Voltage Divider Blows Fuse

Hi, I have a simple circuit that monitors 3no 12V signals from a burglar alarm that are constantly running HIGH.

When they fall to LOW they trigger an input pin on the Arduino and send a signal.

I have a simple voltage divider 10K (R1) and (3.3K) R2 that provides 3.3V to each input pin, with the remaining 9V connected to ground.

The Arduino is powered at 5V from the 12V circuits using a switching power supply.

The problem is that although the system runs fine initially, eventually the fuse in the alarm blows which I suspect is due to the return voltage/current running through Ground.

What am I doing wrong?

Would I be better with an optocoupler and would say the PC817 accept 12-13v on it's input and would I avoid the grounding issues?

Not giving us a wiring diagram is where you’ve gone wrong .

Also the output specification for the alarm - is it a relay contact or a transistor ? What does it say in the manual about it .

Hi,
Welcome to the forum.

Please read the first post in any forum entitled how to use this forum.
http://forum.arduino.cc/index.php/topic,148850.0.html .

Can you please post a copy of your circuit, in CAD or a picture of a hand drawn circuit in jpg, png?

Can you tell us your electronics, programming, arduino, hardware experience?

I have a simple voltage divider 10K (R1) and (3.3K) R2 that provides 3.3V to each input pin, with the remaining 9V connected to ground.

The underlined has me worried.

Thanks.. Tom... :slight_smile:

The circuit diagram is attached, there are 3no identical voltage dividers connected to each respective pin (3.3V) with a common ground.

Correction, 9V does not go to ground but I have measured the current drain to ground at less than 1mA.

The alarm outputs toggle between 13.3V (normal) and 0V (alarm state), therefore, the code detects the absence of a input to the PIN.

Where are the capacitors, required on the input and output pins of the LM1117?

How is GND in your circuit connected to GND on the Alarm?

The LMS1117 is a module, containing all the capacitors etc.

The ground is missing from my sketch ( :kissing: ) but connects back to the - supply (GND) of the alarm from the GND of the LMS1117 module.

I think this may be my issue?

The voltage divider(s) won't blow a fuse, if the resistor values are as you claim. A bad connection, or the power and ground connections to the LM1117 module and the Arduino are suspects.

This statement does not agree with your schematic.

The Arduino is powered at 5V from the 12V circuits using a switching power supply.

The schematic suggests that you are applying 5V from a linear regulator (the LM1117) to the 3.3V Arduino output.

Check all voltages, resistor values and connection continuity with your multimeter. Verify that the Arduino still works if powered separately (e.g. program the Blink example).

FYI - The voltage divider CAN'T blow the fuse because the current through the voltage divider doesn't go through the fuse... And ~1mA is what you'd expect (12V/13.3K) so if the values are correct and it's wired correctly it won't draw enough current to blow the fuse. And, I assume you are also measuring about 3.3V.

All of the grounds should be connected together. But, missing grounds would not cause a blown fuse. Of course a wrong ground connection... A short to ground... can blow a fuse or cause other problems.

Too much current would indicate a bad or mis-wired voltage regulator module or a mis-wire to the Arduino, or the Arduino is simply bad (and shorted out).

Since you apparently have a multimeter, disconnect the Arduino and if you have time before the fuse blows, check the voltage into and out-of the LM1117 voltage regulator. If it's OK, check the current into the voltage regulator without the Arduino connected. If you are getting excess current without the Arduino, the voltage regulator is dead (shorted) or mis-wired.

If you are getting excess current through the Arduino it's most-likely dead, although it might be mis-wired in a way that draws excess current without getting destroyed.

Usually when a fuse blows, something else has already blown. There's an old joke, "The transistor blew to protect the fuse". the purpose of a fuse is usually to prevent a fire, or to prevent further damage after something has gone wrong. The fuse in your circuit should protect the power supply but it may not protect the Arduino.

Your schematic doesn't show any connected Arduino outputs. A microcontroller can be useful with no inputs, but it's kind-of useless without outputs. I assume there are outputs (unless maybe just the USB serial monitor or watching a built-in LED?) and maybe the problem is with something you didn't show us.


You didn't say which Arduino you're using, but the standard Uno has a built-in voltage regulator so you can power it directly off of 12V as long as you are ONLY powering the Arduino (and maybe a few LEDs, etc.) through that built-in regulator.

Logic-high is "nominally" +5V. 3.3V will work but it's "cutting it close" for a logic-high. Anything greater than 3V (0.6Vcc) is read as logic high and anything less than 1.5V (0.3Vcc) is read as low. Between 1.5 and 3V is undefined and it may read high or low.

The "absolute maximum" to an input is 5.5V (Vcc+0.5V), but there are (low current) over-voltage protection diodes on the I/O pins so with the 10K voltage divider resistance limiting current, you can actually go over a "calculated" 5.5V, and the protection diodes will safely limit the voltage. So IF there is current limiting, you don't have to be too paranoid about accidently putting more than 5V on an input pin.

ushills:
The problem is that although the system runs fine initially, eventually the fuse in the alarm blows.

Probably a good idea to elaborate on 'runs fine initially'.

Eg. what does runs fine initially mean? For 1 minute? 10 seconds? 20 hours?

Which Arduino?

I really need to get better at drawing schematics.

Also, missing from my sketch is the switched power converter taking the 13.3V down to 5V to supply the power regulator.

The board is a NodeMCU (ESP8266) module that for an output triggers a webhook to prompt a twitter notification, other than that there are no outputs electronically.

I have rewired the board with the ground from the voltage dividers going straight back to the alarm ground, the LMS1117 ground & board ground now connect back to the voltage regulator ground and therefore all grounds are in fact connected. Following this, the board has been stable for 12 hours.

Previously the voltage dividers grounds were connected, as in the sketch, back to the ground of the LMS1117, but I could see no reason why this would cause a fuse to blow in the alarm supply to the board.

I am using the LMS1117 rather than the regulator on the board as initially the LMS117 was connected directly to the alarm, without the switched supply, but generated too much heat dropping to 3.3V.

ushills:
Previously the voltage dividers grounds were connected, as in the sketch, back to the ground of the LMS1117, but I could see no reason why this would cause a fuse to blow in the alarm supply to the board.

I am using the LMS1117 rather than the regulator on the board as initially the LMS1117 was connected directly to the alarm, without the switched supply, but generated too much heat dropping to 3.3V.

The thing here is ----- ‘previously’. When things are changed and different from what was previously presented, then it becomes difficult for people here to track or trace or keep an account of what’s going on.

So - after this change or modification - is the fuse still blowing?

Also - if the LMS1117 is over-heating, then does it require cooling/heat-sink etc?

Hi,
Can you post a picture of your project so we can see your component layout please?

Thanks..Tom.. :slight_smile: