Topic: SHT31

[Jester]

Quick question for those who operate a VP2+ and have an SHT31 installed. Do any of your stations ever hit 100% humidity. Best mine has done is 97% Right now both ASOS sta. in my area, one just over a mile away as the crow fly's are reporting 100% and I'm 94%. I'm running a 12VDC fan at 9VDC in my FARS. My SHT15 hit 100% from time to time.

[mcrossley]

Very occasionally, yes.

[drew1021]

Yep, same here. I have been following my station with the ASOS station in town and have observed the exact same results as you. One thing to keep in mind is the accuracy of the ASOS sensors, they aren't perfect. The best way to check your own accuracy is to wait until dark and get a known accurate psychrometer and whirl away. I own a Belford aspirated psychrometer accurate to +/- 0.3 degrees Fahrenheit. Mine did fall to within 2% on the low side, 96%, during light fog. At the same time the airport was at 100%. On a bright note the other day we had humidity levels between 5 and 10%. My station was within 1% of the airport ASOS. I believe ther is some variability between Sensiron sensors.

[Jester]

Yep, same here. I have been following my station with the ASOS station in town and have observed the exact same results as you. One thing to keep in mind is the accuracy of the ASOS sensors, they aren't perfect. The best way to check your own accuracy is to wait until dark and get a known accurate psychrometer and whirl away. I own a Belford aspirated psychrometer accurate to +/- 0.3 degrees Fahrenheit. Mine did fall to within 2% on the low side, 96%, during light fog. At the same time the airport was at 100%. On a bright note the other day we had humidity levels between 5 and 10%. My station was within 1% of the airport ASOS. I believe ther is some variability between Sensiron sensors.

Might have to try that...thanks

[ValentineWeather]

I have 4 SHT-31's at 2 different locations and very rarely hit 100%. One time I did on 1 of the 4 sensors. 

There are some 3rd party softwares like Cumulus that have offsets for calibration. How it works to get say 100% you would subtract a few % and then use multiplier

Example 100-2=98 x 1.036= 101.52...Doing this when sensor reaches actual 98 cumulus will display 100%.  Mid scale at 50-2=48 x 1.036= 49.7 so its still showing actual 50% when rounded off.

[CW2274]

About a month ago it was WOXOF (goggle it) which is SUPER rare here, and my 31 refused to leave 99%, but all the ASOS's and RAWS around me were 100%, something I've never seen here before. I wanted to join the party as being the lone 99% guy on the NWS web site was bumming me out, so I bumped up the humidity 1% for sh!ts and grins. It stayed at 100% for the next 6 hours, til 10 in the morning when the fog finally burned off. Granted, I had to fudge it 1%, but it was fun seeing everyone of us at 100%.

[drew1021]

About a month ago it was WOXOF (goggle it) which is SUPER rare here, and my 31 refused to leave 99%, but all the ASOS's and RAWS around me were 100%, something I've never seen here before. I wanted to join the party as being the lone 99% guy on the NWS web site was bumming me out, so I bumped up the humidity 1% for sh!ts and grins. It stayed at 100% for the next 6 hours, til 10 in the morning when the fog finally burned off. Granted, I had to fudge it 1%, but it was fun seeing everyone of us at 100%.

Problem with putting in an offset of +1% is that it will only make readings at the lower end of the scale that much higher. Personally I think the ASOS sensors are too quick to saturate anyway.

[CW2274]

About a month ago it was WOXOF (goggle it) which is SUPER rare here, and my 31 refused to leave 99%, but all the ASOS's and RAWS around me were 100%, something I've never seen here before. I wanted to join the party as being the lone 99% guy on the NWS web site was bumming me out, so I bumped up the humidity 1% for sh!ts and grins. It stayed at 100% for the next 6 hours, til 10 in the morning when the fog finally burned off. Granted, I had to fudge it 1%, but it was fun seeing everyone of us at 100%.

Problem with putting in an offset of +1% is that it will only make readings at the lower end of the scale that much higher.

Of course it will. Soon as it fell below 100%, I took the offset out and it remains out. Like I said, I did it for the fun of seeing us all at 100%, that's all.

[CW2274]

About a month ago it was WOXOF (goggle it) which is SUPER rare here, and my 31 refused to leave 99%, but all the ASOS's and RAWS around me were 100%, something I've never seen here before. I wanted to join the party as being the lone 99% guy on the NWS web site was bumming me out, so I bumped up the humidity 1% for sh!ts and grins. It stayed at 100% for the next 6 hours, til 10 in the morning when the fog finally burned off. Granted, I had to fudge it 1%, but it was fun seeing everyone of us at 100%.

