POE CAT6 Tests OK Yet No Camera Connectivity

BlaineBug

Young grasshopper
Joined
Jun 27, 2022
Messages
56
Reaction score
18
Location
United States
I have installed a Reolink 6 security camera system on an 8 channel 2TB NVR about 13 months ago in June of 2021. The system was installed for about 1 full year and everything worked flawlessly. I used my own FastCat CAT6 solid core ethernet cable for the installation, and up until 2 or 3 weeks ago haven't had an issue.

In June of 2022 one of my cameras I noticed had no signal. If I unplugged the camera from my NVR, I could get it to resume functionality, although this didn't last for good. Connectivity from the camera could range from anywhere from a few minutes to a few hours. So, since my system had a 24 month warranty, I took advantage of that and Reolink shipped me a replacement camera.

I get the camera and promptly install it. Everything seemed ok, but once again less than 2.5 hours after installation I once again lost camera feed. I unplugged the camera and could not re-establish a connection.

So I had both cameras, I used short leads and tested them right in my basement and connected them directly to the NVR. Both the OLD camera and the NEW camera function.

I had a cheap ethernet continuity tester that has the blinking LED light, continuity checked out OK. So I purchased a more advanced ethernet tester with POE testing functionality and I tried this as well.

One again continuity checks out good with the new ethernet tester, but I'm showing some odd stuff for the POE test. Can you make any sense of what is going on? According to the manual, there are no shorts because if there were, I would be seeing a different display on the continuity test if so!

I am attaching some pictures showing the LCD display of the new tool. When in POE testing mode, the ethernet cable is plugged into the NVR and my device on the camera end. When it continuity testing mode, the device and a receiver are connected to the ethernet cable without having the NVR connected at all.

Continuity test

20220707-133124[1].jpg

POE test

20220707-132604[1].jpg
20220707-132615[1].jpg
 
Joined
Dec 28, 2019
Messages
11,508
Reaction score
27,487
Location
New Jersey
Continuity is nice to be able to test for as a basic test, but it doesn't actually test for real data connectivity. If both cameras work with a shorter cable on the test bench the only thing different is the cable. I'd suggest re-terminating the cable, try the far end first. It doesn't take a lot of corrosion, and it may not even be noticeable to the eye, to mess up data transmission.

Assuming the camera is actually outside, use some dielectric grease on the RJ45, just a small dab, before plugging it into the camera. Best practice is to use dielectric grease, install the cable gland, wrap with self amalgamating tape like Coax Seal then wrap with a quality electrical tape like 3M 33+ or 88. Even the minor condensation caused by temperature changes can cause corrosion over time.
 

BlaineBug

Young grasshopper
Joined
Jun 27, 2022
Messages
56
Reaction score
18
Location
United States
Continuity is nice to be able to test for as a basic test, but it doesn't actually test for real data connectivity. If both cameras work with a shorter cable on the test bench the only thing different is the cable. I'd suggest re-terminating the cable, try the far end first. It doesn't take a lot of corrosion, and it may not even be noticeable to the eye, to mess up data transmission.

Assuming the camera is actually outside, use some dielectric grease on the RJ45, just a small dab, before plugging it into the camera. Best practice is to use dielectric grease, install the cable gland, wrap with self amalgamating tape like Coax Seal then wrap with a quality electrical tape like 3M 33+ or 88. Even the minor condensation caused by temperature changes can cause corrosion over time.
Thank you for the reply. Last week to rule out the possibility of corrosion I did snip back the wire about 6" and re-terminated with a new plug. No change.
 

mat200

IPCT Contributor
Joined
Jan 17, 2017
Messages
10,945
Reaction score
16,642
What else can it be?

What changed in June?

wonder if there is some RF interference ?

Have you added anything else electrical to your home in June?
 
Joined
Dec 28, 2019
Messages
11,508
Reaction score
27,487
Location
New Jersey
If you already re-terminated then something has gone wrong with the cable. It would take a TDR capable tester to determine the likely cause and it would also show you where the problem actually is. Unfortunately a TDR capable tester is well into the hundreds or even thousands of dollars. Not worth it for us home users.
 

