Circuit Protection Help please
Brian Carpenter
Posts: 728
I have a circuit that operates off 5v regulated wall power supply.· It is a unit that will be sold commercially.· my question is, how do i protect against someone plugging in a power supply greater than the 5v one supplied with the product.· There will mot be any internal regulators in my design, so i was wondering about·a fuse and a Zener diode.
See the attached pic.
Would i fuse it at 300 mah?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
It's Only A Stupid Question If You Have Not Googled It First!!
See the attached pic.
Would i fuse it at 300 mah?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
It's Only A Stupid Question If You Have Not Googled It First!!
bmp
385K
Comments
You'll want the fuse to clear (open) before you exceed the maximum Zener current.
-Phil
You could go with a 6 volt zener which would give you some leeway as to input voltage. Also, you also have to allow user access to change the fuse which raises the spectre of the user putting in a much larger fuse. You could use a self-resetting fuse which will clear and then reset after it cools down. Again, a cost issue.
So, at the end of the day, isn't an on-board regulator just as cheap?
Cheers,
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tom Sisk
http://www.siskconsult.com
·
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
It's Only A Stupid Question If You Have Not Googled It First!!
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
It's Only A Stupid Question If You Have Not Googled It First!!
-Phil
We use the Raychem Polyswitches at work all the time. They are readily available at Digikey in radial form and SMD.
The trick to these is to understand the given current rating is the current this device FOLDS BACK TO. That is, the RXE030 (rated 300mA) is a device that will pinch off, not cut off, starting when the current reaches 600mA. And then it only goes down to 300mA. There is still some small voltage left. Just to complicate matters, it is all a matter of the heat generated by the current passing through it. If the current rises near instantly to the max, it will pinch off in less than a second, but not as fast a fuse link. If the current climbs slowly above its hold rating, it will eventually fold back, even if not at maximum, but will take several minutes to reach the trip point, then take a second or two to drop to mininum current. And this puppy gets HOT! You'll have to kill the power and give it a few minutes to cool down before its ready to run normally again. Also, it has some resistance all the time, which drops some portion of the power voltage. Look these up on DigiKey and check out the charts.
The following online page has the RXE series data:
http://dkc3.digikey.com/PDF/T063/1499.pdf
Only one part looks like it is designed to fold back at 300mA. It is way at the page bottom. If you want to view a picture of the part you will have to change the web address from 1499.pdf to 1500.pdf.
So, these parts may be your answer, if you don't need fast, pinpoint current control, and can live with some small voltage drop.
Good luck on your project,
kenjj
It was pointed out to me that small value fuses [noparse][[/noparse]like 250ma] will actually cause a significant voltage drop that in itself can cause problems.
In truth, the real protection of a microcomputer circuit is in a good balanced design. About the only exception to this truth is some hardware failsafe features to prevent you from pluging in wrong polarity and from hot plugging interfaces that cannot tolerate such.
The area where the most damage is done is in assembly and on the breadboard. These places always allow you to get it wrong. I just had one of Gunther's Multiboards creating 13volts DC from 8 volts because inserted a diode in reverse and it appears I created a charge pump. [noparse][[/noparse]I bet that Gunther doesn't even know it is possible to do so.]
So it really comes down to being a habitual double checker and slowing down what you are doing in order to check. In the beginning I would just plug something in and watch it smoke. Now, I check the power rails for voltage level and polarity before I fully populate my boards. Also, I now know that anything over 220ohms will somewhat protect a BasicStamp or SX-28 or Propeller I/O. So when I see a value below that, I ask myself is this in the right place?
In sum, faultless design is a mental attitude of design and construction, not one special schematic. BTW, finding high current capacity zener diodes is not easy to do. While they may work in theory, the parts may not be available.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"If you want more fiber, eat the package.· Not enough?· Eat the manual."········