Personally I think the ASOS sensors are too quick to saturate anyway.

Maybe, but certainly not in this scenario.

[ValentineWeather]

About a month ago it was WOXOF (goggle it) which is SUPER rare here, and my 31 refused to leave 99%, but all the ASOS's and RAWS around me were 100%, something I've never seen here before. I wanted to join the party as being the lone 99% guy on the NWS web site was bumming me out, so I bumped up the humidity 1% for sh!ts and grins. It stayed at 100% for the next 6 hours, til 10 in the morning when the fog finally burned off. Granted, I had to fudge it 1%, but it was fun seeing everyone of us at 100%.

Problem with putting in an offset of +1% is that it will only make readings at the lower end of the scale that much higher. Personally I think the ASOS sensors are too quick to saturate anyway.

What happens with the small amount of offset used it makes readings at bottom slightly lower, not higher. It actually helps out because they tend to run slightly high down on bottom, at least my experience. Kills 2 birds with one stone you might say.

Take 30%-2=28 x 1.036=29%

[CW2274]

About a month ago it was WOXOF (goggle it) which is SUPER rare here, and my 31 refused to leave 99%, but all the ASOS's and RAWS around me were 100%, something I've never seen here before. I wanted to join the party as being the lone 99% guy on the NWS web site was bumming me out, so I bumped up the humidity 1% for sh!ts and grins. It stayed at 100% for the next 6 hours, til 10 in the morning when the fog finally burned off. Granted, I had to fudge it 1%, but it was fun seeing everyone of us at 100%.

Problem with putting in an offset of +1% is that it will only make readings at the lower end of the scale that much higher. Personally I think the ASOS sensors are too quick to saturate anyway.

What happens with the small amount of offset used it makes readings at bottom slightly lower, not higher. It actually helps out because they tend to run slightly high down on bottom, at least my experience. Kills 2 birds with one stone you might say.

Take 30%-2=28 x 1.036=29%

I actually considered leaving the +1% in because when we drop down to the 1-2% humidity range, the 31 gets flagged on the dewpoint for being too low, but I'm still at zero correction as a choice. Between 1-2% and 99%, the 31 is money, at least mine is that way.
Be nice to know how a 35 would fair......

[ValentineWeather]

About a month ago it was WOXOF (goggle it) which is SUPER rare here, and my 31 refused to leave 99%, but all the ASOS's and RAWS around me were 100%, something I've never seen here before. I wanted to join the party as being the lone 99% guy on the NWS web site was bumming me out, so I bumped up the humidity 1% for sh!ts and grins. It stayed at 100% for the next 6 hours, til 10 in the morning when the fog finally burned off. Granted, I had to fudge it 1%, but it was fun seeing everyone of us at 100%.

Problem with putting in an offset of +1% is that it will only make readings at the lower end of the scale that much higher. Personally I think the ASOS sensors are too quick to saturate anyway.

What happens with the small amount of offset used it makes readings at bottom slightly lower, not higher. It actually helps out because they tend to run slightly high down on bottom, at least my experience. Kills 2 birds with one stone you might say.

Take 30%-2=28 x 1.036=29%

I actually considered leaving the +1% in because when we drop down to the 1-2% humidity range, the 31 gets flagged on the dewpoint for being too low, but I'm still at zero correction as a choice. Between 1-2% and 99%, the 31 is money, at least mine is that way.
Be nice to know how a 35 would fair......

  You've got a different problem...1-2% doesn't happen here at least so far. I have seen 10-12% however.

[drew1021]

Have any of you that are using the SHT31 noticed more fluctuations of the dew point  values when looking at a graph?
I have my sensor in the 24 hour FARS

[ValentineWeather]

Only above about 90°F my SHT31's become erratic with dew point making wild swings. All of them act the same some are worse than others.

 [ You are not allowed to view attachments ]

[drew1021]

Only above about 90°F my SHT31's become erratic with dew point making wild swings. All of them act the same some are worse than others.

 [ You are not allowed to view attachments ]

I wonder if its due to the fact that the sht31 sensor is more sensitive than previous versions of the sensiron sensors? I thought the purpose of the fars was to smooth out fluctuations in temp/dew point.

