Re: LM324N op amp and toolchanger sensors

-

> Here is info on the PMDX 120 board;
>
> http://www.pmdx.com/PMDX-120/PMDX-120_Manual_12.pdf
>
> page 7 "contact closure inputs"
>
> "The return pins are common to all 5 inputs but isolated from PMDX
> ground."
>
> "Inputs are opto-isolated and accept contact closure switches or

open-

> collector sensor outputs."
>
> "Over-voltage protection tolerates direct application of 110VAC to
> switch inputs without damage to your computer or the PMDX-120

board."

>
> "The PMDX 120 supplies approximately 12V with an open circuit, and
> will source up to 20mA max to each contact closure pin when

connected

> to the common return."
>
> Does this information change what you said above? I think you can
> tell me what I need to know, you just need to know what I'm working
> with.
>
> I've not yet figured out what supply voltage is needed for the
> circuit (LM324 and 4 opto's).

Okay, in reading the manual I have to make some assumptions on
the "switch closure" pins since the input parameters are not stated.
Based on the statement of "open collector" inputs the optos are
connected with their anode to a plus voltage and the cathode with a
series resistor (or the resistor in the anode leg, it works the same)
so that taking the pin low turns on the opto LED. Since we don't
know what voltage the opto supply is we cannot hook up anything that
will "source current" (i.e. be at a higher or lower potential than
the Opto.

The good news is that the inputs are isolated, the bad news is your
LM324 is NOT open collector and the pin outputs cannot be directly
connected. You will need to add an open collector buffer chip (or
the transistors I mentioned). The transistor setup is simple. 4 NPN
small plastic transistors (2N4400, 2N3904, etc). All of the
emitters are common to the circuit ground (power supply gnd on the
LM324 board). The base of each transistor has a 4.7K to 10K resistor
in series with a connection to the four signal leads from the sensor
card (18K to 24K if the final operating voltage ends up to be 24VDC -
see explaination further down). Each respective collector is tied to
a switch closure pin on the PMDX card. There will be an inversion.
(high signal in = low signal out). Since we don't know if the output
on the sensor card goes high or low when it's sensing we will have to
test and possibly change the polarity of the signal using the PMDX
dip switches to give the parallel port the right polarity. (on the
other hand, your controller software will probably let you define the
polarity so it may not matter). Mount the transistor circuit close
to the sensor card. The leads to the bases should not be run close
to motor or AC wires.

The PMDX switch closure ground goes to the sensor card ground and you
will need a separate small DC supply with it's ground to the same
spot and it's + to the sensor baord.

Now you have one problem left: What voltage supply for the sensor
card? This will take some head scratching. Each sensor will be made
up of an emitter and a receiver (usually UV). If you can figure out
which side is which, you can look for a resistor that is tied to the
positive board supply that is the current limit for the emitter.
It's value will give a clue as to the needed supply voltage. Use
Ohm's law to work bacwards. You have two pieces of the equation: the
resitor (R) value in ohms and you can use 10ma (.01 A). Now use the
equation E = I X R to find the voltage.

Can't find the resistors?

If you have access to a small +5 supply, hook up the sensor card
(before the transistor additions) and using an LED with a 470 ohm
resistor in series connect the anode to + 5 and the cathode (through
the resistor to one of the output pins. Pass something opaque
through the sensor window. Does the LED change from light to dark or
vice versa?. If it does, then you can stop and hook up the sensor
card to +5 and proceed with the transistor part of the project. If
it fails to change (make sure your LED "probe" works by touching the
cathode through the series resistor to the ground; it should light.

If it goes not give any or gives weak transistions change the supply
out for a +12 and the series resistor on the LED probe needs to be 1K
(brn-blk-red). Do the test again. Hopefully it will work at 12V.
The next (and last voltage) to try would be 24 and once again double
the LED series resistor value.

BTW what you are doing with the LED and resistor is a simulation of
what is happening inside the opto, so if you can make it work with an
LED you have most of the problem solved.

Hope you can follow this ramble. If you still can get it sorted out
contact me off list and I will try to get you up and running.

Tom C