Page 1 of 2
Kids, don't try this at home
Posted: Tue Nov 04, 2014 10:46 am
by lligetfa
I had terminated a pair of drops in a PVC LB using keystone jacks and neglected to tape over them. I hotted one up with 24V and the other with 48V. The 24V is feeding a UVC DomeIR and I was trying to bring up a UVC Pro on the 48V drop. Cable test was showing a disparity in the pair lengths so I opened up the LB to check connections and made sparks when I shorted the 48V drop to the 24V one. The UVC DomeIR video feed went black but still showed link.
Power cycled the UVC and all is well.
Re: Kids, don't try this at home
Posted: Tue Nov 04, 2014 11:12 am
by sirhc
Yea, there is this ability to turn OFF POE to a port in the UI or even CLI prior to doing such things!
Also before you apply POE to a port/cable for the first time it is a good idea to run the Cable Diagnostics utility in the switch with the device connected to verify you have no cross shorts.
Did you fry the port in the switch?
The poly fuses are too slow to open all the time before damage can occur to the switch, normally the Ethernet transformers get burned out. We have found that under most circumstances 48V and especially 48VH will damage the Ethernet transformer effectively killing that port of the switch. 24V is a crap shoot but we know that repeated shorting of a 24V enabled POE port will after several shorts take out the transformer plus the transformers can be left in a state of degraded operation manifesting with Rx and Tx errors on that port.
At least a dead short on our switch does not damage the entire switch, just the port shorted. We destroyed several ToughSwitches while testing them, a dead short tended to take out the entire TS8.
We have a design on the table for the next generation of our switches (not any time soon) that uses intelligent circuits along with current sensors to detect a short or excessive current load and turn off POE before and damage can occur and eliminate poly fuses as the safety mechanism. It had to be done with hardware making the decision as software polling was too slow to achieve this. Next generation, if we are successful with this one anyway.
Most WISP's are smart enough to avoid too many shorts especially if they use the Cable Diagnostic tool provided with the WISP Switch to test cables BEFORE they apply POE power anyway.
There is far less of a chance of damaging a switch from cable damage shorts that occur while already in service as the poly fuse is already warm and trips much faster than a cold poly fuse. A cold poly fuse short almost always results in the Ethernet transformer getting damaged killing the port.
Re: Kids, don't try this at home
Posted: Tue Nov 04, 2014 11:18 am
by rockhead
We have a design on the table for the next generation of our switches (not any time soon) that uses intelligent circuits along with current sensors to detect a short or excessive current load and turn off POE before and damage can occur and eliminate poly fuses as the safety ecumanism
LOLOL, autocorrect of the year for that one ! Remember kids if you use a penny for a fuse always use an american penny cuz it says 'in god we trust' !
Re: Kids, don't try this at home
Posted: Tue Nov 04, 2014 11:36 am
by lligetfa
sirhc wrote:Most WISP's are smart enough to avoid too many shorts especially if they use the Cable Diagnostic tool provided with the WISP Switch to test cables BEFORE they apply POE power anyway...
So, you're saying I'm dumb? I'm OK with that hence the title. My error was in not taping over the keystone. Normally I use a larger NEMA box where I have enough room for a surface mount wallplate to put the keystone jacks into. I could have used two LBs but the second drop is temporary.
I did use the cable diagnostic tool but it showed a length disparity on pairs even with no jumper plugged into the keystone. Anyway, I had hotted it up at 24V first and tested at the LB using an old airCam Dome on a foot long jumper and it was fine. The UVC Pro doesn't have any port LEDs for verification of power and link so I use the airCam as a quick check. I sure do miss the Fluke tools I had on my last job.
Re: Kids, don't try this at home
Posted: Tue Nov 04, 2014 11:41 am
by sirhc
LOL - NO but I was poking at you a little for working on a live POE cable!
Re: Kids, don't try this at home
Posted: Wed Nov 05, 2014 11:04 am
by lligetfa
OK, I guess I should tie up a few loose ends...
First off the short was very brief. As I was moving the keystone jack, it made contact with the other one and made a spark. Neither port was damaged.
I mentioned that I tested the drop before I put PoE power to it but got disparity on length. In fact both drops tested out that way and so I used an airCam Dome on a foot long jumper to verify both drops. Later, in further testing I found that using a short 3 foot jumper between the switch and the patch panel caused the anomaly. A 6 foot jumper gave good results.
After testing the second drop at 24V with the airCam, I switched it to 48V. Let me preface to say that I have a UVC Pro mounted on a heavy umbrella stand that I move around to where I may want to record activity. It has a long jumper connected to the UVC Pro's dongle with a weathertight seal that is a royal pain to open. I've known that the ends of this long jumper are overcrimped and that it can be a bit of hit-and-miss when plugging it in. Anyway, it is a case of "the cobbler's kids go barefoot" and "do as I say, not as I do". My other UVC Pro has the same issue with an overcrimped end and I've been remiss to buy a new crimper to re-make the ends.
I do good work for others, really. It's just myself that I shortchange.
Re: Kids, don't try this at home
Posted: Wed Nov 05, 2014 11:37 am
by sirhc
lligetfa wrote:Anyway, it is a case of "the cobbler's kids go barefoot" and "do as I say, not as I do".
WOW, "cobbler", now there is a word I have not heard in a long time. Leave it to Les to mind us all of our age! - THANKS BUDDY!
Anyway I agree with your analogy, the painter's house is never painted, the mechanic's truck is always broken, and yes, "the cobbler's children go barefoot"!
We all tend to ignore what we preach, it's in our nature.
Re: Kids, don't try this at home
Posted: Wed Nov 05, 2014 12:10 pm
by lligetfa
I am a victim of my own frugality. That, and I hate to throw anything away.
At work I had encountered overcrimps on two fronts. One was with the crimper so I bought a good quality Greenlee ratcheting crimper. The other was with factory made jumpers. When we churned hundreds of computers, the built-in ethernet port on the new model had slightly heavier gauge wire spring contacts that caused the overcrimped factory jumpers to fail. I bought all new jumpers and tossed the old ones.
Now here is where my frugality and reluctance to throw anything away comes into play. Being too cheap to buy good quality jumpers for at home and being too honest to steal good ones from work, I took home some of those old overcrimped jumpers that were heading for the ewaste bin. Also I used the bad crimper from work at home. Now, the road to Hell is paved in good intentions, and I planned to crimp new ends on but first I need a new crimper.
Re: Kids, don't try this at home
Posted: Fri Nov 07, 2014 7:01 pm
by lligetfa
The following pic is from the UVC Pro that I mentioned in this thread. When I saw how the wife was about to use that rolling garden seat as a step stool, I thought to myself... Kids, don't try this at home. LOL
Re: Kids, don't try this at home
Posted: Fri Nov 07, 2014 7:13 pm
by sirhc
Nice! I take it and hope she lived through the experience?
Remind me not to break into your house, is that a puppy or a small horse? Is the puppy mean?