[ValentineWeather]

Not sure, it really messes with my data like heat index when it spikes like that. 

[CW2274]

Only above about 90°F my SHT31's become erratic with dew point making wild swings. All of them act the same some are worse than others.

 [ You are not allowed to view attachments ]

I thought the purpose of the fars was to smooth out fluctuations in temp/dew point.

Not the way I see it. I think the fan brings more fluctuations than stagnant air in the chamber. That being said, my dewpoint is rock solid (except when I'm down in the 1% humidity range), the temp, that's a different story, especially in the summer when the heat really gets moving across the desert floor.

[Old Tele man]

Hot summer desert air = eddy-current 'bubbles' of hotter rising air being filled in behind by cooler air.

[CW2274]

Hot summer desert air = eddy-current bubbles of hotter rising air being filled in behind by cooler air.

And being dry makes it even more pronounced.

[WxLover16]

Hot summer desert air = eddy-current bubbles of hotter rising air being filled in behind by cooler air.

And being dry makes it even more pronounced.

But you consider it to still be very accurate, right? since we all know the 31 has a faster response time than previous versions.

[CW2274]

About a month ago it was WOXOF (goggle it) which is SUPER rare here, and my 31 refused to leave 99%, but all the ASOS's and RAWS around me were 100%, something I've never seen here before. I wanted to join the party as being the lone 99% guy on the NWS web site was bumming me out, so I bumped up the humidity 1% for sh!ts and grins. It stayed at 100% for the next 6 hours, til 10 in the morning when the fog finally burned off. Granted, I had to fudge it 1%, but it was fun seeing everyone of us at 100%.

Problem with putting in an offset of +1% is that it will only make readings at the lower end of the scale that much higher. Personally I think the ASOS sensors are too quick to saturate anyway.

What happens with the small amount of offset used it makes readings at bottom slightly lower, not higher. It actually helps out because they tend to run slightly high down on bottom, at least my experience. Kills 2 birds with one stone you might say.

Take 30%-2=28 x 1.036=29%

Between 1-2% and 99%, the 31 is money, at least mine is that way.

Requote from my post above.

[WxLover16]

About a month ago it was WOXOF (goggle it) which is SUPER rare here, and my 31 refused to leave 99%, but all the ASOS's and RAWS around me were 100%, something I've never seen here before. I wanted to join the party as being the lone 99% guy on the NWS web site was bumming me out, so I bumped up the humidity 1% for sh!ts and grins. It stayed at 100% for the next 6 hours, til 10 in the morning when the fog finally burned off. Granted, I had to fudge it 1%, but it was fun seeing everyone of us at 100%.

Problem with putting in an offset of +1% is that it will only make readings at the lower end of the scale that much higher. Personally I think the ASOS sensors are too quick to saturate anyway.

What happens with the small amount of offset used it makes readings at bottom slightly lower, not higher. It actually helps out because they tend to run slightly high down on bottom, at least my experience. Kills 2 birds with one stone you might say.

Take 30%-2=28 x 1.036=29%

Between 1-2% and 99%, the 31 is money, at least mine is that way.

Requote from my post above.

I was more or less referring to the temperature. I know you have a ton of huge spikes in summer cause of heat eddies on the hot dessert floor, which the spikes would be due to the 31's fast response time. Just was wondering if you consider those to be accurate.

[CW2274]

Oh. My reply had the word "dry" in it so I thought accordingly. Anyway, yes, I consider it accurate, the sensor's merely doing it's job. It just yo-yo's a lot, the big fan probably exacerbating it as I can see swings in temp a good 3-4F in minutes.

[WxLover16]

Oh. My reply had the word "dry" in it so I thought accordingly. Anyway, yes, I consider it accurate, the sensor's merely doing it's job. It just yo-yo's a lot, the big fan probably exacerbating it as I can see swings in temp a good 3-4F in minutes.

The only way that would drive me crazy is if it were just the lack of a good radiation shield and the sun making for erroneous readings, thus driving the temp up and down. But like you said, the fact that it's because you have a highly sensitive accurate sensor with a fan added in to make readings even more accurate, would make it quite interesting to watch since you know it's registering the upmost in accuracy. Now, I'm just waiting on a 35 to get even more accuracy if that's even possible.

[CW2274]

With that being said, if I want an average representation of the temp, I need to study it for a few minutes (in other words, look at WL, browse station data; or stare a lot). No averaging algorithm for us, what you see is what it is, at that moment.