Advanced RFID System (Uses PIR and Status LED's) - C# .NET source, bs2, schemat

JamesDougherty

INTRODUCTION
On Friday I picked up some parts for something I want to design and I decided to·do·this as a learning project. I have programmed in basic languages before (PICK/D3 BASIC as an example) so I picked up this flavor all right. Of course there is going to be a learning curve, but that’s half the fun. I’m not going to go and build deep blue anytime soon, but we all have to start somewhere. Anyway, on with the project...
·
PROGRAM OVERVIEW
I read that the power consumption for the RFID Reader was 10 in idle mode and 100 in active mode. There are two ways that I know of to enter these states.
·
1)···· Ground the /ENABLE so it always stays in the active state (consumes constant power)
2)···· Control the idle/active states via the DTR pin on the serial port. However, this still requires that it be done manually.
·
This got me thinking and I knew that I wanted to try a different approach so I fiddled around with the idea of using a PIR sensor. The idea was to leave the RFID Reader in idle mode until motion has been detected. If motion is detected then it will set the RFID Reader to active state. If no tag is detected in the specified amount of time them it will put the RFID Reader back into idle mode. Not only will this save a lot of power consumption, but it is completely automated in changing states. Finally, I have LED’s showing when the system is busy, For instance, I have the PIR sensor go through a warm-up process and during this time the red LED will be light up and when it is finished the green LED will light. Likewise, when the system detects a tag the red LED will light up while the tag data is being processed and then the green LED will light when the processing has completed.
·
C# .NET
I wrote a C# .NET class that handles the serial communications. I originally wrote it specific for the RFID, but I expanded it to allow reading, writing, events, etc. so more projects can be built with minimal effort. I attached a simple program to test it and demo it. If you find any bugs or have any suggestions then please let me know. I wrote this in Visual Studio 2008 Professional, but you should be able to compile/run it in the express edition as well. If you can’t then please let me know and I will change it so it will run. I put some comments in the code, but the documentation is still not 100% done and I apologize for that. Feel free to use the source code anyway you see fit, just leave the copyright in place. This code has been tested, but still may contain some bugs so Ariel Productions has no warrenty at this time.
·
PLEASE READ
It works pretty well and I learned a lot from it. This is my first time doing any BS2 programming, my first time doing any kind of hardware design, my first time doing any circuit board schematics, etc. so please do not be too harsh.
·
I included a picture of what my board looks like with everything in place just in case I messed up the circuit board schematics. Like I said, it was my first time doing this so please let me know if you have any questions.


[noparse]/noparse][b][color=blue]EDIT[/color][/b
I almost forogt... I wanted to use a bi-color LED instead of using two seperate ones, but I don't have one at the moment. This would be easy to change so feel free·


Sincerely,

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
James Dougherty

Ariel Productions
http://www.arielproductions.com/

Post Edited (JamesDougherty) : 1/6/2009 3:28:54 AM GMT

sam_sam_sam

JamesDougherty

This got me thinking and I knew that I wanted to try a different approach so I fiddled around with the idea of using a PIR sensor. The idea was to leave the RFID Reader in idle mode until motion has been detected. If motion is detected then it will set the RFID Reader to active state. If no tag is detected in the specified amount of time them it will put the RFID Reader back into idle mode. Not only will this save a lot of power consumption, but it is completely automated in changing states. Finally, I have LED’s showing when the system is busy, For instance, I have the PIR sensor go through a warm-up process and during this time the red LED will be light up and when it is finished the green LED will light. Likewise, when the system detects a tag the red LED will light up while the tag data is being processed and then the green LED will light when the processing has completed.


I like the way you think and how you solved your problem

I like your Project and your ··Very Nice first Project

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
··Thanks for any··that you may have and all of your time finding them

·
·
·
·
Sam

SRLM

I like the idea of using the PIR to turn the RFID reader on and off. However, why not simply go all electronic and use the direct high/low to turn the RFID reader on and off? I'm not a EE, but I'm sure that you could do something with a transistor or two and maybe a couple of resistors. Anyway, nice project.

JamesDougherty

@sam_sam_sam - Thanks. I shouldn't say very first, but rather "real" first project of my own. I did the project out of the book that controls which way the servo turns via the push buttons, but that is about it. The C# wasn't a first though since I have been programming for over 10 years. Thank you a lot though; I appreciate it.

