Searching \ for '[PIC] Pin Protection' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/microchip/devices.htm?key=pic
Search entire site for: 'Pin Protection'.

Exact match. Not showing close matches.
PICList Thread
'[PIC] Pin Protection'
2005\05\24@045852 by Francois Robbertze

flavicon
face
I still have a problem with my protection to the circuit.

The LM317 give me a constant approx 1mA current. When the wire gets cut the
output of the LM317 should go up to 12V. (Thanks Herman)

I have upload a schematic at http://www.eskulaap.co.za/schem.gif .
Will this circuit give me relative good protection against ligtning.

The fence is a 1.6mm wire around a residential home and probably a very good
candidate for a lighting strike.

Should I connect more transorps, MOV's, caps or inductors. Where and what
values?

Thanks

Francois



2005\05\28@042911 by Herman Aalderink

picon face
Francois Robbertze wrote:

>I still have a problem with my protection to the circuit.
>
>The LM317 give me a constant approx 1mA current. When the wire gets cut the
>output of the LM317 should go up to 12V. (Thanks Herman)
>
>I have upload a schematic at http://www.eskulaap.co.za/schem.gif .
>Will this circuit give me relative good protection against ligtning.
>
>The fence is a 1.6mm wire around a residential home and probably a very good
>candidate for a lighting strike.
>
>Should I connect more transorps, MOV's, caps or inductors. Where and what
>values?
>
>Thanks
>
>Francois
>  
>

The problems and solutions  .......

aa. You cannot protect your PIC system (any electronic system)  from a
direct hit unless you provide for a professionally-installed
lightning-protection system.
As direct hits are rare it is cheaper to instead make 2 the same systems
(have a spare). Install the spare after a direct hit.
Hint: Include a weak link that will get destroyed first (can be as
simple as one or more 1/8 Watt low-Ohm resistors in series), thereby
limiting the damage to the electronic system.

bb. Induced lightning-induced spikes (many thousands of volts at worst,
limited by the actual resistances to ground along the loop). Count on
1000's of volts for your design.

To get rid of kiloVolts you have to combine these 2 ways of protection :
1. Block the spike from getting to your PIC/electronics. (Inductors in
series.)
2. Lead the spike to ground. Capacitors and MOVs to ground.

1. Block the spike from getting to your electronics. (Inductors in series.)
You did not show any inductors to stop the kiloVolts from reaching your
PIC. Still some work to do there. You can make your own inductors by
winding a coil over a low-resistance (preferably carbon) resistors. Use
several in series.

2. Lead the spike to ground. Capacitors and MOVs to ground.
You show a 17Volt MOV leading the kiloVolts to ground. Good !
You can also use capacitors, same type as used for by-passing, discussed
often on the PIClist.

Combine 1. and 2.
Put an inductor in series followed by a capacitor or MOV to ground. I
did so 3 times (3 inductors in series, followed by bypass-caps to
circuit-ground) and lost a bipolar transistor (I did not have a good
ground). My wires among the trees outside were maybe 200 meters long,
twinlead.

Back to your circuit:

Your MOV conducts the charge to ground (but not till after the spike has
reached 17 Volts).
So your electronics will see 17 Volts at best.
Lightning has very-high-frequency characteristics. It will jump all over
the place because of serial capacitances, incl capacitance between
wiring of a inductor. It will get to the electronics/PIC unless you stop it.
Often via the powersupply.
By the way .... if you have a powersupply or charger connected :
    Expect the same lightning-spike to come via the 'backdoor' ......
via the (overhead?) powerlines.

You can guess the results.
Note: 17 Volt assumes a perfect MOV. Actual voltage is higher. Assume
MUCH higher.

You did not provide protecion for your MOV.
As the MOV conducts the charges it is likely to get damaged sooner or later.
An MOV will :
    fail OPEN (burned out, maybe literally). > you have no more protection.
    fail shorted (dead short, but more likely with considerable
resistance left > you think all is well.
>>>>>>> some more work to do Francois

Professionals put a low-wattage, low resistance resistor in series with
the MOV. Like when MOV is used for over-voltage protection in
powersuppies. To avoid burning-up of the MOV (I have seen MOVs on fire,
during tests).