IAmATeaf

Known around here
Joined
Jan 13, 2019
Messages
2,782
Reaction score
2,348
Location
United Kingdom
How long are the runs?

Also would POE be via the pairs 3, 4, 5, 6?

Lastly what happens if you run a continuous ping/trace to the cam, look at the total return times, they should be fairly similar.
 
Last edited:

ThomasCamFan

Pulling my weight
Joined
Dec 14, 2020
Messages
96
Reaction score
128
Location
USA
I'd say the POE tester is reporting "Non-Standard!" because of the low wattage (9W) measurement.

Standard POE is typically 15W and should not be less that 12W on longer runs. If the tester is accurate then either the cable has deteriorated or the POE Switch's port is in trouble.

For comparison, test a different camera cable run that is about the same length and see what the wattage is.

BTW, the POE tester appears to show your cable is wired for T568A, whereas most of us use T568B. But either wiring format is acceptable. And some say that T568A is better for long POE runs (I have no idea if this is true).

I used my own FastCat CAT6 solid core ethernet cable for the installation, and up until 2 or 3 weeks ago haven't had an issue.
I looked at the reviews and most are positive. But there are some reporting sharp kinks out of the box, undersized wire gauge, bad performance, and even a comment about brittle copper.

- Thomas
 
Last edited:
Joined
May 1, 2019
Messages
2,093
Reaction score
3,211
Location
Reno, NV
plug a different network device in, instead of the camera (laptop, another camera, etc). this will determine if it's a faulty camera
you could bring the camera in and plug directly into switch/router port you are currently trying to use. this will determine it is a wiring issue
 
Joined
May 1, 2019
Messages
2,093
Reaction score
3,211
Location
Reno, NV
oh...missed the 'new' camera for replacement part and the bench testing.
I'd bet...POE issue with the NVR. Try different port.

When network devices for for weeks/months/years...and then something goes amiss, it's 99.9% never the cabling. Could be the wall plate jack got messed up or other vendors stepping on cable in ceiling or accidentally cutting it. But usually it's the router/switch/POE issue.
 
Joined
May 1, 2019
Messages
2,093
Reaction score
3,211
Location
Reno, NV
I'll bet cable. Just a feeling although trying another port can't hurt and costs nothing to try.
I do admit....because my work is a professional workplace, we do buy top notch cabling manufactures, jacks, RJ45 mod plugs, etc. I take it for granted that, at least, my work comes with lifetime guarantee because of using the right stuff. Unsure of off-brand quality.
 
Joined
Dec 28, 2019
Messages
11,508
Reaction score
27,487
Location
New Jersey
I always use, and always used, the best grades I could find, too. Even with that it doesn't take much for Murphy to show up and spoil the party. Worst cases were the squirrels and mice eating the cable. Back then I had a TDR and could get a really good idea where the problem was. I'll never forget the first time I traced one out and found the cable totally cut by squirrels.
 

wittaj

IPCT Contributor
Joined
Apr 28, 2019
Messages
14,967
Reaction score
27,853
Location
USA
+1 above - we have seen many people where over time the POE ports degrade on cheap NVRs and cannot push the same power thru at distance when it was new. Doesn't take much deviation for a camera to not work.

I had a consumer grade NVR once (okay maybe twice or three LOL) and after awhile it couldn't push enough power thru 60 feet of cable to power the camera. If I put a POE injector on one end and the camera on the other it was fine, so it wasn't the cable. I have had that happen, and the HDMI port not able to keep up with the distance after a years use.

And maybe among other things, Reolinks get more power hungry as they age :lmao:
 

looney2ns

IPCT Contributor
Joined
Sep 25, 2016
Messages
13,773
Reaction score
18,602
Location
Evansville, In. USA
I do admit....because my work is a professional workplace, we do buy top notch cabling manufactures, jacks, RJ45 mod plugs, etc. I take it for granted that, at least, my work comes with lifetime guarantee because of using the right stuff. Unsure of off-brand quality.
Pray tell, please share those brands.
 

wittaj

IPCT Contributor
Joined
Apr 28, 2019
Messages
14,967
Reaction score
27,853
Location
USA
Just to keep things in one place, he basically has the same thread going twice...

