Indoor heat readings increases with Nano mounted

[nia]

Hi.

Just today, for my Davis VP II, I recieved my shiny new and impressively small Nano

While going through the configuration, I noticed the indoor temperature registered at 30C

It is clear, that it is the Nano that dissipates heat - opening up to the Nano, and putting a finger on the unit's CPU reveals that it's warm to the touch, quite clearly enough to throw off the indoor temperature sensor.

I have tried to remove the rear battery cover, but that only helps reducing it by 1C to 29C. The real temp is around 24,5C

Of course, I would like the indoor sensor to be correct.

Any suggestions?

Best regards

Niels

[SkidMark]

Niels,

Had the same problem with my Davis WeatherLink IP. A correction can be made to offset the change using the Davis program but I did not want to go that route just yet.

I came across this page http://wx.annoyingdesigns.com/DavisSPI.pdf Most of the info was getting over my head but may be of use to someone. It did help me find some parts to remote my logger out of the console.

I used the following: Samtec Inc. TCMD-10-T-06.00-01-N and ESQT-110-02-G-D-760 found at Digi-Key...

https://www.digikey.com/product-detail/ ... ND/1106729

https://www.digikey.com/product-detail/ ... ND/1995334

I cut the Header as needed to make my connection in the console to allow the battery door to fully close. The cable was
just able to fit into the opening that is already there.

The cable IS NOT PLUG AND PLAY! I took the female connector off of the cable then split the cable into 10 pairs.
Made a 1/2 turn of each of the pairs and then put the connector back on. I'll add a photo to show.

There are no guide pins to keep you from plugging it up wrong when in this configuration. Plugging it up wrong will most likely cause the temperature to increase in the wrong places. I may be able to post a photo Sunday of how it looks when plugged into the unit.

Most likely there is a better options but so far it has been working fine since 11/2016.

Did I mention the cable IS NOT PLUG AND PLAY! You will have to make the change to it.

Mark

[nia]

Hi Mark.

Thanks for your quick and elaborate resonse!

I was actually also thinking along the lines of an extension cable - really good to see that it can be done and how to go about it.

I'll be really interested in seeing your installation photo, particularly how you have dealt with the housing and aestaethichs. As it was supposed to be a very clean install on an otherwise clean wall - this is a challenge to say the least...

I'm somewhat disappointed that this is an issue though, and that it is not mentioned anywhere.

/Niels

[Mattk]

Has always been a known issue as even turning on the console backlight affects internal temperature which is why Davis included an additional connection to attach a temp sensor that works outside the console/envoy which effectively replaces the on board internal temp sensor.

[admin]

Regarding the additional heat there is no perfect way to avoid this. As said, also original Davis products add to that, but of course at a lower rate.
@Mattk: I was not aware that an external temp sensor can be attached to the console. How can one do that?

[Mattk]

The current Envoy definitely have a RJ11/12 connector beside the external power connector into which an external type Davis temp sensor can be plugged which overrides the on board sensor. Console version I will need to check but I'm fairly sure same applies for the same reason, I will confirm

[admin]

My (may be) older Vantage PRO2 console does not have a RJ11 or alike connector.

I just updated the SW, so that one can easily define a calibration of the indoor temp sensor of the console. To me -6° Fahrenheit looks like a reasonable adjustment, but this can be adapted to anyones needs. Calibration is done inside the console itself (not just in the Meteobridge SW), so all programs connecting to the console will benefit from it.

calib.jpg (234.62 KiB) Viewed 5107 times

[nia]

Console version I will need to check but I'm fairly sure same applies for the same reason, I will confirm

Would be interesting if it was so. At least, it is not immediately visible from behind the battery door like it isn't for admin, but then again, there may be a way to connect deeper in the device. My Pro2 is brand new, so nothing apparently changed there compared to admins station...

Otherwise, I guess the best option would be the extension cable approach.

As I see it, it should at least be long enough to be able to route the cable behind the console so the device can be somewhere above it (to avoid rising heat if it would be placed below).

