Junebug help??

Status
Not open for further replies.
It is a great learning tool

In the short time I have had the unit it has worked great. Just wish the directions on usage were a little bit more descriptive.
I learned that you turn off all the di switches to program an in circuit PIC
I need to use it more often. Just as soon as I finish up building and testing my CRITTER RIDDER proto type.
Tested it today, seems to work.Want more testing.
found several small problems. Had to change a line of code, found several questionable solder joints, need to make some small revisions to pc board lay out.
 
CONGRATULATIONS. Your long correspondance served you well. Nice to note.
Perhaps it may be time to close this post and start fresh one for newer designs.
 
I am contemplating adding a deer counter

to count how many times the PIR is triggered.
but a method to proceed.
assuming I can use the 18F1320 memory? or add additional memory?? with a hearing aid battery as backup power to keep the memory alive??
any thoughts
 
I assume the processor is either sleeping or running at a low clock rate when not generating sounds. No need for external memory or batteries. Normal RAM will not be lost unless the batteries go dead.

Set a counter to zero on startup and bump it once each time the PIR is triggered. Have the user press a switch which will have the uC blink out a code to indicate the count.

If for some reason you want to keep the value when the power is gone store it to EEPROM.

================= but ===================

This is starting to look like feature creep. I suggest you focus on getting the unit into the hands of your customers.


MrDEB said:
to count how many times the PIR is triggered.
but a method to proceed.
assuming I can use the 18F1320 memory? or add additional memory?? with a hearing aid battery as backup power to keep the memory alive??
any thoughts
Click to expand...
 
PIR is triggering in daylight

indoors the unit will shut off but outside it keeps triggering.
going to add a CDS cell bu for right now I am going to insert some code to trigger the unit every 15-20 minutes. Just for test purposes until I get the PIR working right.
now for code. Will research on swordfish but any help would be appreciated.
looking at a foot step detection using homemade geophone
Circuit - Human Traffic Footstep Detecto - Circuits designed by David A. Johnson, P.E.
Circuit - Earthquake Alarm Hobby Circuit
 
is there a better way to time event?

working with a code snippet on Junebug.
GOAL = to insert a timed event (take pin 9 RB1/INT1) LOW FOR A FRACTION OF A SECOND TO TRIGGER THE OUTPUT ON PIN 18(RB3/CCP1)this simulate the PIR until I can get the unit to work in daylight. TESTING effectiveness of unit in relation to repell deer from ones garden.
using Junebug to test code by turning on one LED(porta.0) for 20minutes then turning off for 10ms.
present code is LED on for 40 seconds then off for 10ms(just guessed on the 40 seconds)
expanding the 40seconds on to 20 minutes on (no real accuracy needed, just about 15-20 minutes between events(PIR simulation)
QUESTION = a more efficient method to time this sequence?
Code: 
Device = 18f1320
Clock = 8
Config OSC = INTIO2, WDT = OFF, LVP = OFF
OSCCON = $62
TRISA=%10111110         //A0 & 6 output
PORTA=1
While (true)
    Toggle(PORTA.0)
    DelayMS(100)
    toggle(porta.0)
    delayms(20000)
Wend
End
 
milli second math?

am I assuming correctly that 1 ms = 1/1000 of a second
in my code snippet I have 20,000ms equals about 40 seconds
so if I want 20 minutes then 600,000ms = 20 minutes?
OR
is there a more efficient method of deriving at 20 minute delays?
after I get this snippet into my PIC for real life testing I plan on adding a cds to control the unit in daylight (triggers every 20 minutes) then at night switch to the PIR.
this is my next code endeavor but for now going with every 15-20 minutes.
 
QUESTION why is timer better?

more efficient or just more accurate?
I really could care less about the accuracy.
Perhaps the PIC would get "lost" with say a 600,000 ms delay??
will take a closer look at the code snippet but it looks long
 
more efficient or just more accurate
Click to expand...
That's not it your delay will set and count for 20 min. And that's all it will be doing till it's done. Timers are better for that. You don't have to use all that code It just shows you how it works. Set up the timer. That's the part you need .
 