To easily check if your protection is still working, you probably want
to add an easy way of testing it. Indicator or a test-pushbutton.

Good luck ! (and dont get too close while lightning is around !)

Herman in PHL.

2005\05\28@110143 by Spehro Pefhany

picon face
At 10:58 AM 5/24/2005 +0200, you wrote:
>I still have a problem with my protection to the circuit.
>
>The LM317 give me a constant approx 1mA current. When the wire gets cut the
>output of the LM317 should go up to 12V. (Thanks Herman)

The LM317 can't give you 1mA constant current- it needs about 5mA to work.
It can also be destroyed by reverse current unless you put diodes around
it, which allows current to flow back into the 12V supply...  so I suggest
another method.

See this quick sketch:

http://www.speff.com/Fence_break.pdf

The physically large 10W wire-wound resistors can typically withstand
short transients in the thousands of volts. The lightning choke and spark
gap (as for electric fence controllers) deal with the higher voltages
and currents. This should be pretty bulletproof with good layout and
proper grounding, although nothing is for sure with lightning). The key
is to provide a more attractive path for the current than through the
sensitive circuitry, and to deal with the high, but manageable, voltages
and currents that result.

Hope this helps,
Best regards,

Spehro Pefhany --"it's the network..."            "The Journey is the reward"
spam_OUTspeffTakeThisOuTspaminterlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com
->> Inexpensive test equipment & parts http://search.ebay.com/_W0QQsassZspeff


2005\05\28@112528 by Russell McMahon

face
flavicon
face
>>I still have a problem with my protection to the circuit.

Completely opto isolating the input circuitry from the processing
curcuitry with a coupler (possibly home made) rated at 10's of
kilovolts will help greatly and not cost much relative to everything
else involved.

Where the cable leaves the fence in an insulated pair (or more) of
wires, run it down to ground level and wind the cable in a two or
three turn loop of around 150mm diameter with the bottom of the coil
touching the ground or even partially buried in it. A full lightning
strike will burn through to ground at this point. Mathematical
calculations show that a coil like this is not an effective bar to
lightning BUT practical experience shows that it works. You still have
to deal with a very nasty spike on the line beyond this point but it
"takes the edge off it".  I don;t think anyone mentioned gaseous
arrestors (gas discharge tubes) which are specially made for lightning
protection and are arguably THE most effective means available of
limiting lightning potential rise. Google will know about them.




       RM

2005\05\30@214853 by Chen Xiao Fan

face
flavicon
face
Oh yes. Gas arrestors are one of the best device for lighten
protection. The company I am working for have lots of
Over-voltage suppressor (lightning-protection barrier).

Take a look at my company's web site.
http://www.pepperl-fuchs.com and search for P-LB and K-LB.
The discharge current is up to 10KA.

Xiaofan

{Original Message removed}

2005\05\31@053859 by Francois Robbertze

flavicon
face
Dear Spehro Pefhany,

Thanks, I have looked at your circuit. Just want to clear up some confusion.

According to the schematic there are a ground(0V track on the PCB) and a
earth(close to the spark gap).

The transistor, 10K Resister, 10uF cap and TVS are all connected to the
ground.

There must also be a short(thick) track from the TVS to the pad that will
connect the ground to earth.

Do I understand this correctly?

So when the loop is fine there will be a approx 1mA current and the PIC
input will be low. When the loop gets broken there will be approx 0.23mA
current. The Transistor base will see approx 2.3V and the PIC input will
read High

Francois


{Original Message removed}

2005\05\31@083004 by Spehro Pefhany

picon face
At 11:29 AM 5/31/2005 +0200, you wrote:
>Dear Spehro Pefhany,
>
>Thanks, I have looked at your circuit. Just want to clear up some confusion.
>
>According to the schematic there are a ground(0V track on the PCB) and a
>earth(close to the spark gap).