@SRLM - I am still learning all of these terms so I don't know what EE means · If you look at the bs2 source I call low/high enabled, which is actually the /enabled pin on the RFID Reader (I don't do it via the DTR pin). The C# is only a wrapper for the serial port with all of the delegates, error handling, etc. and it is only used to read the output buffer (I mentioned it has the ability to write as well, but that is for future projects). I hope I understood you and if not then I appologize. I am still learning about how all of the components work so I would love if you could explain a little more. Any and all suggestions are great. Thank you.


[noparse][[/noparse]EDIT]
On a side note, don't forget to turn off debug mode in the bs2·before running the C# program. If you have debug mode enabled then it will send that output through the serial which will cause garbage and any tag reads will be ignored. Also, you will most likely receive a port error and possibly a pin changed, which you will see in the message box. So a heads up if you run into this problem (it took me a few hours to track that one down).

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
James Dougherty

Ariel Productions
http://www.arielproductions.com/

Post Edited (JamesDougherty) : 1/6/2009 4:33:36 AM GMT

SRLM

I was thinking about using the output of the PIR directly to power the enable pin without going through any sort of computation. I tired it tonight, but as I mentioned my EE skills are almost non-existent. I'll give it another shot tomorrow after I ask the resident EE guru.

Anyway, the basic circuit would be something like this:



The first half (left side) inverts the signal, since the PIR outputs high when it senses, and the RFID need a low to be active. The second half amplifies the signal. I tried it without the inverter, but I only got 2.8 v from the output. My poor circuit design speaking, or something else? As a side note, the jumper on the PIR allows you to choose a cycle action when it detects: this could be used to save even more power if you're willing to (possibly) hold the card up a little bit longer.

soshimo

You could also use a hex buffer like the 74ls125 - they are cheap and easy to find.

JamesDougherty

@SRLM - Interesting idea. When I get home tonight I will toy around with it and see what I can come up with. What transistors did you try; the 2N3906?

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
James Dougherty

Ariel Productions
http://www.arielproductions.com/

SRLM

I used a 2N3904, but like I said, my circuit didn't work right. I may change that later today (when I experiment more).

soshimo

Forget what I said - I'm getting old and haven't had my coffee yet - for some reason I though they were hooked up with emitter to 5v.

Post Edited (soshimo) : 1/6/2009 7:57:20 PM GMT

JamesDougherty

@SRLM - I am still new to all of this transistor stuff, but what about using the 2N3904 for the PIR output and the 2N3906 for the enabled? Here is my thinking (even if it is wrong):

The 2N3904 is NPN so it would get triggered when the base is high relative to the emitter. The emitter is always 0 since it is grounded so this transistor should always get triggered anytime motion is detected like you thought (since the PIR put out HIGH like you said). Since the 2N3906 is PNP it will be triggered when the base is low relative to the emitter. So when it gets triggered the emitter will push the +5v needed to the collector, which in turn triggers the RFID.

Like I said, this could be completely wrong since I am still learning, but it works in my mind



@soshimo - I'm on my second pot already and I still can't get with it

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
James Dougherty

Ariel Productions
http://www.arielproductions.com/

SRLM

Okay, I worked some more with the circuit, and here's what I've got:



The circuit works when I don't have the PIR in there: I used a jumper between the PIR input on the left and 5V or Ground. When I did, it worked well. However, when I put the PIR in there, it began to cycle high-low with about a 1 second interval between changes, regardless of local motion. I may fiddle some more later...

JamesDougherty

I had something come up tonight so I have been too busy to try this, but I will tomorrow. I had that problem a few times and that is the reason I had the warmup in there, which seemed to helped some. During the warmup I couldn't flintch a muscle or it would cause undersired affects like this as well. What did you have the jumper set at on the PIR? If it's set to high then the PIR will keep reporting HIGH until there is no more motion detected. If the jumper is set to low it will keep going HIGH to LOW so I wonder if this could be a factor. Finally, you may want to play around with the PAUSE calls and make it pause·a little longer as that may help with the sensitivity of it.

Almost forgot... There is a timeout for the RFID reader. If no tags are detected then it will go back to idle mode to help save the consumption. The default was set at two seconds... You could try increasing this and seeing if that helps as well. Like I said, tomorrow I will have some free time so I will try some configurations as well. I need to swing by radio shack tomorrow and get some resistors (like the PN2222, 2N2222, etc.)