@Admin: Would you consider making an option consisting essentially of something like the cable above, where the end point for the Nano being a very small box with the bus terminated inside, and a sticky-pad on the back to put it on the wall above the station...?
(Probably not worth the effort, but you would have at least one buyer here )

...and while writing the above, @admin posted about calibration. That's an impressingly fast reaction to the issue , and certainly a workaround worth investigating. I'm just worried about variables like load on the Nano (varying heat) or varying air movement around the console (varying effect of the internal heat buildup) can lead to inaccutate results. But I will set up a field experiment next to my old weather station (LaCrosse). Its internal sensor has always appeared to be quite accurate. I'll try to set up two instances of WD to log each station for comparison...

[admin]

Otherwise, I guess the best option would be the extension cable approach.

As I see it, it should at least be long enough to be able to route the cable behind the console so the device can be somewhere above it (to avoid rising heat if it would be placed below).

@Admin: Would you consider making an option consisting essentially of something like the cable above, where the end point for the Nano being a very small box with the bus terminated inside, and a sticky-pad on the back to put it on the wall above the station...?
(Probably not worth the effort, but you would have at least one buyer here )

Please don't do that. It will look awful. If you really insist, take the box the NANO was shipped in and cut something out to have the cable passing through. But come on, calibration is the better way. And it is just about indoor temp, nothing even a serious weather enthusiast would care too much.

[nia]

Please don't do that. It will look awful. If you really insist, take the box the NANO was shipped in and cut something out to have the cable passing through. But come on, calibration is the better way. And it is just about indoor temp, nothing even a serious weather enthusiast would care too much.

Fair point, quite possibly overthinking this massively - the wife does care a bit about the indoor temp, though, and it's always a good idea to keep them happy - not least when replacing a (on that particular parameter) well functioning weather station with a new setup, spending on the wrong side of €1k. But that can be fixed by other means

Actually, I was more just considering the cable, and then make a square hole behind the unit in the drywall just large enough to be able to push it in there - should do the trick...

But for now, I'll just start with the calibration approach while I focus on other aspects of setting the new station up; and then probably be all happy with it after calibration and a few days to accept that it's just as well.

[nia]

The new option to calibrate installed on the Nano at reboot just as it should (if only every software supplier had this turnaround time... )

It does its magic to the numbers visible in e.g. WeatherDisplay, although I need to adjust by around -5,3C to be about right - pending longer term observations. I have the Davis surrounded by 5 other thermometers currently , but things havent quite stabilised yet...

One question: Is it in any way possible to also calibrate the displayed value on the Console itself - i.e. for the console to display the calibrated value in stead of the raw sensor value

[SkidMark]

Hi Mark.

Thanks for your quick and elaborate resonse!

I was actually also thinking along the lines of an extension cable - really good to see that it can be done and how to go about it.

I'll be really interested in seeing your installation photo, particularly how you have dealt with the housing and aestaethichs. As it was supposed to be a very clean install on an otherwise clean wall - this is a challenge to say the least...

I'm somewhat disappointed that this is an issue though, and that it is not mentioned anywhere.

/Niels

Niels,

This with the battery door off. If you have it on the wall its not going to look very nice. Mine sits on a shelf and is behind the console.

Mark

[nia]

This with the battery door off. If you have it on the wall its not going to look very nice. Mine sits on a shelf and is behind the console.

Thank you for the effort of taking the photo. It would definitely require something built to hide it to make an external solution using the cable work

I'll roll with the new software update with calibration for the short term. Fortunately, it looks like exactly 5C is what needs to be subtracted, which should be easy to do at the console once getting used to it. If it gets too annoying, there is always the option of making a hole in the wall behind the station to put the unit in...

[admin]

When you use the Meteobridge NANO web interface this is just a few clicks and you are done. It does not need to repeated, as the offset is stored in the eeprom of the console.

[nia]

The calibration thing as stated works as a charm towards anything that uses the data...

One question: Is it in any way possible to also calibrate the displayed value on the Console itself - i.e. for the console to display the calibrated value in stead of the raw sensor value

I'm still wondering if it's possible, although I guess it's not...