PNP Proximity Sensors

[wingnut]

Hello,
Wondering if anyone has successfully replaced the NAIS NPN prox sensors with PNP? This is on an older ShopBot with the g4 upgrade. If anyone has any experience with this or suggestions I would greatly appreciate hearing from you.

I have the sensors on hand so that is why I am going this route.

The replacement sensors are NO (same as the existing) and they function properly at 5V.

Thanks.

[Gary Campbell]

Mike...
The control board is looking for current to be sunk (-5V or ground) therefore the NPN. PNP output source current (+5V) when active. You could use an opto gender changer or a relay to achieve the proper signal. May be best to use the proper speced parts.

[wingnut]

Unbelievably fast response.
Thanks, but thats no fun. Could you expand on the "opto gender changer" idea?

I am really hoping to take something very simple like using the correct parts and turn into a 8 hour project at 5 times the cost

[Brady Watson]

An 'opto gender changer' basically converts NPN to PNP or vice versa.

E.G. - Positive Negative Positive to Negative Positive Negative

-B

[Gary Campbell]

Mike...
You should be able to find an "opto isolator" that, based on the specs of your prox, will change the PNP signal to an NPN signal that is compatible with your board. They can also change NO to NC and visa versa. Since I dont know any of the board or prox specifics (voltage, etc) you will have to do the homework.

I often have <$100 projects that have $10K labor in them

BTW generic, compatible prox can be had for under $20

[dana_swift]

An NPN sensor can be used with nothing more than a pull-down resistor on the output and a diode. The value of the resistor will need to be fairly low, around 200 to 470 ohms. It should pull down to 0 volts, NEVER -5. The diode prevents +12 from being passed to the logic inputs.

The reason people don't do this, is noise causing a false input is far more probable. And the reliability issues cause the NPN preference. Much better noise immunity.

The amount of energy required to fool a digital circuit from seeing a 1 when it should be 0 is different than that required to go the other way.

That is why the sensors are "1" when they are inactive, and "0" when they are active. It seems illogical, but it is more reliable.

But technically a PNP can work.

I suggest unless this all makes sense, just get an NPN sensor. They just hook up and work.

D