Have·a good night SRLM.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
James Dougherty

Ariel Productions
http://www.arielproductions.com/

SRLM

Ah, that might explain some things:

I knew about the jumper, just forgot to check it. I think it's set on the hold high, but I'm not sure (I messed around with it some yesterday). Anyway, the warm up period might be part of the problem. I immediately tried to use it as expected, without giving it the warm up time. That may be the problem.

By the way, this is a completely hardware level solution, so there are no PAUSE statements (no code at all in fact).

I'm not quite sure where you got the RFID reader timeout from. If it's in the code, than that's not an issue here. I looked in the datasheet, and there wasn't anything mentioned. A quick hardware test verified that the RFID reader does not have any built in timeout. Try it: hook up VCC, ground, and tie the enable pin to ground. The red light stays on as long as the power is applied.

James said...
I need to swing by radio shack tomorrow and get some resistors (like the PN2222, 2N2222, etc.)

I think you mean transistors? [noparse]:)[/noparse]

As a side note, one reason why I'm so interested in this is because the PIR would make a nice simple, hardware type project. I was thinking hooking it up to an LED and battery, and have a simple motion detector. It could also be used (for example) to turn on a light when somebody enters a room (via a relay) with no microcontroller. That part is the interesting bit: by cutting the costs $30 or so, more projects open themselves up. Plus, I just want to practice my soldering...

soshimo

@SRLM - I had the same thought. I wanted to hook up the PIR to a relay which will trip to apply power to some flood lights. I'm just not comfortable working with mains power - as close as I get to 120VAC is my surge strip where I plug in my wall wart. When you do come up with something would you mind sharing?

JamesDougherty

@SRLM - Yes I meant transistors, it's been a very long week.·· Anyway, I·am not sure how you are reading the tag information, but there is a timeout the way I had it via the SERIN. SERIN RFID,396,ReadTimeOut,TIMEOUT,[noparse][[/noparse]WAIT($0A),STR TagID\10] If the tag information is not read in the timeout period then it will jump to the TIMEOUT label. This is the part I was talking about and I was not sure if you were doing it this way or not. I put this in because once the RFID was in active mode it would stay in active mode until a tag was read and this is not what I was after. I was thinking along the lines of "Ok, if someone gets close then they may have a tag so we had better turn it on. If this was not the case and no tag was detected then we want to go back to idle mode for the power consumption". Anyway, if you look at the TIMEOUT: part you will see in there that it sets the RFID back to idle mode.

Thats funny you mention that about the lights because I was wanting to try that as well. ·I have not done much soldering either, but I am going to pick up some copper CAD and some etching liquid while I am at Radio Shack today so I can make a little board so I can A) make my first board and sharpen my soldering skills. It's not that hard at all to created a PCB using the toner transfer method so I figured I would give it a shot.



@soshimo - I was kind of weary about that too. God only knows what I am going to burn down or blow up. However, I have some flood lights on my garage that I was thinking about tearing down and tearing apart just to see how they handled the voltage. Giving power straight to the light bulbs is a no brainer, but there has to be some kind of processor in it since it has a timer, the PIR, etc.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
James Dougherty

Ariel Productions
http://www.arielproductions.com/

SRLM

This was the tutorial that got me thinking on how to control a light (Sparkfun has some really good stuff). Just power the relay with the PIR, and you have a motion activated something. Of course, it's only as good as you are moving around, so it might not be best for room lights. What it would work for though is something like a motorized doggy door or electronic exercise equipment.

As I mentioned before, the hardware solution doesn't have timeout, unless you count the emergent behavior when you stay still for a while. So no serin timeout. If you want to implement it, I'd say just stick with code, since I'm guessing hardware timeouts will be a pain (555 timer chip or capacitor probably). Still, I like doing stuff in hardware, since there's less code to worry about getting right (plus, it looks more 'techie' [noparse]:)[/noparse].

JamesDougherty

That's a pretty cool link, thanks for sharing.

I went to Radio Shack today and bought some parts so I can play around with this, but now I just need more time Have you seen this link? Check out the "Time Delay Circuits" section. I think it may have potential if you replace the switch with the PIR output. http://home.cogeco.ca/~rpaisley4/LM555.html#20

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
James Dougherty

Ariel Productions
http://www.arielproductions.com/

soshimo

You beat me to it SRLM. I'm going to give that circuit a try. I think you can probably tie the input directly to the PIR enable output to trip the relay. DIY motion detect flood lights! Oh, how the kids will fear my house next halloween!

sam_sam_sam

James Dougherty and··SRLM

·Please Keep us posted how your testing going with ....>>>>>>>>>


I like the idea of using the ......>>>>>>PIR to turn the RFID reader on and off.


