Voltage problems

shrtcirkt

Well-known member
Joined
Dec 22, 2001
Messages
328
Reaction score
0
Ok, here is what I think has been a tough problem, but may be fairly simple for someone out there.
I have several PIR boards using the master chip that I mentioned in my other post today. I have joed's timer, an optocoupler forthe cam, a voltage regulator supplying the chip and a voltage inverter coming off the PIR because it;s trigger is an active high.
Tinhorn may remember this setup from a few months ago.
What the problem is when the system is in camera mode it goes into a continous loop. Everytime the micro chip finishes it's delay count and resets it causes a trigger in the PIR that starts the cycle over again. If I use two seperate power sources the system works great, but when I switch back to one power source supplying both the timer chip and the PIRI get the continous loop again.
I added a 47k resistor and a cap to the + side ofthe PR, but the voltage drop across the resistor was too great and the PIR would not turn on at all.
could a voltage regulator be used on the PIR as well to smooth out any spikes?
I have a copy of my schematic available if anyone wants to see it, or I might be able to post it if necessary.
Any thoughts anyone?:confused-yellow::what-do-u-think:
 



Archilochus

Well-known member
Joined
Mar 27, 2001
Messages
1,285
Reaction score
0
Hi shrtcirkt,
Do 'ya have access to a 'scope to find out where the problem is happening??  Need to find out if it the PIR is getting tripped or if it's the micro itself getting reset - that would make tracking down the problem easier.

Archilochus
 

shrtcirkt

Well-known member
Joined
Dec 22, 2001
Messages
328
Reaction score
0
Arch, during my troubleshooting I had connected an LED to the output of the PIR. And the LED was coming on as the camera circuit re triggered.
Unfortunately I dont have access to a scope any more, But I could probably bring the board to a tv repair shop if necessary.
What do you think?
Thanks
 

Archilochus

Well-known member
Joined
Mar 27, 2001
Messages
1,285
Reaction score
0
Hmmmm.... Well... I'll take a look over the data sheet for the PIR control chip - but a diagram of how you have everything wired up would be heplful.

Archilochus
 

shrtcirkt

Well-known member
Joined
Dec 22, 2001
Messages
328
Reaction score
0
here is the schematic, sorry bout the quality, I did this up real quick this morning.

 

shrtcirkt

Well-known member
Joined
Dec 22, 2001
Messages
328
Reaction score
0
I wired in a 5volt regulator to the input of the PIR board and that seems to have solved the problem. So far with the test unit it has been working good.
$.25 fix, but nowI need to re-etch boards. hope it holds up.
 

Brian

Inactive
Joined
Jul 30, 2001
Messages
1,194
Reaction score
0
Which PIR are you using?

The oscillation is most definitely the PIR because the PIC will not work if the voltage to the chip is in what is called a brown out situation.  You can test the voltage at the trigger and see if it is changing from high to low.

According to the comparator specs I have looked at, the load capacitance should be around .01uF.  If you exceed this then the comparator will oscillate.  I used this concept to improve the TS part numbers Arch was having trouble with in his circuit.  It cured the problems for my boards using the TS part numbers.  Try putting a .01uF cap on the PIR trigger to ground right before the base resistor for the inverter.  This may not work since I don't know what PIR you are using but it can't hurt to try and it is easy to add to a completed board.

Arch did you get a chance to try the stuff I sent you.  Email me if you did.

(Edited by Brian at 10:37 pm on Mar. 14, 2002)
 

Tinhorn

Well-known member
Joined
Mar 13, 2001
Messages
3,516
Reaction score
0
when I build a Refresh Circuit out of a TLC555, I sometimes get a "Power Line" spike from it that triggers the PIR.  

I've found that if I add a diode in the (+) power lead that is going to the Refresh Circuit (with the Kathode towards the Refresh Circuit) it blocks any spikes created by the refresh circuit from getting into the power supply and ending up at the PIR.

You can see the diode (D3) in this schematic:

http://www.nightowl.net/~tinhorn/JoeDRefresh.jpg

Only trouble is, I think shrtcirkt and I already discussed this method months ago and it didn't fix his problem ! ! !

But since the 5v reg you added works, It still leads me to believe that a spike is coming from the timing chip, the regulator is just chopping off the spike (as you figured it would do)

A few times the "TEST" LED created a spike that triggered the PIR (believe it or not) and I had to place a diode in series with it to prevent it ? ? ?

I never did drag my scope out to look for the spike because I figured it was so fast I'd not see it, too bad it aint a recording scope.  

Tinhorn
 

Archilochus

Well-known member
Joined
Mar 27, 2001
Messages
1,285
Reaction score
0
Hi Tinhorn,
You should be able to see even fast spikes with your 'scope if you set the triggering mode and source to the appropriate settings and adjust the the level and slope a bit.

>>>>
shrtcirkt......
Sounds like you have it worked out - but I'll just toss in a few more bits of info for future use.  Might save you a few ICs, caps and some board space.

I'm assuming you're using a micro-power, low-drop CMOS voltage regulator, right??
If so, these buggers can sometimes be a bit touchy about what types of in/out capacitors are used - particularly on the output side (problems increase at temp extremes). I noticed in your schematic that you have shown a single electrolytic cap on the output side of the reg.  The data sheet for your regulator should give some info about what characteristics the in/out caps should have - one of the more important being ESR - Equivalent (Effective) Series Resistance.  
Whenever one of my circuits needs a low-drop regulator, I use "low ESR" aluminum electrolytics (from Mouser) on the output *and* input in parallel with a small value ceramic caps for higher frequency noise.  I've not had any problems with switching noise on the regulated power rails - even when switching over 60 mA.

Archilochus

(Edited by Archilochus at 9:00 am on Mar. 15, 2002)
 


Top Bottom