Topic: New Ecowitt HP2561, dead sun sensor

[HaraldS]

Got everything working nicely including additional sensors and integration with Wunderground, Ecowitt app and site, and Home Assistant. Labels and calibrations are set up, too.
Unfortunately I just discovered that sun and UV have been flatlined. I checked for protective foil, reset the array, checked calibration for default, etc.

That sensor is dead. We got lots of sun in California!

I send a note to Ecowitt on their site. How responsive are they? I would like a new array and then send back the old one, if needed. I do not want to spend a month without the rest of the setup. Any other experiences with them?

-- Harald

[Gyvate]

A note on their site will not be the proper approach.
Send an email to support@eowitt.com and describe what doesn't work and what all you have done.
They will come with some suggestion or even right away send you an replacement. Either the full array or the upper part which can be easily (with instructions) be exchanged.

[HaraldS]

Thanks, sent them a note via email.

[Gyvate]

A note on their site will not be the proper approach.
Send an email to support@eowitt.com and describe what doesn't work and what all you have done.
They will come with some suggestion or even right away send you an replacement. Either the full array or the upper part which can be easily (with instructions) be exchanged.

The replacement part applies to the WS80 or WS90 arrays - with a WS69 (HP2651 = HP2560 + WS69) it would probably have to be the whole piece as its solar sensor is not so easily replaced if you don't have a certain expertise.

[HaraldS]

Ecowitt support responded swiftly to the email.
They would like me to do the following steps. These make sense and are good to know:
1.     Take down ws69 sensor. Press “reset” button to see if the device is working for light&uv.
2.     Take out battery and put the array inside a dark room for 48hrs until the solar energy has been 100% discharged.
3.     Inside the dark room, put the battery in to start up.  Check light and uv is normal .

I'll report my results in 48 hours.

[Gyvate]

correct - pressing the reset button only is what we call a "warm" or "hot" reset
removing the batteries and letting the super-capacitor die (discharge completely) and re-inserting the battery is the "cold" reset.
If you want to continue receiving the other data until it finally "dies", take out the batteries and tightly cover the solar panel with a few layers of black tape. The array will contnue sending the data from the other sensors until the super-capacitor is expired.

The black tape is the "online" version of the "offline" dark room.
It can take up to three days if the supercapacitor was fully charged.
The sign of complete discharge is: no more other sensor data received by the console and the red LED of the array stops blinking (every 16 seconds).
Not 48 hours are the criterion - the complete discharge of the supercapacitor is !
(But often 48 h will do the job).

[Rover1822]

I am not familiar with the unit , but on the reset push, most take a hold down for > 5 seconds , sometimes 10. But I defer to others that may know

[Gyvate]

you need a staple end to insert into the reset hole until the LED lights up - then release
it's described in the manual 
@Rover1822: @HaraldS has a HP2561 station = HP2561 console + WS69 outdoor array - in principle the same as the WS-2000 7-in-1 outdoor array
reset works the same way

[HaraldS]

correct - pressing the reset button only is what we call a "warm" or "hot" reset
removing the batteries and letting the super-capacitor die (discharge completely) and re-inserting the battery is the "cold" reset.
If you want to continue receiving the other data until it finally "dies", take out the batteries and tightly cover the solar panel with a few layers of black tape. The array will contnue sending the data from the other sensors until the super-capacitor is expired.

The black tape is the "online" version of the "offline" dark room.
It can take up to three days if the supercapacitor was fully charged.
The sign of complete discharge is: no more other sensor data received by the console and the red LED of the array stops blinking (every 16 seconds).
Not 48 hours are the criterion - the complete discharge of the supercapacitor is !
(But often 48 h will do the job).

This was the point of the support instructions. It is somewhat unusual. Normally a reset forces the SoC to completely reset and that's that. I suspect this is due to having multiple components involved.

[Gyvate]

[It is somewhat unusual. Normally a reset forces the SoC to completely reset and that's that. I suspect this is due to having multiple components involved.