I would like to do something like this my self

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
··Thanks for any··that you may have and all of your time finding them

·
·
·
·
Sam

JamesDougherty

I got it to work finally using soshimo's suggestion. However, I ran into the same problem that SRLM is running in to... The PIR·is too sensitive for it to be effective! All I did was put the output from the PIR directly to the enabled pin (the jumper on the RFID was set to high); no resistors, no transistors, no timers, no nothing.

To test it I used this (the RFID SOUT is on PIN 0):

LoopCounter VAR Byte
TagID       VAR Byte(10)
 
'--Warmup the PIR sensor
FOR LoopCounter = 40 TO 1
  PAUSE 1000
NEXT
 
DO
 
  '--Poll for the tag (Set a timeout so it doesn't get stuck on active mode)
  SERIN 0,396,4000,TIMEOUT,[noparse][[/noparse]WAIT($0A),STR TagID\10]
 
  '--Display the data
  DEBUG CLS,HOME,STR TagID
 
TIMEOUT:
  PAUSE 500
 
LOOP

What is making this so sensitive? Even after the warmup and zero movement it doesn't help

Is there noise (or other disturbance)·in the line that would cause it? I will have to play around with·a timer or something and see if I can't adjust the sensitivity some how. On the flood lights I have there is a sensitivity knob on it. Do you think there is a timer delay·on it that would be something like "If the PIR outputs high for this amount of time then there is positive movement, otherwise it must be a false positive"? Now all we need to do is figure out the sensitivity...

[noparse]/noparse][b]EDIT[/b
soshimo, do you know those poppers you buy around New Years where you pull the string? You should have one of those inside of a ghost or pumpkin and when it detects someone have it kick on a servo to pull the string. That will surely get their attention (they will probably think you are shooting at them or something so maybe you better not LOL)

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
James Dougherty

Ariel Productions
http://www.arielproductions.com/

Post Edited (JamesDougherty) : 1/8/2009 5:06:10 AM GMT

SRLM

Success! A hardware level solution to enabling the RFID reader:


I think I found the problem: when in normal operation, the PIR outputs about 3.3 volts. When hooked up to my transistor circuit shown above, the PIR outputs about 2.2 volts. My guess is that it's drawing too much current. So, I increased the 2kOhm resistor to 10kOhm. Also, I made sure the jumper was in on the H side. When built like this, the circuit works perfectly.

I found that the PIR requried a warmup of about 30 s, so I placed an open book over the module and sat still. Eventually, it settled at no motion. I then removed the book, and waved my hands: Solid red light on the RFID module! Whoho! Once I sat still for a moment, it turned green for about 1s, turned red for about 1s, and then back to green solid. Further tests with the module continuously on did not duplicate this action, so I'm assuming that the PIR has some sort of built in learning over time.

@Sam:
When I get a chance tomorrow(maybe Friday[noparse]:)[/noparse], I'll solder up a small circuit that will toggle a pair of LEDs and post it with the circuit in the completed projects forum. You should be able to replace the LEDs with whatever you want to drive.

On a slight side note, when measuring a resistor with my multimeter to find its value (10 kOhm listed) I got a perfect score: 10.00. No variance! I think this deserves a drink [noparse]:)[/noparse]

SRLM

A couple more notes on the PIR:

1) It seems to work best when the first 30 seconds or so it is covered with a still object. I noticed that without this 'perfect' warm up the PIR was delayed several seconds when changing states. This tendency may disappear over time.
2) The PIR has built in sensitivity training. It waits a minimum of about 1/2 second (guess) before it changes states (either from LOW to HIGH or HIGH to LOW). This should eliminate the need for a 555 sensitivity timer.
3) The only resistor value that appears to matter is the 10 kOhm on the base of the transistor. The other value I tried at 2kOhm and 220Ohm and it still works (for an LED and buzzer)

sylvie369

SRLM said...
This was the tutorial that got me thinking on how to control a light (Sparkfun has some really good stuff). Just power the relay with the PIR, and you have a motion activated something.

I built one of those yesterday. Yup, it's pretty interesting.

