Small AC Signal input protection

[luisjoserod]

Hi, I've been reading this https://www.ccsinfo.com/forum/viewtopic.php?p=127332 but still don't sure if it's ok to apply something like this to my purpose.

My ideal signal is a 5Vpp @ 1khz , then biased to 2.5V (with voltage divider and a voltage follower opamp) which is within the limits 0 - 5 V of the PIC.

So I'm trying to protect from signals above 5Vpp but still small signals, like 10Vpp that could take negative values even when biased and also got higher than 5V. What do you suggest?

I wonder if the idea could be done with some diodes, I'm trying to avoid the use of op-amps because my available supply for the circuit is 0V and +5V.

Thanks in advance.
_________________
Luis,
IEEE Member at University of Zulia
Electronics Engineering Student at URBE

[newguy]

Perhaps a TVS (transient voltage suppressor) will work for you?

https://www.digikey.ca/products/en/circuit-protection/tvs-diodes/144?k=&pkeyword=&sv=0&pv1730=i1&sf=0&FV=ffe00090%2C1c0002%2C1f140000&quantity=&ColumnSort=0&page=1&stock=1&pageSize=25

[dluu13]

Tvs diodes don't have very sharp corners on the iv curves so they will start conducting before you might expect and changing your input signal

I personally use Schottky diodes like the bat54 or bas70 up to the vdd rail to clamp to vdd+0.3v

Problem is this isn't always foolproof as it depends on your vdd rail having a giant capacitance to absorb the current flowing into it when your Schottky is conducting

[temtronic]

hmm, 'old school' trick is to use a real MC1489 'RS-232' receiver to convert the AC signal into DC. If you're old enough you'll remember, it was used with a PIC in a 'line frequency' measurement unit in Circuit Cellar Ink. It displayed 59.9 Hz on 3 digit LEDs. All machine code (CCS wasn't around back then..)
Well pretty sure it was CCInk... it was 2 decades ago more or less... PICs had windows on them back then and 15 minutes to erase.
Jay

[Ttelmah]

Er. If the only supply is 0 to 5v, and this is powering the op-amp,
then it's output can't swing outside this range, unless it's input is being
massively overdriven, and it's internal circuitry is becoming reverse
biased. You can protect the op-amp from this using diodes:
<//www.analog.com/en/analog-dialogue/articles/op-amp-input-overvoltage-protection.html#>
This does introduce increased load on the signal though.

[asmallri]

[Ttelmah]

Some extra comments:

If you clamp using a supply rail, remember the energy has to go
somewhere. If the rail is only powering low current devices, the energy
being delivered by the clamp diodes can raise the rail voltage. There
has to either be something to limit this like a zener, or a large enough
capacitance that the voltage increase is limited and can then be
discharged slowly by the load.

There are some ultra low leakage diodes available that can keep
the current and capacitance imposed on the signal to sensible levels.
Unfortunately Schottky diodes though they have low Vf figures, usually
have much higher leakage than a simple PN junction. So a diode like
a basic 1N914 can be a better clamp for low current signals.

For a complete clamp with low leakage, I've used the ESD9R3 very
successfully on a low current input.

As Asmallri says, If the input is high impedance, and to limit side
effects, series resistance can be your friend, however you clamp.
A resistor in the line limits the current that can flow.

[luisjoserod]

Hi, i've been reading you all and make this:

[img]https://imgur.com/2OPZ7MG[/img]

R10 and R11 are for offset,

Can you give me any feed back on my design?

Thanks
_________________
Luis,
IEEE Member at University of Zulia
Electronics Engineering Student at URBE

[asmallri]

[PCM programmer]

imgur.com is a legitimate image hosting website.