I actually do a very fast, software defined, overcurrent shutdown in the product of interest here. It's been a while since I worked on that code but it will shut down a short circuit in a few ms or so.
The software defined part allows some flexibility in the shutdown which is important in that I also have to not shut down for inrush currents. The algorithm is such that if the overcurrent is small it shuts it down slower than a, large one. Roughly, it trips once a certain amount of excessive energy is seen. I'm guessing in this case the amount of energy we let through is more than the windings on the magnetics can handle. If I have time I'll grab a set of magnetics and see if I can characterize this. I'm also surprised that the 100 series radios died as well as they should appear as a dead short to the injector and have no magnetics on those pins to blow up. On Apr 24, 2015 8:58 AM, "Chuck McCown" <[email protected]> wrote: > Forrest, please forgive me for even thinking this thought... but I > wonder how much current it takes to blow a phy transformer and how hard it > would be to have an over current shutdown. I have done over current > shutdowns before and have used those over current passive devices that self > heal. Polyfuse I think is the name.... > > Like you need more ideas... > > *From:* Craig House <[email protected]> > *Sent:* Friday, April 24, 2015 5:26 AM > *To:* [email protected] > *Subject:* Re: [AFMUG] Packet flux question > > Thanks Forrest. It was a mix of 450 and 100 series. They all appear to > have been damaged. The only thing I get an Ethernet link light on is a > bh50 radio > > Sent from my iPhone > > On Apr 24, 2015, at 05:06, Forrest Christian (List Account) < > [email protected]> wrote: > > Hmm.. I love (not) autocorrect on android... port, not portal... But > now I'm on a real computer and have a chance to re-read the original > message, and think a bit... I think I need to change my final answer. > > If these were 100 series radios I'd say "that seems rather odd that this > would cause a failure". BUT... I'm assuming these are 450 radios. > > With the 450, there's a ethernet transformer on each pair. To DC, this is > effectively a short. Or since these are made with very thin wire you > could probably more accurately call it a 'fuse'. So if you take a pair > and put say the + lead of a 24V power source on one wire in the pair, and > the return (-) on the other pair, you'd find that the wire in the > transformer would melt, and would probably do so very quickly. This is > *exactly* the wiring that the 320/430 radios used. In addition, there is > every possibility that the current being drawn before melting is smaller > than the amount of current needed by a real 320 or 430 radio on power on. > So, when this got plugged in, there's a good chance that you melted the > ethernet transformers. > > The good news is if this is what has happened, it should be a fairly easy > fix by almost any electronic repair shop which knows how to rework surface > mount boards - just remove the magnetics and replace them. > > Unless of course there was another cause. > > -forrest > > > On Thu, Apr 23, 2015 at 9:58 PM, Forrest Christian (List Account) < > [email protected]> wrote: > >> Due to the odd wiring the radios probably shorted out the overcurrent >> protection in the injectors, turning off the portal and I'd not be >> surprised if the radios are just fine. Especially if they were never >> plugged into an already on injector. >> So we accidentally put sync injectors on to a din rail today that were >> for the 320/430 radios. Oops >> >> Both of the injectors were powered by a 24 V 10 amp power supply >> All of the radios that were plugged into those injectors no longer appear >> to boot up which wouldn't surprise me if there had been a 56 V power supply >> or 48 V power supply powering them. However since they were powered by a >> 24 V power supply how could that have damaged the radios? >> >> Sent from my iPhone >> > > > > -- > *Forrest Christian* *CEO**, PacketFlux Technologies, Inc.* > Tel: 406-449-3345 | Address: 3577 Countryside Road, Helena, MT 59602 > [email protected] | http://www.packetflux.com > <http://www.linkedin.com/in/fwchristian> > <http://facebook.com/packetflux> <http://twitter.com/@packetflux> > >
