Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Need help - 5x7 Dot Matrix Display - Automotive application

Status
Not open for further replies.

rmuthuku

New Member
I have herewith attached a circuit diagram for reading an analog voltage from an automotive sensor and displaying relevant information on a 5x7 dotmatrix display. The circuit works and functions on the bench as designed. However, after connecting the circuit to the chassis of the motorcycle and having all the connections, and while starting the motorcycle, the PIC is reset and toasted. First I thought it could have been due to D10(BAT54S) and so I removed it and tried again. The result was the same. So far I burned out 3 PIC16F88. Can someone help me identify what's wrong with this HW design ? Thanks in advance for any help.
 

Attachments

  • Dotmatrix-Display.pdf
    51.3 KB · Views: 204
Your diagram is incredibly cluttered and difficult to read but .....

You are using a 7805 as your regulator to the PIC - at the minimum you'll want to try swapping that for an automotive rated regulator. I use a 7805 in one of my product designs but also have an RC filter, zener protection and a tranzorb (as well as a fuse) on the input and have never had one fail yet.

After that, I'm afraid I gave up trying to work out what pins you were using for the A2D etc.
 
I am using 780M05 as of now. RA1 and RA7 are used as analog inputs. J3 is shorted to 12V input.

If the PIC is powered with 5V, no pin can ever be tied to +12V; that will instantly smoke the PIC. Why is pin 10 ever tied to +12V???
 
The datasheet of 2803 recommends that pin10 to be connected to the supply voltage of any device it is supposed to drive. Also, I thought it would be useful for driving 12V relays. I tried pin 10 tied to 12V and open, both times it smoked the PIC.
 
When your PIC is being "smoked" - test the voltage between each pin and ground and reply back.

In fact, try it without the PIC in place in the circuit and just test the voltages that would be present on each pin (during idle, running and cranking)
 
Any CMOS circuit (like a PIC) specifically prohibits connecting ANY pin to a voltage > (Vdd+0.5V) or < (Vss-0.5V). If your PIC is running on 5V, that means any connection to any pin cannot go above 5.5V or below -0.5V!!! Read the data sheet
 
None of the PIC's pin is connected to any other voltages. Only 2803's pin is planned to be connected to 12V.
 
Any CMOS circuit (like a PIC) specifically prohibits connecting ANY pin to a voltage > (Vdd+0.5V) or < (Vss-0.5V). If your PIC is running on 5V, that means any connection to any pin cannot go above 5.5V or below -0.5V!!! Read the data sheet

Well technically ..... if you current limit the input you can get away with higher voltages on some of the input pins as they have clamping diodes. One microchip application has mains voltage applied directly to a pin via a very high value resistor (if my memory serves me rightly).

Realistically its not very good practice at all.

*This might be the one ...... http://dangerousprototypes.com/2012/02/19/app-note-lamp-dimmer-using-pic12c508/
 
None of the PIC's pin is connected to any other voltages. Only 2803's pin is planned to be connected to 12V.

Your schematic shows a path from +12v to PIC pin 10 via J3. You said J3 is connected! If that is correct, that is why you are blowing up PICs.
 
Well technically ..... if you current limit the input you can get away with higher voltages on some of the input pins as they have clamping diodes. ...

If you put a current limiting resistor between the PIC pin and the external voltage source (> Vdd), then the internal input protection diode will clamp the voltage at 0.65V>Vdd. Do it without the resistor, and likely you will destroy the chip.
 
Bikes can be really bad noise wise, radios even when powered by their own battery can have a hard time.
You could try powering the circuit from its own 12v supply, with just the gnd connected to the bike (all other inputs unplugged) and see if the circuit runs with the bike running.
You might need a low pass filter on the input to the 7805.
Some smaller engine bikes can have several volts of noise on them, that would uspet a pic but not neccesarily blow one up.
The supply to the 62783 is confusing, you've shown a transistor shorting out the lm317 with its base powring the 62783, its marked up link, is this a link or a tranny, if its a link is it fitted properly?
 
Last edited:
The circuit runs perfect when connected to the battery terminals of the bike in both ignition off and the engine running. No other inputs(neutral, clutch & the gear sensor) are connected. Yes, the supply to 62783 is confusing as I am using a transistor(symbol) for a 2 way jumper in the schematics. The idea behind this is to have an option of powering 62783 either thru reduced voltage or 12V for connecting any relays in the future. Also, I have a low pass filter thru L2(80uH) & C1. Is this not sufficient ?
 
You might have an error in the layout causing contention.

80uh allthough better than nothing may not be a high enough inductance, however you've allrerady proven that its not an incomming power issue, rather an i/o one.
 
I figured out the causes of PIC burnout. EMI coupling and noise on the +5V line caused the the PIC burnout. After disconnecting(or for that matter separating +5V wire from the connector bundle) stopped the PIC burn out. Thanks for all the suggestions.
 
Never heard anyhting like that before, noise has messed up pics running software before, nut not actually toasted the chip.
Maybe it was so bad the polarity was reversed or something.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top