I don't think so - you do the same with a computer/server.
Restart or Shutdown and New start.
The restart only doesn't re-initialize the whole hardware - only part of it respectively only partially. Whereas a shutdown and a complete new start re-initializes the whole hardware.
Reset under power and full re-initialization are not the same. For the full re-initialization (power cycling) the piece (whatever SoC, mainboard ...) will be set off-power first.
E.g memory (RAM) may be wiped during a reset, but it will be completely re-initialized with power-cycling.
So not really unusual but just different levels of reset. That's why such procedures are dubbed "warm" and "cold" in IT and they serve different purposes.
Often a restart is enough - but sometimes a full re-initialization is needed.

[HaraldS]

[It is somewhat unusual. Normally a reset forces the SoC to completely reset and that's that. I suspect this is due to having multiple components involved.

I don't think so - you do the same with a computer/server.
Restart or Shutdown and New start.
The restart only doesn't re-initialize the whole hardware - only part of it respectively only partially. Whereas a shutdown and a complete new start re-initializes the whole hardware.
Reset under power and full re-initialization are not the same. For the full re-initialization (power cycling) the piece (whatever SoC, mainboard ...) will be set off-power first.
E.g memory (RAM) may be wiped during a reset, but it will be completely re-initialized with power-cycling.
So not really unusual but just different levels of reset. That's why such procedures are dubbed "warm" and "cold" in IT and they serve different purposes.
Often a restart is enough - but sometimes a full re-initialization is needed.

A computer or server is an aggregate device with multiple subsystems. Most IoT devices are not.

[Gyvate]

whatever you say ....
But I'm sure the principle still applies.

[HaraldS]

Used the blackout tape approach and took out the batteries. Three days later and the capacitor battery is still going strong. Wow!

[Rover1822]

Used the blackout tape approach and took out the batteries. Three days later and the capacitor battery is still going strong. Wow!

It can take some time

[HaraldS]

Ok. Just in case the dark tape leaked (don't see how,) I put it in a dark closet. It is still sending. The two batteries are on my desk, I know I took them out.
Started this Saturday AM, it is now Thursday PM.

Five days and counting. That's a hell of a capacitor.

[Rover1822]

Just want to make sure you are checking using the WSView app , not the Ecowitt. net site to check status.

[HaraldS]

I look at the HP2561 panel and its connection status/strength plus changes in temp.

[Rover1822]

OK, that is good enough

[Gyvate]

As long as the reception symbol right of the WiFi symbol (that's the WS69/WH65 array - see also the console display pictogram in the MUST READ thread) shows bars and no red battery symbol appears, the array is still transmitting. In principle the battery symbol should appear and turn red, and the bars and the reception symbol should disappear when the array stops transmitting.
I say in principle as you may not see it happen one after the other ...

[HaraldS]

So...
This is embarrassing but I will fess up to it as a lesson for other users. I was thinking about the issues and the bulb went off in the middle of the night on what was happening.

When the HP2561 panel came up originally it linked to my old Ambient array that has been sitting on the side of the house for recycling. This is why there was no solar and UV (it was sitting in the shade) and so little wind reporting. But things had been calm around here. It's been bright enough to keep that array going with jsut indirect sun light.

I cleared the sensor entry and fired the new array up and all is fine. To make sure I spun the wind direction and speed sensor and ran some water through the gauge before mounting it back on the roof.

[Gyvate]

you didn't mention the old array in your post before ....
otherwise one would have checked the sensor ID of the outdoor array as a potential source of trouble.
When two are active at the same time, the console chooses randomly - and you can register the "right" one only if you know its sensor ID, but you wouldn't have had a chance for this as long as the old one was transmitting as there was no chance to get to know the sensor ID of the new one.

So both arrays had either to "die" or being taken at least 1/2 mile away from the console so the console could connect to the wanted one. It's good to keep a record of ones sensor IDs (also in other contextes).

If you had known/remembered the sensor ID of the old array, you could have realized that the old was still connected ....

[HaraldS]

Yup. Embarrassing. But I am sure I will not be the only one to ever do this.

This is why a fessed up to it.

 

[Rover1822]

This is not an uncommon occurrence, just be glad it was "your" unit. As these become more popular, we have had instances where the console picks up a neighbor's array, bit more fun . A few times it happens because one neighbor sees the other neighbor's weather system and wants one themselves.