Best Battery-Friendly Outdoor Humidity Sensor for 80-100% RH: Alternatives to HS3001, BME280, Si7021

I have made an Outdoor weather station with a lot of sensors. All of them working well except for the humidity sensor. The weather in my country have a humidity between 80% and 100% during the winter season. The current sensor is HS3001, during the high humidity period it has settled at 99.9% without changing. 
I have also tried BME280 and Si7021, the latter with the best result. It seems that all these sensors having problem working continuously in this high humidity range.
Does anyone in this forum have knowledge about an other sensor that can be use for this purpose? An other requirement is that it has to be battery friendly.

I believe this a common problem. Heating the sensing element to keep it above the due point, which the Si7021 part does, is a partial solution, which may explain why it behaved better, assuming you put it through a heating cycle. Of course the heating cycle uses more power.All sensors will loose accuracy at the humidity limits to some extent, depending on environmental and sensing technology used. Additionally, living near the sea, or in industrial areas will also create further problems.There are companies that claim to have sensors that will cope better with your problem, but they are stand alone specialist pieces of equipment. So other than that, I can offer little in the way of support.Cheers,Richard

Yes, perhaps heating is the answer, but which algorithm should be used? I have tried the following heating procedure:  1. Wait until measured RH is above 80% for a time period (5 hours or so) 2. Start heater 3 Stop heater when the RH is below 50%  (measured with the sensor it self)4. Wait until the measured temperature has settled again.
But is it the right way to do it. I couldn't find any example or description on the internet, and the datasheet doesn't give much hint. Si7021 or HDC1080. The best description I can find is AN607 (si)

Most humidity sensors claim to work over 0-100%, but 20-80%  seems typically for the specified accuracy, so you are really fighting an uphill battle. You will have to play around with the algorithm, I can't say what will work. The time resolution will be poor, you are already talking about 5 hours with the algorithm you are using, what is acceptable?It is also worth looking at ST semis device, and in particular the app note, which has a lot of useful information. App note here. This note will apply to all of the sensors you have I would think.There is another paper here, it is 15 years old, but the methods for humidity measurement have not changed much, the last method remains the most accurate  over the full range.Also note this company has some interesting sensors, here, which maintain good accuracy over the full range.
Cheers,Richard

Get to us & get the most reliable equipment next to your door.