Reverse Polarity Protection

cstan_02 · Joined: 24 Jul 2005 Posts: 13

Okay this is what my perception is ..
When we connect a diode on the ground for protection, there is a voltage drop on that terminal and the ground does not become the true ground.. Since this app is powered by a small battery all the time, it will drain the battery even when it is not powered on.. Thus putting the diode becomes less apt here..

gs · Joined: 22 Aug 2005 Posts: 30 Location: Ioannina - Greece

by putting a diode in series with a power supply terminal you will have a votage drop on the diode.

A suggestion for no voltage drop, for low current circuits, is a fuse with a reversed diode across the power terminals. My english is bad so I'll try to draw it

cstan_02 · Joined: 24 Jul 2005 Posts: 13

Hello gs, it is a good solution but we can't afford to let the customers change the fuse themselves.. it's a kind of hidden thing and it's supposed to be fail safe all the time, no matter what they do with it.. I just hope for a fuseless solution.. But there is limitation also.. Anyway thank you so much.

kender · Joined: 09 Aug 2004 Posts: 768 Location: Silicon Valley

I like gs' solution too. You can use a resettable fuse. It will short by itself sfter it has been blown. User doesn't wouldn't need to change it.

PCM programmer · Joined: 06 Sep 2003 Posts: 21708

I don't think it's a good solution, because fuses can take a while to blow.
During that time, you have reverse current flowing through your chips.

Suppose you have a 3AG type fuse, and it blows quickly enough to
save your circuits. The customer may easily assume that the "fast blow"
fuse is the problem and replace it with a "slow blow" type. Then when
they plug it in again, your board is damaged before the fuse can blow.

The resettable fuses can take a long time to trip. The data sheet
lists 200 or 300 ms, depending on the type.

The original poster never really adequately explained his circuit.
I don't know whether the antenna has some electronics in its box,
such as a pre-amplifier, or if it's just a metal box with nothing else in it.

cstan_02 · Joined: 24 Jul 2005 Posts: 13

Hello PCM, sorry if i have never made it clear.. The circuit is just too simple and that's why i omitted it.. The antennae are mounted on the box which originates from the board. These antennas are for GPS, GSM and RF(All OEM modules on the board). So all share the same ground through their connectors attatched to the casing(box)..

Thanks

kender · Joined: 09 Aug 2004 Posts: 768 Location: Silicon Valley

PCM programmer · Joined: 06 Sep 2003 Posts: 21708

I mostly agree.

I had overlooked the fact that the reverse voltage applied to the circuit
will be limited to the diode's forward drop. But there are still issues
with this approach:

The forward current due to the reverse voltage is going to be large,
because the only thing limiting it is the fuse. So the voltage drop
across the (silicon) diode will be -1.0 to -1.2v for a short time. This
will exceed the spec for Absolute Maximum Ratings on most silicon
devices, such as the voltage regulator on the board. If a fast blow
fuse is used, maybe nothing will be harmed, but if the fuse takes
a while to blow there could be some damage.

S1B Diode data sheet. Graph of Vf is on page 2.
http://www.diodes.com/datasheets/ds16003.pdf

LM317 voltage regulator data sheet. Absolute Maximum Ratings on page 4.
http://www.national.com/ds/LM/LM117.pdf

Storic · Joined: 03 Dec 2005 Posts: 182 Location: Australia SA

the link http://cache.national.com/ds/LM/LM2594.pdf is the regulator I use now, with my application I was able to save 50ma and low to no heat generated. it had replace the LM7805 Reg.

I had feed AC and DC via a Diode and Filter cap works well. I think it is important to have a diode in series to the REG to ensure if you reverse the polarity, you dont take out the circuit.

other reverse polarity circuits I have seen is the diode across the +Ve and -Ve and a fuse for protection. how about a full wave bridge rectifier, dose not matter which way the connection is, it will always be correct on the way out. Add a filter cap and you can have AC and DC in?

Andrew
_________________
What has been learnt if you make the same mistake? Wink

ckielstra · Joined: 18 Mar 2004 Posts: 3680 Location: The Netherlands

For circuits that can't afford the voltage drop over the forward diode there is an alternative setup using a FET: http://www.elecdesign.com/Articles/Index.cfm?AD=1&ArticleID=9945

libor · Joined: 14 Dec 2004 Posts: 288 Location: Hungary

I don't understand that in an automotive environment where the supply voltage can anyway vary from as low 7-8V (when starting) up to 14-15V, why can you not afford the small voltage drop of a series diode. You should design your power supply with that much tolerance anyway, so a diode's voltage drop should not affect it that much.
Schottky diodes can have a forward voltage drop as low as 0.3V btw., why not use a Schottky ?

cstan_02 · Joined: 24 Jul 2005 Posts: 13

The above mentioned app is always standby and if the mains is cutoff, it has to survive with a single nokia Li-ion battery for a few hours as well..
Easy to make a work around for this, but still it might drain the car/truck battery in long run.. Is there any MOSFET with a VGS of upto 24volts ??
The problem is not protecting positive input.. The main concern is protecting the ground since it is directly hooked to the chasis ground.. So in case of reverse polarity, it'll short circuit and burn the tracks.. Easy if i could put a fuse on the ground side, but..

Thanks

kender · Joined: 09 Aug 2004 Posts: 768 Location: Silicon Valley

PCM programmer · Joined: 06 Sep 2003 Posts: 21708

I think the following diagram describes his circuit. On the left side, the
board plugs into the cigarette lighter socket and connects to +12v and
ground. I've shown an inline diode on the +12v line. Normally, this diode
would protect against the cigarette lighter being wired in reverse.
But the problem is that he has a 2nd connection to ground on his board.

There's a cable associated with his antenna that has a ground wire
connected to his board at location "(X)". This ground wire then goes out
and connects to the metal chassis on the car somewhere. So now, if the
cigarette lighter is reverse-wired, and you've got +12v over on the GND
on the left side, current is going to flow right through the track on his
circuit board to the chassis ground, and the track will act as a fuse and
burn up. I think that's his problem.

libor · Joined: 14 Dec 2004 Posts: 288 Location: Hungary

Just another thought: What about a DC isolation on the antenna wires made with capacitors in series ? This would also be useful to protect your radio (gsm) module from any accidental (or the burglar's deliberately malicious) DC bias on the antenna wires.