Author Topic: Advise needed > lightning vs 7824  (Read 3327 times)

0 Members and 1 Guest are viewing this topic.

Offline Bloch

  • Supporter
  • ****
  • Posts: 449
  • Country: dk
Advise needed > lightning vs 7824
« on: November 10, 2010, 06:16:06 pm »
Hi All

This is my first post here so be kind to me  ;D

At work I have this circut "24V power supply" see drawing.

It have been built i many units with no problems.

It powers some wind sensors there is very likely to get stiket by a lightning.

My problem is that i would like to change it to a LM7824.

I need a current limiter and  need to make a new PCB and price / speed is always important.   

Pro.
  • Cheaper time and parts
  • Curent liniter
  • Smaller space

Con.
  • Will it survive the lightning as well as the old tried circut

And no i don't have the possibility to get it tested at a lab.


 

Offline tecman

  • Frequent Contributor
  • **
  • Posts: 435
  • Country: us
Re: Advise needed > lightning vs 7824
« Reply #1 on: November 10, 2010, 06:54:48 pm »
Your existing design has the merits of being robust and inexpensive.  The 7824 will give you tighter regulation and current limiting (sort of).  It is not true current limiting, but it limits based on thermal, so you will reach some steady state current on an overload.  The other caution is that the 7824 have a max Vin of 40 volts, which could be an issue with surges that might occur with lightning.  You could add parts to mitigate the surge sensitivity, but the cost will go up.

I would be tempted to stay with what you have and just ass a polyfuse (PTC fuse) to the DC line feeding the transistor.  Again thermal protection that will limit current.  Simple and robust.

paul
 
 

Offline Simon

  • Global Moderator
  • *****
  • Posts: 16652
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: Advise needed > lightning vs 7824
« Reply #2 on: November 10, 2010, 09:56:35 pm »
surely R1 needs to come from the positive supply not negative, like that the transistor will just not pass current
 

Offline oPossum

  • Super Contributor
  • ***
  • Posts: 1349
  • Country: us
  • Very dangerous - may attack at any time
Re: Advise needed > lightning vs 7824
« Reply #3 on: November 10, 2010, 11:52:09 pm »
Current limiting could be easily added to the existing design. A TVS (transient voltage suppressor) may improve robustness.


 

Offline Bloch

  • Supporter
  • ****
  • Posts: 449
  • Country: dk
Re: Advise needed > lightning vs 7824
« Reply #4 on: November 11, 2010, 06:38:43 am »
Quote from: Simon
surely R1 needs to come from the positive supply not negative, like that the transistor will just not pass current
Thanks. Yes i did miss i a wire.

Quote from: tecman
Your existing design has the merits of being robust and inexpensive.  The 7824 will give you tighter regulation and current limiting (sort of).  It is not true current limiting, but it limits based on thermal, so you will reach some steady state current on an overload.  The other caution is that the 7824 have a max Vin of 40 volts, which could be an issue with surges that might occur with lightning.  You could add parts to mitigate the surge sensitivity, but the cost will go up.

I would be tempted to stay with what you have and just ass a polyfuse (PTC fuse) to the DC line feeding the transistor.  Again thermal protection that will limit current.  Simple and robust.

Thanks. It is also my gut feeling that the 7824 is not robust enough.

I have tried the polyfuse solution.

The circuit powers 4 wind sensors. 

On installation they did an error (shorted) one sensor. Then they did see that they did it wrong. The polyfuse was so hot that the consumption from the 4 sensors did keep the polyfuse hot  >:( and newer got reset before they changed the PCB.

So the polyfuse is not a fix here.

Quote from: oPossum
Current limiting could be easily added to the existing design.

Thanks.

Why did i not think on that ? I am surely getting a little rusty.
I was thinking of adding a current limiter. But thinking in the direction of an extra power transistor+current sensor.

Thanks this must be  the answer to my "little" problem. Only 3 cheap/easy components more.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf