Cable splitters and poe problems.

That looks good! I'm envious!
What I did, to reduce the chance of damaging the solder pads, was to snip the legs off and desolder the stubs one by one.
Not quite the same approach as removing an SMD device.
 
Toan the Moan said:
Orange and Orange white become positive and Green and Green white become Negative when pin checking the Green out of the splitter.
Click to expand...
Sure looks like you've found the answer. If you're able to swap the transformer and get the camera fixed I'll consider that a victory for the good guys, as the saying goes. I personally don't have a very good record working with the new micro geometries and SMDs, probably having destroyed more than I've fixed. Back in the days of tenth-inch grids I was a soldering whiz.

Having said the important things, my OCD is screaming that there's something fishy going on, in that the power should have never appeared on the green and orange pairs. Looking at the power/data combiner adapter:
splitter-1.jpg
I believe (and told you earlier) that the data exits the adapter on the orange/green pairs, and power on blue/brown. Then looking at the ali express combiner:

splitter-2.jpg
Both of the male RJ45s should connect to pins 1,2,3,6, where the green/orange pairs normally connect to. I think I can see that in the picture. Therefore, power should have been not connected at this point, and the power should have never been sent down the long ethernet cable.

So now I'm wondering if that first power/combiner adapter is scrambling the pin assignments in a way I think they shouldn't? If the accompanying splitter on the other end is compatible it doesn't matter, but with the ali express splitter in the middle, what you're seeing would happen.
 
  • Like
Reactions: Toan the Moan
tigerwillow1 said:
Sure looks like you've found the answer. If you're able to swap the transformer and get the camera fixed I'll consider that a victory for the good guys, as the saying goes. I personally don't have a very good record working with the new micro geometries and SMDs, probably having destroyed more than I've fixed. Back in the days of tenth-inch grids I was a soldering whiz.

Having said the important things, my OCD is screaming that there's something fishy going on, in that the power should have never appeared on the green and orange pairs. Looking at the power/data combiner adapter:
View attachment 74605
I believe (and told you earlier) that the data exits the adapter on the orange/green pairs, and power on blue/brown. Then looking at the ali express combiner:

View attachment 74606
Both of the male RJ45s should connect to pins 1,2,3,6, where the green/orange pairs normally connect to. I think I can see that in the picture. Therefore, power should have been not connected at this point, and the power should have never been sent down the long ethernet cable.

So now I'm wondering if that first power/combiner adapter is scrambling the pin assignments in a way I think they shouldn't? If the accompanying splitter on the other end is compatible it doesn't matter, but with the ali express splitter in the middle, what you're seeing would happen.
Click to expand...
I had concerns when the yellow splitter router rj45 terminated at the green out of the splitter T. I should never had tried the 12v power injectors after that.
 
dahukvision said:
Is this what your thing is supposed to do?
Click to expand...
Might have worked if I'd used one of the 48v to 12 v converters, and done it like that. I used a pair of 12volt dc injector adapters as I was testing the cable splitters at the router on the bench and not on the POE hub which is in another building. Why I decided to test the splitters with the 12volt injectors god knows as I'd already opted for a direct cable from camera to hub and use the injector adapters, which after seeing your post I should have used one of the 48v to 12 volt adapters I have . I feel a right Rodney plonker Trotter now
 
Toan the Moan said:
I feel a right Rodney plonker Trotter now
Click to expand...
Being on the left side of the Atlantic I had to look him up. My response is that some of the best things I've learned have happened while watching a perfectly good device of some sort go up in smoke. Painful at the time, of great value in the long run.
 
  • Like
Reactions: alastairstevenson
tigerwillow1 said:
Being on the left side of the Atlantic I had to look him up. My response is that some of the best things I've learned have happened while watching a perfectly good device of some sort go up in smoke. Painful at the time, of great value in the long run.
Click to expand...
I forgot many members were from the other side of the Pond. Here's a link to the UK's famous comedy show Only fools and horses

One of the most famous sketches
And probably the most famous sketch triggers broom
 
Last edited:
  • Haha
Reactions: Ckb3
alastairstevenson said:
The ethernet signals are transferred in via an isolation transformer.
If indeed 12v has been applied between 1,2 and 3,6 then it's possible / likely that the ethernet transformer has been damaged.
If so - the camera will not appear on the network.
You may be able to see evidence of heat damage if you inspect the transformer.
It will be a plastic block about 12x8mm about 5mm height, 16 legs soldered down.
I don't see it on your camera image - it may be under the system board.