A word of warning - there's a dead-end trace (mistake) in the board files provided for that story (it's supposed to go to the diode). I sent them up to BatchPCB (Sparkfun's service), and later got an email from someone there pointing out the problem and asking if I wanted corrected versions. He did send me good working boards, and told me that the person who wrote that tutorial plans to upload corrected versions soon. Of course you could correct it yourself as well.

JamesDougherty

Great job!

I am curious though as to what the advantages/disadvantages are to using the transistors and resistors versus just going straight from the output on the PIR? It works both ways that's why I am curious.

I still want to play around with it some to try and come up with a sensitivity setting for it like to lights have. I have a big business meeting/project coming up so I won't be able to play with it till this weekend, but until then I will research it. However, one thing that did cross my mind was the lights... When the PIR detects motion it triggers the RFID Reader to go into active state. When this happens the LED changes to RED which could trigger the sensor into thinking there was motion. I wonder if this could be one of the things causing the PIR to "toggle" the output. I wonder if it would help to put something in between the PIR and the RFID Reader so·the PIR doesn't see the light.

I did a quick little google and I found a PIR·Controller Chip that has a pin just for sensitivity. The·chip is the KC778B·and you can buy the chip and PIR separately, but they also have a board with the chip already in place.

Take a look here: http://www.electronics123.com/s.nl/it.A/id.72/.f

Here is the board: http://www.electronics123.com/s.nl/it.A/id.408/.f

Do you know if Parallax has any chips like this?


[noparse]/noparse][b]EDIT[/b This is interesting... I took a look at the data sheet for that chip and there is an anti-alias on it to filter the noise like (kind of what I was thinking).·Here is the datasheet for it http://www.electronics123.net/amazon/datasheet/kc778b.pdf

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
James Dougherty

Ariel Productions
http://www.arielproductions.com/

Post Edited (JamesDougherty) : 1/8/2009 2:24:15 PM GMT

SRLM

The transistors and resistors do two things: 1) invert the signal so that when the PIR senses motion it pulls the RFID on 2) it provides more current. My guess is that the PIR can only source a couple of millamps. As for the RFID light, that may be the case, but I doubt it. The PIR uses incoming infrared to create the maps, and LEDs generally have very narrow regions of wavelength emitted. If you want to adjust the sensitivity of the module, you could perhaps put a semi-opaque covering over the PIR.

sam_sam_sam

SRLM

This would be great Thanks

@Sam:
When I get a chance tomorrow(maybe Friday[noparse]:)[/noparse], I'll solder up a small circuit that will toggle a pair of LEDs and post it with the circuit in the completed projects forum. You should be able to replace the LEDs with whatever you want to drive.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
··Thanks for any··that you may have and all of your time finding them

·
·
·
·
Sam

soshimo

@James - I took a look at that datasheet you posted, not sure if that's the PIR you are using or not but it shows an impedance on the OUT pin of 35ohms. You probably need a buffer amplifier. You can use a transistor circuit set in the emitter follower configuration - it's non-inverting with a 1:1 voltage gain. What it does do is amplify current which sounds like what you need to do. Alternatively you could go with my first suggestion of using a hex or octal buffer (which has the same purpose, only in an IC package). You tie the OUT of the PIR to the enable pins of the buffer- then you set the input to one of the buffers to the opposite of your desired output if it's inverting, or the same if it's non-inverting, i.e. if you want the RFID reader to feel +5v then you should tie the input accordingly (GND if inverting, +5v if non-inverting). When I get home later I'll try to whip up a circuit to test my idea - I don't have your hardware but it should work for other devices (such as an LED).

Post Edited (soshimo) : 1/9/2009 6:32:08 PM GMT

JamesDougherty

Thank you for the information. I'm like SRLM as in mainly being a programmer so I am kind of learning as I go and any information is always welcome. I appreciate it.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
James Dougherty

Ariel Productions
http://www.arielproductions.com/

SRLM

A little cross promotion: here is the thread where I posted my completed project of using the PIR as a trigger. Such a circuit could be used to pull the enable pin of the RFID reader originally posted.

JamesDougherty

Indeed. I should have posted the link to it, sorry.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
James Dougherty

Ariel Productions
http://www.arielproductions.com/

Manetheren

I have done the same thing with a laser beam into a photoresistor.· Instead of direct motion it uses the light from the laser to drive the photoresistor and activate the RFID reader when someone walkes infront of the laser.· (as I picked up a couple of cheap 3v lasers).

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Manetheren