Analogue Pin over voltage protection

[Blob]

Hello,

I am working on a system that monitors the output voltage of a regulator on a MX bike.

Normally the regulator will give an output of 15V max.

So I use a voltage divider with 2 resistors to translate it to 0 to 5 V.

In some cases the regulator on the bike breaks, causing an overload on the divider input. and thus on my analogue input.

What is the standard or simplest way to protect your analogue pins from being blow?

thanks in advance,

Blob

[Ttelmah]

Provided the resistances in the voltage divider are reasonably large, the internal protection diodes are all you need.
If you use (say) 5KR, and 10KR, then you are comfortably below the recommended source impedance for most chips, and the existing protection _provided there is sufficient load on the 5v supply to 'sink' 20mA_, would happily handle over 200v...
If you allready have resistors in this sort of scale, and are having problems, think 'outside the box' slightly:
Are you sure your 5v supply will remain properly regulated?. What will handle HF spikes?. How much load is there on the 5v rail?. What causes the regulator to blow?. What is the nature of the generator output when the regulator blows? (in many cases, you effectively have a magnetic circuit, with energy stored in it, and nowhere for the energy to go - behaves then like a 'flyback' transformer, and voltages can go through the roof - they will rise, _till_ something dumps the energy, either distructively or safely....).

Best Wishes

[Blob]

Hello Ttelmah,

thanks, I now use 22K and 10K in series.
So that will be ok.

Our company is specialised in motor electronics, we see a lot of broken regulators, stators, flywheels,...

And I just want my new PCB to survive the "garage handyman's magical touch" (thats what causes the problem in most cases)

Thanks,

Blob

[SherpaDoug]

_provided there is sufficient load on the 5v supply to 'sink' 20mA_

This is an important part of Ttelmah's reply. Make sure there is something to prevent the 5V rail from rising like maybe a 5.3V zener or enough capacitance to eat the current for the length of the bad input pulse.
_________________
The search for better is endless. Instead simply find very good and get the job done.

[languer]

Keep in mind that the ADC acquisition time is proportional to the source impedance.

[bkamen]

[Blob]

thanks guys,

I use a decent voltage regulator to get my 5V supply. (LM2936MP)

I need this value only at low rpm, so the acquisition time is not that important.

best regards,
Blob