I've replaced several of these on cameras and NVRs - here is a damaged one from a DS-2CD2385G1 that I fixed up - you can see the fried windings on the first toroid.
Click to expand...
Massive thanks to Alastair my ebay ethernet transformers arrived today replaced the fried one and we are all fully operational again. £1.80 for 5 plus £3.60 postage.
 

Attachments

  • 20201125_144546.jpg
    20201125_144546.jpg
    1.8 MB · Views: 28
  • 20201125_144620.jpg
    20201125_144620.jpg
    4.2 MB · Views: 27
  • 20201125_144440.jpg
    20201125_144440.jpg
    2.6 MB · Views: 27
  • 20201125_144459.jpg
    20201125_144459.jpg
    2 MB · Views: 27
  • 20201125_144516.jpg
    20201125_144516.jpg
    1.4 MB · Views: 28
  • Like
  • Wow
Reactions: sebastiantombs, alastairstevenson and tigerwillow1
alastairstevenson said:
That is a brilliant result, well done. And compliments on the soldering if that's the finished result.
That failed transformer had been through hell.
Click to expand...
Thankyou, wouldn't have been able to fix it without yours and Tigers knowledge. Was quite easy to replace. I have had to change far worse SMT chips in my automotive career. Won't be using splitters and injectors together ever again. I shall be installing it 30 feet up over the weekend or should I say my lad will as I'm too old to climb up that high these days
 
I'm impressed. I think I've mentioned that I've destroyed a few things trying to swap surface mount chips. I suppose it might help if I had the right equipment.
 
Toan the Moan said:
I have had to change far worse SMT chips in my automotive career.
Click to expand...
I've done a bit of that, hobby-wise, but my results though functional haven't been that pretty.

Here is something I did the other day, fixing up a short circuit on a power rail.
The dug out through-hole was done to isolate the backside components and figure the location of the short.
It usefully turned out to be one of the 2 ceramic smoothing capacitors, not the tiny SOP6 next-level power conversion chip I'd expected.
Then the patch wire to replace the opened through hole and it burst into life.
I was dead chuffed!
 

Attachments

  • PB220005.JPG
    PB220005.JPG
    7 MB · Views: 34
  • Like
Reactions: fenderman
tigerwillow1 said:
I'm impressed. I think I've mentioned that I've destroyed a few things trying to swap surface mount chips. I suppose it might help if I had the right equipment.
Click to expand...
It certainly does, that JBC soldering station is brilliant had it for over 25 years was seriously expensive with all the tips I have for smt stuff back then. Certainly makes the job so much easier
 
Last edited:
Since I have a captive audience here.....
This is an oven control board that was designed to fail after a few years. The transistor hanging on its wires replaces a surface mount transistor on the back side of the board that got hot enough to start burning the board before it failed. The resistor at the top of the picture was also turning the board brown which is why I added the heat sink to it. I would have just bought a new board except it was NLA new or used for a GE oven only a few years old. It's been running ok for about 3 years now. I wouldn't want a job where a primary design goal is failure after a certain amount of time.
newTransistor_2 (Small).jpg
 
I see the faulty chip was a Glory and it burnt a hole. I take it from that your board had a Glory hole. It's no wonder you're happy it's working again. Just don't tell the wife!

On a more serious note, glad you got it working again. Only aspect that would worry me is what caused the chip to burn out in the 1st place. Hopefully it was just a co-incidence, although I'd have to worry if maybe the splitter was wired wrongly or had a short and sent power down the wrong conductors. If it stays fine, I wouldn't worry too much. If it doesn't then I suggest you revisit the splitter. If you have a means of testing, it might be worth doing a sanity check by checking which wires in the ethernet are live at the camera end.
 
CCTVCam said:
I see the faulty chip was a Glory and it burnt a hole. I take it from that your board had a Glory hole. It's no wonder you're happy it's working again. Just don't tell the wife!

On a more serious note, glad you got it working again. Only aspect that would worry me is what caused the chip to burn out in the 1st place. Hopefully it was just a co-incidence, although I'd have to worry if maybe the splitter was wired wrongly or had a short and sent power down the wrong conductors. If it stays fine, I wouldn't worry too much. If it doesn't then I suggest you revisit the splitter. If you have a means of testing, it might be worth doing a sanity check by checking which wires in the ethernet are live at the camera end.
Click to expand...
With tigers help we established after I did some pin out checks that the issue was caused by sending 12 volts dc up the orange wire. This was caused by using the splitters in conjunction with rj45 power injector adapters. It appears you can use one or the other but not both.
 
Last edited:
  • Like
Reactions: CCTVCam and alastairstevenson
You must log in or register to reply here.
Top Bottom