 

BlaineBug

Young grasshopper
Joined
Jun 27, 2022
Messages
56
Reaction score
18
Location
United States
What else can it be?

What changed in June?

wonder if there is some RF interference ?

Have you added anything else electrical to your home in June?
No, nothing new electronic in my home in June. Nada. The only difference is this is when we experienced a lot of 90+ degree days. That would be the only difference. However we had 90+ days last summer as well when all of these cameras were installed as they were installed June of 2021, before the peak of summer heat.

If you already re-terminated then something has gone wrong with the cable. It would take a TDR capable tester to determine the likely cause and it would also show you where the problem actually is. Unfortunately a TDR capable tester is well into the hundreds or even thousands of dollars. Not worth it for us home users.
True.

How long are the runs?

Also would POE be via the pairs 3, 4, 5, 6?

Lastly what happens if you run a continuous ping/trace to the cam, look at the total return times, they should be fairly similar.
I have not yet run a continuous ping/trace to the cam, I'm not exactly sure what that is, I am not a pro. As far as the run length I am unsure in all honesty, however this particular camera in question IS the furthest run. And according to my meter it appears that wires 1,2,3,6 are the ones carrying the power (as illustrated by those numbers having a black background.)

I'd say the POE tester is reporting "Non-Standard!" because of the low wattage (9W) measurement.

Standard POE is typically 15W and should not be less that 12W on longer runs. If the tester is accurate then either the cable has deteriorated or the POE Switch's port is in trouble.

For comparison, test a different camera cable run that is about the same length and see what the wattage is.

BTW, the POE tester appears to show your cable is wired for T568A, whereas most of us use T568B. But either wiring format is acceptable. And some say that T568A is better for long POE runs (I have no idea if this is true).


I looked at the reviews and most are positive. But there are some reporting sharp kinks out of the box, undersized wire gauge, bad performance, and even a comment about brittle copper.

- Thomas
It's possible that there is a copper problem, there are some pretty tight bends in the cables especially where they enter conduit and make their way from the second floor soffit on my home down to the basement. Most of the runs are enclosed in the aluminum soffit/fascia on my home and then transition into PVC conduit down the siding and into the basement. However all 6 cameras have their wire routed exactly the same way.

plug a different network device in, instead of the camera (laptop, another camera, etc). this will determine if it's a faulty camera
you could bring the camera in and plug directly into switch/router port you are currently trying to use. this will determine it is a wiring issue
I have not yet tried another device like my laptop but I do have 2 cameras here, the "old" and the brand new replacement, both are non-functional when using the CAT6 cable outside of my home but both work when connected directly to the NVR in my basement.

Given that both cameras work on the bench the cable has gone bad for some reason. Whether it's a PoE problem on the cable or a data problem on the cable, it's academic. The cable is shot.
Possibly, although why would I see continuity on all 8 wires?

oh...missed the 'new' camera for replacement part and the bench testing.
I'd bet...POE issue with the NVR. Try different port.

When network devices for for weeks/months/years...and then something goes amiss, it's 99.9% never the cabling. Could be the wall plate jack got messed up or other vendors stepping on cable in ceiling or accidentally cutting it. But usually it's the router/switch/POE issue.
I have no wall plate jacks, both ends of CAT6 are terminated and that is that. No keystones or adapters or anything else. Each run of CAT6 simply has 2 RJ45 plugs and that is all. Also no one else stepping on the cable, it's mostly run in the soffit/fascia of my home and then in PVC conduit where it comes down from the second floor fascia and then into my basement.

I'll bet cable. Just a feeling although trying another port can't hurt and costs nothing to try.
Cameras work indoors when connected with short ~5' cable. I've tried plugging the full length run CAT6 into a variety of other ports on the NVR, which doesn't affect anything.

I do admit....because my work is a professional workplace, we do buy top notch cabling manufactures, jacks, RJ45 mod plugs, etc. I take it for granted that, at least, my work comes with lifetime guarantee because of using the right stuff. Unsure of off-brand quality.
Ok.