Yes, they are connected together, close to the where the fence loop return
connects.

>The transistor, 10K Resister, 10uF cap and TVS are all connected to the
>ground.

Yes.

>There must also be a short(thick) track from the TVS to the pad that will
>connect the ground to earth
>Do I understand this correctly?

Correct.


>So when the loop is fine there will be a approx 1mA current and the PIC
>input will be low.

High.

>When the loop gets broken there will be approx 0.23mA
>current. The Transistor base will see approx 2.3V

0.6V, because it's a BJT with ~250uA flowing into the base

>  and the PIC input will
>read High

Low.

Best regards,

Spehro Pefhany --"it's the network..."            "The Journey is the reward"
.....speffKILLspamspam@spam@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com
->> Inexpensive test equipment & parts http://search.ebay.com/_W0QQsassZspeff


2005\05\31@123949 by Peter

picon face


On Tue, 31 May 2005, Chen Xiao Fan wrote:

> Oh yes. Gas arrestors are one of the best device for lighten
> protection. The company I am working for have lots of
> Over-voltage suppressor (lightning-protection barrier).
>
> Take a look at my company's web site.
> http://www.pepperl-fuchs.com and search for P-LB and K-LB.
> The discharge current is up to 10KA.

And the minimum firing voltage is 80V so you need something else after
the gas arrestors to stop the damage.

Peter

2005\05\31@195645 by Chen Xiao Fan

face
flavicon
face
Oh yes. You will have your usual protection circuits
in your device (TVS, zener, fuse, resistor, varistor, etc).
That is the normal thing you need to take note anyway.
You need to design according to the standards which apply.

The lightening barriers add protection to the voltage
differential caused by lightening strike. Actually
it will not even protect the connected device from
direct lightening strike.

Xiaofan

{Original Message removed}


'[PIC] Pin Protection'
2005\06\01@085256 by Howard Winter
face
flavicon
picon face
On Tue, 31 May 2005 19:39:47 +0300 (IDT), Peter wrote:

> On Tue, 31 May 2005, Chen Xiao Fan wrote:
>
> > Oh yes. Gas arrestors are one of the best device for lighten
> > protection. The company I am working for have lots of
> > Over-voltage suppressor (lightning-protection barrier).
> >
> > Take a look at my company's web site.
> > http://www.pepperl-fuchs.com and search for P-LB and K-LB.
> > The discharge current is up to 10KA.
>
> And the minimum firing voltage is 80V so you need something else after
> the gas arrestors to stop the damage.

Yes, that's where you'd have something like a MOV - you need protection in depth for this sort of thing.  

The gas-discharge tubes do work - the Telcos have been using them for decades, at least.  I can remember
hearing the ones in my telephone master socket give off a loud BZZZZT during a close-by thunderstorm - nothing
got struck, it was just the induced voltage in the line.  A MOV on its own would be likely to be blown apart
by the energy involved, whereas a gas discharge tube takes it in its stride.  

Of course we're not talking about direct lightning strikes here - in that case whatever you do in the
electronic domain, you're still likely to end up with a box-full of carbon and deposited metal vapour!  :-)

Cheers,


Howard Winter
St.Albans, England


2005\06\01@085715 by Gerhard Fiedler

picon face
Chen Xiao Fan wrote:

>>> Gas arrestors are one of the best device for lighten protection.

>> And the minimum firing voltage is 80V so you need something else after
>> the gas arrestors to stop the damage.

> Oh yes. You will have your usual protection circuits in your device (TVS,
> zener, fuse, resistor, varistor, etc). That is the normal thing you need
> to take note anyway.

I think anything in the range of lightning protection has to be
multi-staged. You usually go for more robust and/or slower devices on the
outside (to pick up the high peaks of voltage and the bulk of the energy)
to more precise and/or faster devices on the inside (to get the fast peaks
and make sure the specs of the connected circuit are not exceeded).

Gerhard

More... (looser matching)
- Last day of these posts
- In 2005 , 2006 only
- Today
- New search...