and
+5V & 0V that powers the PC side of the BOB

noise picked up on the proxy sensor wiring has an easier path into a PC printer port or motion controller
than it would with the jumpers removed and a second isolated 5V power supply used
just to power the Vdd & Vss terminals

that way you have galvanic isolation that's lost with the jumpers installed

John

Hello John,
I have a separate dc 5v for the bob as I was thinking better this way than the pc powering it.
But I have to check the jumpers as I can’t remember.

First of all Merry Christmas to all :)

I have checked the bob JN1&JN2 was on.
so i have removed the jumpers and and fitted a 2nd dc5v 1.5A psu which solved the problem but another appeared.

basically I have 4 nema32 hooked up x,y,z,and A
Y&A are linked together and when I run the motor now the stepper motor on A axis start to move very slowly :/
nothing on the Y just A and after stopping the motor A still active seems like now I have induced voltage on this output. I am not sure it is the output or wiring at this point. I have more output which I could eventually swap if thats the case, but if anyone have any idea what could cause this would be a lot of help.
thanks :)

I have just checked with my multimeter, and I have about 100v AC induced between the motor housing and the cnc framework, if I ground the frame to the vfd ground terminal it drops down to 45v AC. Any thoughts how to get rid of this completely ?
Also fitted a ferrite ring magnet on all my motor wiring and limit switches but still no joy :/
After adding a layer of silicon washer and electrical tape around the proxy sensor mount the voltage reduced to 20v ac

Quote:

---

Originally Posted by Jprorok

I have just checked with my multimeter, and I have about 100v AC induced between the motor housing and the cnc framework, if I ground the frame to the vfd ground terminal it drops down to 45v AC. Any thoughts how to get rid of this completely ?
Also fitted a ferrite ring magnet on all my motor wiring and limit switches but still no joy :/
After adding a layer of silicon washer and electrical tape around the proxy sensor mount the voltage reduced to 20v ac

---

It sounds like you have something connected to Ground that should not be , post some photo's of your wiring may be able to see what is wrong

one problem with a machine built from anodised aluminium parts is the aluminium oxide film will act as an electrical insulator

I have seen a few photos of Chinese machines with earth / ground wires screwed to the inside of the control box
and fail to make a connection !


you have to remove the anodised surface to make a connection to the aluminium metal underneath


check the continuity from the earth / ground terminal on the VFD to the motor cable earth / ground wire

then to the 4 pin spindle motor connector

then check the continuity from the earth / ground pin is internally connected to the motor case
it can be missing on Chinese spindle motors !

John
PS

saved from a post here on cnczone

Attachment 433718

I guess if you connect in series you mean connecting the ground of sensor 1 to the output of sensor 2, and the ground of sensor 2 to the output of sensor 3. When this NPN NC proxy is active, the power supply voltage can be measured by measuring the output voltage. Because most of these proxies have inner tensile strength, they do. In this case, I would try to use a residential proxy and it should solve the problem. This will fix the connection that shows up as an error.