A simple mod for Passive IR sensors... A toggle switch setting clock to: ON / PIR-controlled / Off

[Robert L]

Hi folks,

This is a *VERY* simple modification I'm making to the PIR sensors on my clocks. I was going to add this as a response to Edwards build thread but thought it might be of more general interest... The mod is simple and I'm finding it quite useful.

I've added a center-off SPDT toggle switch on my Passive IR sensor modules with positions for "On", "PIR controlled" and "Off". The switch allows me to put clocks in my shop space to sleep when I'm working for a time with my back to the clocks... allows me to easily test PIR response on controllers, etc. 

See the attached photo for the details. Label added for clarity.

Note that the PV Electronics PIR module case top piece shown has an extra hole added above the second PIR module trim pot to allow me to adjust the module hold-on time. I've retrofit some of my clocks with HV power supply disable lines under PIR control. These modified clocks need the PIR module to control on-time when activated. I also change a cap from 10 nF to 100 nF on the PIR module I use extending the max hold-on time to about 42 minutes... the cap is connected to pin 4 of the BIS0001 IC. Modules differ in layout so I won't post a photo showing the cap location... The BIS0001 data sheet details the function of this cap.

Best regards,

Bob

[gregebert]

Most PIR sensors should  run on 3.3V; most likely yours is one of those  'jelly bean' units available from dozens of sellers on Ebay/Amazon.

If there are 3.3V devices in your clock, there will be a 3.3V source somewhere; In my case I provide +5V to a RasPi, but it's onboard 3.3V regulator's output is exposed, so it's used to power the PIR sensor, which requires less than 1mA of current.

[Robert L]

A friend made a good point by private e-mail... The mod as originally shown will potentially place +5 or GND directly on a processor GPIO pin. Not a good idea...

All of my devices have a series resistor on their PIR input pins - typically 10k. In most cases there are ESD clamps on my designs as well. A pull-down to assure no false triggering is also present on some of my clocks. One or more of these are quite possibly not the case for devices you are using.

For this reason, consider adding a series resistor on the output of the modified PIR detector downstream of the switch. Look at the input circuit for any device you are driving with the modified PIR module and decide what value of output resistors on the module make sense for your device. Consider current driven into the receiving device - especially through input protection diodes. Also watch out for voltage dividers that may be formed with input networks on the target device - as an example, there might be a pull-down or pull-up on the connected device input to prevent false triggering. 2K into a 10k pull-down meets minimum voltage high and low levels on one of my clocks. (Try sending PIR and true RS-232 down a shared cable... The cap coupling from GPS data edges in the cable can be enough to trigger a PIR input without a modest pull-down. Ask me how I know?)

B