Email: Password: Remember Me | Create Account (Free)

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
12/08/11 06:11
Read: times


 
#185030 - Not Necessarily
Responding to: ???'s previous message
Kai Klaas said:
Nevertheless, you end up with a destroyed fuse, which is no good idea in an industrial application...


Crowbars do not need to be designed as units to blow out the front end fuse. The technique to blow out the fuse is pure and simply a cheap way to protect things with minimal circuitry. There are other ways that can protect the load, after a crowbar has tripped and shorted the output that are quite agreeable to industrial applications. There are two general types of these that I am aware of.

One type uses a scheme called "foldback" that causes a power supply to cut back on how much current the output can deliver into the short circuit load created by the crowbar. The current may be reduced by a factor of 10 or 20 times so that there is a dramatic reduction of power dissipation in both the crowbar and the power handling components in the power supply. Foldback schemes that I have seen are most often on linear type power supplies. The power supply stays in the foldback state as long as the crowbar short circuit load persists. Usually an input power cycle is needed to drop the delivered current to zero and release the crowbar.

Another type is a system that causes the power supply regulator to shut down completely when there is an overload on the output caused by a short circuit or a crowbar trip due to an over voltage event. The regulator shutting down is a latched event that will many times turn on an LED to signal the overload shutdown. When the regulator is shut down there is no power delivered to the load at all. If a crowbar had tripped the output the crowbar will automatically reset itself due to the cutoff of the output current from the power supply. The power supply itself will stay in the latched off state until there is a power cycle at the input to the power supply. These systems usually require the power supply to have an internal standby voltage rail that is used to power the shutdown latch. In my experience it is more common to see systems like this on switching power supplies although nothing prevents a designer from implementing such scheme for a linear power supply.

As you can surmise, both systems add some additional circuitry and thus raise the cost and complexity of the power supply.

Michael Karas


List of 40 messages in thread
TopicAuthorDate
Crowbar circuit using TL431            01/01/70 00:00      
   possibly worthless circuit            01/01/70 00:00      
      Speed not a problem            01/01/70 00:00      
         maybe it is            01/01/70 00:00      
      re: possibly worthless            01/01/70 00:00      
         no, the fuse blew            01/01/70 00:00      
            Polyfuse - OMG            01/01/70 00:00      
               And manufacturing tolerances can ruin the design...            01/01/70 00:00      
   Why not using...            01/01/70 00:00      
      Like shown here, e.g.,...            01/01/70 00:00      
         Possible but..            01/01/70 00:00      
            Tell us...            01/01/70 00:00      
   describe application, and needs            01/01/70 00:00      
      language            01/01/70 00:00      
         sorry            01/01/70 00:00      
            Awesome...            01/01/70 00:00      
         Antique...            01/01/70 00:00      
            The classic approach doesn't use a triac            01/01/70 00:00      
               did not think opf that.            01/01/70 00:00      
         Oh, come on, Erik ...            01/01/70 00:00      
            I do not critisize language problems...            01/01/70 00:00      
               It's a matter of personal pride ...            01/01/70 00:00      
      More info            01/01/70 00:00      
         and values of resistors?            01/01/70 00:00      
            Component values            01/01/70 00:00      
               oscillations?            01/01/70 00:00      
         one alternative circuit            01/01/70 00:00      
   a couple of questions ...            01/01/70 00:00      
   This one's straight from "often-wrong" Wikipedia            01/01/70 00:00      
   Thread wandering            01/01/70 00:00      
      then close it..            01/01/70 00:00      
         Problem now fixed            01/01/70 00:00      
            Are you sure??            01/01/70 00:00      
               Exactly the opposite            01/01/70 00:00      
                  Triac Triggering            01/01/70 00:00      
                     A1 to positive rail is ok            01/01/70 00:00      
                        Yes A1 to ++V            01/01/70 00:00      
                           Nevertheless...            01/01/70 00:00      
                              Not Necessarily            01/01/70 00:00      
                                 Standard for PC PSU            01/01/70 00:00      

Back to Subject List