| Author |
Message |
|
|
 Posted: Apr 16, 2010 - 02:44 PM |
|
Joined: Mar 09, 2006
Posts: 585
Location: Melbourne, Australia
|
|
I have a device which I would like to protect the power input off, since there will be someone who will inevitably plug in power the wrong way or try to connect the wrong power source.
The specs of the system are a 5-6v input, wich tolerates up to 7v input but no more. After this there is a 150mA 3.3v LDO but also a few level shifters
The level shifters are connected to the input supply and can only handle 7v max, therefore I need to ensure that the input never goes over this.
Currently I am thinking of using a 5.0v TVS Diode (such as SMAJ5.0ALFCT-ND). Since the TVS can only handle a finite amount of power for a very short time, adding in a PTC fuse (such as this).
My thinking is that the TVS will also protect against reverse polarity.
So am I on the right track? Any suggestions to making a simple (low part count) but solid protection system?
Cheers. |
|
|
| |
|
|
|
|
|
|
Posted: Apr 16, 2010 - 03:14 PM |
|
Joined: Aug 22, 2004
Posts: 1624
Location: Germany
|
|
I don't know if the circuit below still fits to your low components requirements. But it should produce no heat in the case of overvoltage. T1 simply disconnects your circuit if the input voltage is too high. T2 should switch off T1 if the power supply is approx. 6.5V.
Regards
Sebastian
EDIT: If you replace T1 by a FET, then you could spare D1. |
|
|
| |
|
|
|
|
|
|
Posted: Apr 16, 2010 - 05:54 PM |
|
Joined: Nov 22, 2002
Posts: 12054
Location: Tangent, OR, USA
|
|
Your thinking about a TVS and a PTC is just fine. There are a couple of caveats, however.
1) Make sure you use a "unipolar" TVS. There are some TVS devices that have the same limiting voltage in both directions; these are "bipolar". Unipolar are more like hefty zener diodes with a low forward voltage in one direction and the limiting voltage in the other.
2) Watch out for the TVS tolerance. They have two ratings. One is the max voltage that it WON'T limit. The other is is the max voltage that it WILL limit.
The SMAJ5.0A is guaranteed to NOT break down with voltages up to 5.0V (with a possible "leakage" current up to 0.8ma at this voltage). It is guaranteed to break down somewhere in the range of 6.4V to 7.0V with a current of at least 10mA. Thus, if you use a PTC, rated maybe at 200ma (there is a substantial tolerance to this rating, also), the LDO input voltage could get higher than 7.0V (maybe 7.2V).
I have used this scheme a number of times, and it is often a struggle to juggle the maximum device input voltage, the TVS breakdown, and the PTC current limit so that everything works. PTCs, particularly, are a challenge.
Jim |
_________________
Jim Wagner
Oregon Research Electronics, Consulting Div.
Tangent, OR, USA
"The only thing standing between us and victory is defeat" P.G.Wodhouse in Wooster & Jeeves series
|
| |
|
|
|
|
|
|
Posted: Apr 17, 2010 - 12:52 AM |
|
Joined: Jan 26, 2009
Posts: 534
Location: Thessaloniki Greece
|
|
| What about using a bridge rectifier for protection from reverse polarity input? |
|
|
| |
|
|
|
|
|
|
Posted: Apr 17, 2010 - 12:57 AM |
|
Joined: Nov 22, 2002
Posts: 12054
Location: Tangent, OR, USA
|
|
Then you have two diode drops.
There may also be some problems if you have any possibility of your device ground being connected to "Line Neutral" and the power negative is also connected to Line Neutral or Earth. That will, at a minimum, remove part of the bridge. (Though I don't know how Australian power service is arranged).
Jim |
_________________
Jim Wagner
Oregon Research Electronics, Consulting Div.
Tangent, OR, USA
"The only thing standing between us and victory is defeat" P.G.Wodhouse in Wooster & Jeeves series
|
| |
|
|
|
|
|
|
Posted: Apr 17, 2010 - 03:42 AM |
|
Joined: Mar 09, 2006
Posts: 585
Location: Melbourne, Australia
|
|
Sebastian, that is a very elegant solution, I like the idea of no heat. I'm not sure I can fit that many components onto the PCB although it could be a last resort. 
Jim, your second point is exactly what i'm seeing. I'm using a SMAJ7.0A with a 500mA PTC (what I had on hand). If I apply 12v the voltage is brought down to around 9.5v, but all that extra energy is turned into heat, pretty much de-soldering the TVS, but not blowing it.
I guessing that the PTC is letting through too much current in this setup.
I was under the impression that in the overvoltage scenario, the TVS would go SC and then the PTC would become OC (or close to it) and the system would simply just not be powered. This isn't what I am seeing though, is this thinking wrong?
The LDO input is not a problem, it can handle up to 16v. The problem is the level shifter which is connected to the external (or a protected external) rail. It would be nice if I could just clamp it to 5v. Putting in a small 5v LDO may be an option, but a bit overkill. |
|
|
| |
|
|
|
|
|
|
Posted: Apr 17, 2010 - 04:49 AM |
|
Joined: Jun 22, 2004
Posts: 3849
Location: South West Utah, USA
|
|
|
|
|
|
|
|
Posted: Apr 19, 2010 - 07:11 AM |
|
Joined: Aug 04, 2002
Posts: 1706
Location: Brisbane, Australia
|
|
Jim,
Australian urban power supply is arrnged as symetrical 3 phase system distributed at 415V line votage at 50Hz.
Phase voltage ( line to neutral is 240 V accordingly.
Each consumer typically gets connected to neutral and one, two or all three lines.
Additionally each consumer has own local earth stake which is connected to Neutral. Probably to stop the Neutral floating ( rotating) locally.
Rural powersupplies often come to the consumer via a single phase often unregulated ans sometimes through either a scot or Le Blanc 3 phase to single phase converter transformer. This ensures a balanced load at the 3 phase supply point.
The single phase sipply is ofyten supplied via a single wire with the soil forming both local earth and return path. |
|
|
| |
|
|
|
|
|
|
Posted: Apr 19, 2010 - 07:49 AM |
|
Joined: Mar 09, 2006
Posts: 585
Location: Melbourne, Australia
|
|
Interesting solution Mike, like you say though, too complex for this purpose.
I should clarify that this is a battery powered device. It's not intended to ever be connected directly to mains power.
I have been thinking about putting in a 5v zener for the supply to the level shifter, since this is really the only part I want to protect from greater than 7v. Any thoughts on this approach? |
|
|
| |
|
|
|
|
|
|
Posted: Apr 19, 2010 - 08:23 AM |
|
Joined: May 04, 2007
Posts: 3529
Location: Geelong Australia, Home of the "Cats"
|
|
What about an in-line fuse of correct current say 100mA & a 6.8V />1 Watt zener diode.
On excessive voltage, fuse blows when zener conducts(in zener mode mode)
On reverse voltage, fuse blows through forward conducting diode.
No current consumed by circuit in normal mode. |
_________________
Charles Darwin, Lord Kelvin & Murphy are always lurking about!
Lee -.-
(If you haven't already done so, edit your PostNuke profile and let let us know where you are, what you do & what your interests are.)
|
| |
|
|
|
|
|
|
Posted: Apr 20, 2010 - 04:03 AM |
|
Joined: Mar 09, 2006
Posts: 585
Location: Melbourne, Australia
|
|
That sounds more like what I want Lee, thanks!
I was under the impression that this was basically what the TVS diode does, but it seems it's not as clear cut as that.
I will grab a 6.8v zener and a 160mA PTC and test this out later today. |
|
|
| |
|
|
|
|
|
|
Posted: Apr 20, 2010 - 04:53 AM |
|
Joined: May 04, 2007
Posts: 3529
Location: Geelong Australia, Home of the "Cats"
|
|
The circuit will work with a fast fuse that goes open circuit.
With a PTC fuse , which is a lot slower it may not give the same protection. Let us know how it goes. |
_________________
Charles Darwin, Lord Kelvin & Murphy are always lurking about!
Lee -.-
(If you haven't already done so, edit your PostNuke profile and let let us know where you are, what you do & what your interests are.)
|
| |
|
|
|
|
|
FAQ
Search
Register
Log in to check your private messages
Log in
Maximize
RSS
