Let's Make Robots!

9V battery keep blowing up!!!

Ok I have a problem that just does not make sense but since I am a newbie I thought I would throw this up and see if I can get some pointers.

I had and I say HAD two 9V Duracell batteries (Alkaline not NiMh) that I hooked up to my bread board.  In line I have either a SPDT or a DPDT switch (how do I tell, it has three leads but three positions one marked 1 the other marked 2 with the center position as the third).  Wierd thing is that as long as the switch is in the 'ON' position and my circuit is receiving current, the 9V battery is just fine but once I turn the switch to the 'OFF' position it is only a matter of time before the 9V blows.  I don't think it is a Duracell thing (I read some posts on the Net that said they were prone to spontaneous blowing up) because when I hooked up the second Duracell it also blew.  Again if the switch is in the on position everything is hunky dory but as soon as it is switched off it is like the battery is shorting somehow and getting really hot and then boom (sounds kind of like a firecraker).

I checked voltage coming back to the battery and got back a very minute amount of voltage (I thought ok its somehow getting charge back to the battery and trying to recharge an alkaline).  I am not sure if this is because of my voltmeter or if I can truly trust it.

One last idea, is it possible the switch is defective or is there a limit on the amount of voltage that can be passed through some of them.  Could current somehow jump the open circuit.

I have since taken the switch out of the circuit to protect my new NimH batteries.

Thank you.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Hello Shags,

 Sorry if this is sucking eggs for you.

Another thing to consider is that what are you trying to do with the 2 x 9Volt batteries?

If wired in Series (+ battery - + Battery -) gets you 18Volts, the alternative is parallel, in this case both positive connections are wired to one point and both negative are wired to another point, this gets you 9Volts but twice the amp hours.

If you are trying to drive the Opamp power Ok many Opamps have Pos and Neg power supplies, but with a +9/-9V supply your output signal could swing between + and -9 volts, this is too much for most Microprocessors. You only need to go 0 to +5V. 

Using a 18V power supply is also a bit hard on the regulator, for example a 7805 if you are using one of these they only like to burn off a couple of volts, you are just turning volts in to heat.

In your post above you indicate that you are using a PIR , which one the reason I ask is some of them are digital devices (see http://www.parallax.com/Portals/0/Downloads/docs/prod/audiovis/PIRSensor-V1.2.pdf)

Others are analog, these devices are normally two or more IR Photo detectors with a differential amp between them, thus if there is a difference in temp seen by the detectors you get a signal change. Normally these ones have three leads, one earth, one is the bias pin and last is the Signal pin, normally the bias pin has a suitable resistor (depends on the device, power supply etc..), often a capacitor is placed in parallel to shunt any high freq signals off, another bias resistor goes between the Signal pin and the power supply. You can amplify the signal here using an Opamp or a transistor or two. How are you reading the signal with the Micro, AtoD? Some micros have a number of pins which can be set up as comparators, you set the compare voltage and if the voltage on the pin (i.e. PIR Signal) is greater than the compare voltage then you have a 1 otherwise a 0.

All the best

 

Cliff

 

Just switch ONE WIRE on and off, period. Unless you are trying to reverse polarity, you will never have both positive and negitive going to a switch. Just switch ONE WIRE --I really does not matter which one is switched on and off, as long as the other wire goes straight to the load. ONE WIRE SWITCHED.

I agree with CTC, hooking both positive and negative to a switch, especially when your not certain about how the switch works is asking for trouble.

It pays to draw your circuit first to help get it right in your head before you wire it up. A quick sketch on a piece of paper can be a big help. If your not certain how to draw a schematic then google it. If you want to make robots you need to be able to read / draw simple schematics.

Ok I was wrong on the three positions (sorry about that). It definitely has two positions with three terminals.  I have since seen the error of my ways but I am man enough to post it here to the ridicule of all just so maybe it will help another electronics newbie.  Please correct me if I am wrong here.

I initially had the switch wired like this:  1. I had the positive hooked to the same terminal on the switch, lets call it the left terminal.  To correct this I really needed the positive from the battery hooked to the left terminal and have the positive going to my first component hooked to the center terminal to allow the positive to be connected only when the switch was in the left position.  This issue led me to my second mistake.  2. I had the negative hooked up to the switch as mentioned in two of the replies.  This appeared to work because as I had it hooked up if I moved the switch to the right (positive were both hooked to the center terminal the light would come on my led and if I moved it to the left it would go off.  Unfortunately when it was off the switch was just looping the current back to the battery.  When it was on it was ok because the current was eaten up by the components and the resistor.

Now the question for those more astute in electronics.  Would the circuit have been fine if I had hooked the negative to the left terminal on both the component and the battery side and staggered the the positive by hooking the positive from the battery to the left most terminal and the component positive to the center terminal.  in the position to the right there would not be a completed circuit, but to the left it would complete the circuit.  Because the net back to the battery is always 0 wouldn't this work correctly as well. (I am not saying it is correct but would it work, remember this is just for educational purposes only don't just say don't do it.)

Thanks all the terminal was correct but I also needed to stagger the positive or negative in order to make the switch work correctly which brings up one more question for you all, is it more correct to connect your switch to the negative side or the positive side?

Thanks again

The switch should go in series with the battery. so,

  • one wire from the battery to the middle connector on the switch (doesn't matter if it is the positive or the negative) 
  • one wire from the left or right connector (doesn't matter which one) of the switch to the robot
  • the other wire from the battery (the one that's not connected to the switch) to the bobot
And, that's it.

DO NOT SHORT OUT ANY KIND OF RECHARGABLE BATTERY!!!! YOU MUST SOLVE THIS PROBLEM BEFORE YOU USE THE RECHARGEABLES.

If you think your alkilines are getting hot, a NimH or a Nicad will get about 10 times that hot or worse, will dump out enough current to turn all your wires into toaster elements... They will get very hot, melt the insulation, continue to short even worse and burn your house to the ground. Seriously.

Oh yes I am being very careful with them.  For now I have removed the switch until I get it figured out.  I have a resistor in line to protect my components and bring the Voltage down.  My actual circuit uses a LED to show power is flowing and it then routed through a PIR sensor.  From there I am trying to use an OP-Amp chip I bought to amplify the signal to a point that a micro controller will recognize it.

Anyway, thank you very much for the info on how to verify the pins of the switch.  It has to be something like that thats happening.

There is no reasonable chance that their may be a defect in my breadboard is there?

Ok, what you have here is a simple short. Use your multi tester (measure-thingie) in it's continuity mode, and find 2 contacts that switch on and off when you flip the switch. These are the contacts you will use, lets call them "A" and "B". Now connect the negitive lead from your 9v directly to your test board. Next connect the positive lead from the battery snap to contact "A". Finally, connect contact "B" to your test board. That's it. One more, if you are using a robot brain on that breadboard of yours, remember -a lot of brains and sensors and the like work on 5V! Be sure you are not trying to shove too much power around.