I always use, and always used, the best grades I could find, too. Even with that it doesn't take much for Murphy to show up and spoil the party. Worst cases were the squirrels and mice eating the cable. Back then I had a TDR and could get a really good idea where the problem was. I'll never forget the first time I traced one out and found the cable totally cut by squirrels.
I guarantee you there are no rodents in my attic. No birds. No mice. Nothing.

+1 above - we have seen many people where over time the POE ports degrade on cheap NVRs and cannot push the same power thru at distance when it was new. Doesn't take much deviation for a camera to not work.

I had a consumer grade NVR once (okay maybe twice or three LOL) and after awhile it couldn't push enough power thru 60 feet of cable to power the camera. If I put a POE injector on one end and the camera on the other it was fine, so it wasn't the cable. I have had that happen, and the HDMI port not able to keep up with the distance after a years use.

And maybe among other things, Reolinks get more power hungry as they age :lmao:
It's possible, this is the longest run of cable, I'm not sure of the total length though. Someone else suggested I try a POE injector. I thought of that myself. I'm curious though, since I cannot turn off POE to specific ports on this NVR (it's not a POE switch, per se) can I use a passive POE injector in-line with this POE NVR without being able to disable the POE on the NVR? I thought that might be a cheap test as well, in the event that the camera is no longer receiving the proper power.

Pray tell, please share those brands.
What are you asking?
 

wittaj

IPCT Contributor
Joined
Apr 28, 2019
Messages
14,967
Reaction score
27,853
Location
USA
You would have to run your cameras to a POE switch and then connect the NVR WAN/LAN to the POE switch.

But if you were running just a POE injector on one port, the NVR should recognize that it doesn't need to send out power and just accept the data. But then again with Reolink who knows LOL.
 

CCTVCam

Known around here
Joined
Sep 25, 2017
Messages
1,912
Reaction score
2,392
How long are the runs?

Also would POE be via the pairs 3, 4, 5, 6?

Lastly what happens if you run a continuous ping/trace to the cam, look at the total return times, they should be fairly similar.
It appears to be connected via 1,2,3 and 6.

Shouldn't there be at least 5 active connections for a camera? The diagram I found suggests pins 1,2,3 and 6 are for Rx / Tx and pin 5 is for power. Wonder if there's some issue with power transmission as someone noted above with the NVR? If 5 is missing on transmission but connected in continuity, that suggests a POE problem.
 

BlaineBug

Young grasshopper
Joined
Jun 27, 2022
Messages
56
Reaction score
18
Location
United States
Continuity is nice to be able to test for as a basic test, but it doesn't actually test for real data connectivity. If both cameras work with a shorter cable on the test bench the only thing different is the cable. I'd suggest re-terminating the cable, try the far end first. It doesn't take a lot of corrosion, and it may not even be noticeable to the eye, to mess up data transmission.

Assuming the camera is actually outside, use some dielectric grease on the RJ45, just a small dab, before plugging it into the camera. Best practice is to use dielectric grease, install the cable gland, wrap with self amalgamating tape like Coax Seal then wrap with a quality electrical tape like 3M 33+ or 88. Even the minor condensation caused by temperature changes can cause corrosion over time.
You would have to run your cameras to a POE switch and then connect the NVR WAN/LAN to the POE switch.

But if you were running just a POE injector on one port, the NVR should recognize that it doesn't need to send out power and just accept the data. But then again with Reolink who knows LOL.
As a cheap test I could do that, just a single POE injector on one port which leads to this camera in particular with the issues. How much wattage do you think I should be seeing? Also, would it make a difference that I did not connect the camera to my testing device? My device has two POE test ports so essentially you can put the testing device in-between the camera and the NVR, which I didn't do. As much as I know, I believe POE "negotiates" the power required by the device as most NVRs and Switches are NOT passive in nature. I also read today that Reolink cameras typically use less than 8 watts, so I thought I should be fine even if I am seeing 9 watts with the tester, but who knows!
 
Last edited:
Top