I’ve finally given up on Sensirion in general. I have the good fortune of having held on to my old analog temperature/humidity sensor from the pre2006 VP2 days (the 7346.029). At 14 years old, I wasn’t sure what kind of performance I’d get from it, although for most of that time it hasn’t been exposed to the extremes of outdoor conditions. I plugged it in and it worked and the temperature matched my other sensors. I placed a wet cloth over it and it read 100% humidity within 5 mins. I installed it as my primary sensor on 23Feb. We’ve had unusually dry weather in western Washington for this time of year (humidity down near 30%) the last 2 days. It has performed well.
Previously I was using the SHT75, which for humidity was performing better than the Davis 31, but with the continuous wet conditions, it still suffered from a pretty bad wet bias when things would dry out. I used an offset and calibration on the SHT75 of (1.061x)-5 in Cumulus. With the analog, I am using no calibration or offset. It’s really a shame Davis could not source these humidity elements any longer. I’ll see how it does over the long term, but if you’re interested to see how it compares, the nearest NWS ASOS is KTIW. KSEA and KBFI aren’t too far (see link). Everything prior to 23Feb was the SHT75 with calibration setting. Everything after is the old analog.
https://weather.gladstonefamily.net/qchart/C7491?date=20190227&addnl=KTIW&addnl=KSEA&addnl=KBFI
05Feb really gives a good demonstration of the wet bias after a long wet period with the Sensirion:
https://weather.gladstonefamily.net/qchart/C7491?date=20190206&addnl=KTIW&addnl=KSEA&addnl=KBFI
Man I wish I had that same analog sensor. Looking at your data and comparing it to the ASOS tells me it’s a great sensor. Even though the 75 is better than the 31, I’m still having to do an offset of -3-4%. Even though the sensor has “dried” out for a few days. I wish there was a way to obtain those old temp/hum sensors. If anyone knows the makers of those sensors, please let me know.
I realize information on these old analog sensors is largely irrelevant, but I thought I’d provide an update. After being gone for several weeks, I’ve been back for a few days and combined with what I was seeing remotely and after numerous spot checks with my psycho-dyne, I found that the analog sensor was about 3-4% too dry. The temperature has been very accurate, but in a temperate range from about 43-75F.
The good news about the dry humidity sensor is that it read 3-4% low across the entire scale and topped out at 97% (although I have been able to force it with a warm damp cloth to 100%). I’ve applied a +4% humidity offset, which allows the sensor to easily hit 100% and it has been +/-2% of my psychro-dyne with humidity all the way down into the upper teens (which we just experienced here recently). It recovers from high humidity much quicker than the Sensirion. Sometimes the sensor reads a little wet, other times a little dry. It’s nice to see this random error versus the constant systemic high humidity bias I was used to.
With that said, I did make a side by side comparison with the SHT75 with humidity down in the 20s. They read almost identical (with the +4% offset on the analog) and right on with the psycho-dyne, but with the caveat that the SHT75 had been indoors in a very dry environment and not exposed to the prolonged high humidity that seems to be that sensor’s weakness.
Compared to the nearest ASOS (KTIW), the analog compares favorably. Local effects here can be extreme in warm, dry conditions with proximity to water, wind direction and down and upsloping winds. My psycho-dyne always matches ASOS nearly exactly when I’ve done side by side comparisons, so as a sort of transfer standard, I trust it more than comparing to an ASOS with unique differences over 4 miles away