You'll have to shield it. PIR's don't like direct sunlight, CDS cell isn't going to help with motion sensing.
 
if I shield the PIR then it won't detect

my new plan is to set up a timer that triggers the unit(PIC) every 15-20 minutes then have the cds or photo transistor enable the PIR when it gets dark out side.
posted a short snippet of code but be80be suggests setting up a timer instead of a delay.
trying to figure out the timer?
do I just set up an interrupt , set up a TM0 on RB2/INT2 (RB0/INT0=dip switch output and RB1/INT1=PIR input) so RB2/INT2 (pin 17 on the 18F1320)is only int pin left?
a timer is the better choice over a delay? I assume.
going to play with Junebug to try and get some insight??
 
Shield the PIR from the sun, not from the view.. a simple curved plate over the very top of the sensor. It'll still detect leg motion, and won't go off for birds.
 
I have the top shielded

Are there different PIRs to work in sunlight?
for now I plan on just instituting a 15-20 minute timer for daylight.
I contemplated a DIY geophone. not sure if it would pick up a deers footsteps.
 
getting lost fast

just want to add a 20 minute timer Is this method workable?efficient??
the red are the additions to existing code.
Code: 
[COLOR="Red"]PORTB.1=0[/COLOR]//triggers unit to sound
If PORTB.1=0 Then //if button 1 pressed add PIR=1 here pin 9
For i = 1 To 200 //play 20 tones
    Tone=Rand(5) //each tone is random frequency
DelayMS(Speed) //and for 1.00 seconds
Next //end for loop
[COLOR="Red"]TOGGLE PORTB.1
delayms(600000)//about 20 minutes[/COLOR]
Else //otherwise
     Tone=5 //silence
     i=Rand(255) //make rand more random
EndIf //end if condition
Wend //end of while loop

this is a temp for testing. want to have 20 minute delays only in daylight as PIR dosn't work in daylight.
My plan is to use a photo transistor in the PIR section to disable the PIR durring daylight.circuit board needs to be redesigned.
PIR to be enabled at night. until then I just want to cycle the unit on every 20 minutes.
 
No, delayms won't work on long delays like that, exactly how long you can delay with delayms depends on your clock speed and the compiler you're using, you'll need to do a custom loop or setup a timer with several incremental registers.
 
still trying to figure out basic

what is %10111111 am assuming it points to certain regesters?
using tris = makes port an output?
trying to get a 20 minute loop created but keep getting errors.
using Junebug leds to test but will insert into my critter ridder code after getting a loop set up.
Code: 
Device = 18f1320
Clock = 8
Config OSC = INTIO2, WDT = OFF, LVP = OFF
OSCCON = $62
TRISA=%10111110         //A0 & 6 output
 dim ctr as byte
PORTA.0=1
ctr=1

While (true)
    Toggle(PORTA.0)
    ctr=ctr+1 // want to start a loop that 
    //counts for 20 miniutes
    
    if ctr <100 then false
   
    
Wend
End
 
where to find more info?

like what the numbers refer to
I started changing the numbers and see the results.
somehow I got on the Junebug led 1(A0/A6) to flash then led7(A0/A7) then led led 4 & 5(A7/A6 & A7/A0) came on simultaneously.
tried to figure how but ??
any clue about building a loop that will count for 20 miniutes.
looked at the timer code snippets but ??
I need to find a book on this basic swordfish or a book that will help guide me.
otherwise I am just taking code and fiddeling with it and hoping I stumble onto something.
 
The junebug LEDs are charlieplexed.

Three IO lines (RA0,RA6.RA7) are used to turn on six leds.

To turn on any given LED you make one of the three lines an input. With the remaining two lines you can turn on one LED by making one line HI and the other LOW. You can turn on a second LED by making the HI line LOW and the LOW line HI.

If you look at the schematic in the junebug assembly instructions you can see why. JPUG #1 covers charlieplexing on the junebug in great detail. Maybe too deep for where you are at. Don't know.
 
Status
Not open for further replies.

Similar threads

C
  • Question
Stepper motor help needed
Replies
4
Views
923
camerart
T
  • Solved
Can u help me with this counter!!! (PIC18F46K22)
Replies
8
Views
6K
Ian Rogers
A
  • Question
ASM: 24bit compare help please
Replies
4
Views
4K
augustinetez
A
H
  • Question
I need urgent help
Replies
12
Views
2K
Ian Rogers
H
  • Question
I need help again
Replies
16
Views
3K
Ian Rogers
Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…