Power Over Ethernet camera repair

[tyrel]

Hi,

I try to repair a network camera (Axis Q6032-E) that has died suddenly after more than 10 years of service.

I strongly suspect that the problem is in the Power Over Ethernet power supply board, and more specifically in the flyback converter, since i measure no voltage at the output of the converter.
There is no other way than POE to power the camera.
Visually, i see nothing wrong on the board. The capacitors look good, and their measured value is ok (i have not measured esr).

The power supply is composed of two isolated flyback converters controlled by two NCP1081 (https://www.onsemi.com/pub/Collateral/NCP1081-D.PDF).
The circuit seems very close to the schematic in figure 2 of the datasheet, except that it uses the extra winding (see figure 4), and two completely separate converters in parallel for data pairs and spare pairs.

I have measured several voltage (DC) at various points of the two NCP1081 (see below), and the results look definitly odd, but i can't understand what is going on exactly.
The measured voltages are the same whether the other boards of the camera are connected or not.

converter 1 (from spare pairs)

VPORTP-VPORTN : 53.8V (ok)
ARTN-VPORTN : 52.2V
VDDH : 1.3V
VDDL : -0.1V
VCpd : 1.6V

converter 2 (from data pairs)

VPORTP-VPORTN : 53.8V (ok)
ARTN-VPORTN : ramp down, reset to 53.8V (symetric to VCpd)
VDDH : 0V
VDDL : 0V
VCpd : slowly ramp up to variable voltage (seen up to 42V), go to 0, repeat

Any idea?
Thanks.

[james_s]

Between two friends and I, I think we had 5 Axis 206 cameras fail over the course of about 4 years, I haven't bought any more since then.

[station240]

So converter 1 looks like it's working briefly, but is overloaded somehow.
converter 2 appears totally dead.

If you look closer at the internal diagram, you can see ARTN to VPORTNx is the negative path, but has current limiting.
So the reason converter 2 appears totally dead, is there is a short on the output that is shutting everything down.
Check for bad diodes and capacitors that measure bad or run hot.

[PKTKS]

Check also that so called SHUNT regulator which I will guess should be the TL431 or clone

It that is still matching the proper reference or just gone wild.

Paul

[tyrel]

Thanks for your ideas, i don't have the board right here to test, but i will try tomorrow.

Everything after the flyback looks ok,  when i connect a 19V power supply at the output capacitor, the camera boards boots (at least i see the led turn on).
Also, something i forgot to mention is that it is not one, but two of those cameras (same model, same installation date) that have failed in two months.
I have no access to the second camera for now unfortunalty (11 meters above the ground).

If you look closer at the internal diagram, you can see ARTN to VPORTNx is the negative path, but has current limiting.
So the reason converter 2 appears totally dead, is there is a short on the output that is shutting everything down.
Check for bad diodes and capacitors that measure bad or run hot.

Yes, you are right, there is likely a short at converter 2 output.
I have checked most capacitors and diodes, but i will check again more thoroughly in that area.
There is probably something wrong on the controller 1 too (or he is affected by controller 2), because the ramp slope on VCpd (about 0.3V/s, for a 100uF capacitor iirc) is too slow for a normal current limit.

Check also that so called SHUNT regulator which I will guess should be the TL431 or clone

I did not think to check that, will do, thanks

Between two friends and I, I think we had 5 Axis 206 cameras fail over the course of about 4 years, I haven't bought any more since then.

I don't have had any bad experience like that with Axis, but i have not used their 'low-end' cameras much. Were your cameras indoor or outdoor?
In my case, a lifetime of 12 years for an outdoor camera does not seem too bad for me.

[james_s]

These were all indoor. The Axis 206 was a high end IP camera in its day, IIRC they were over $300 each in 2005.

[Dannyx]

Hi. Have you ever managed to get these cameras working again ? I happen to have the exact same problem, though I do not get a constant, stable 50v on the input caps - it seems to shut down and "retry" every 2 seconds or so, so I was curious if you ever found the culprit. Thanks

[Mavro]

just as a follow up -- I had an issue with PoE not working on a P5414-E but I suspect this is a common issue with the NCP1081 -- it would be into a retry state, never powering up.  Applying power directly camera worked.   I replaced the NCP1081 (the P5414-E) only had 1 NCP1081.  That did the job.

[zener]

I picked up 8 Axis Q6032-E cameras at a government surplus auction.   Out of the 8 only one worked properly.   One worked fully.   One other would come up, and the menus would work, but I would get a camera error(0007) and no video.   By swapping parts, I determined that 7 cameras modules were OK, and 2 power supply boards were OK.

I will get some NCP1081 chips and try replacing them.

There is a 20 pin connector between the camera and power supply.
Connector type: HJST 20
These are the voltages that I read on a good supply board.
They are read with the negative probe on a mounting screw.
The #1 pin is nearest to the straight edge of the board.
The number 20 pin is furthest from the straight edge of the board.

1   3.4v
2   3.4v
3   GND
4   0.23   
5   0.0
6   0.0
7   19.4v
8   11.9v
9   3.4v
10   GND
11   3.4v
12   3.4v
13   0.0
14   0.26
15   0.0
16   GND
17   19.4
18   GND
19   3.4
20   0.0

The 2 good supplies looked like this, clearly a redesign;

The 6 bad supplies looked like this;