Topic: SHT21 as alternative to SHT11

[dataworx]

I came across this sensor board and was considering whether it could be used as an alternative to the SHT11 in the older stations. I appreciate that there are a bunch of technicalities, so if anyone here has some info to share, I'd be most interested:

http://www.ebay.com/itm/SHT21-Digital-Humidity-And-Temperature-Sensor-Module-Replace-SHT11-SHT15-AM-/253076359731?epid=939419517&hash=item3aec86ca33:g:DMsAAOSwPK5ZgwVS

[azchrisf]

This was something I was thinking about earlier, if we can get a list together of sensors that can improve the Davis's accuracy, yet still be pin for pin compatible and work with the software, if possible, that we can replace.

[dataworx]

I've sourced one of the sensors linked to above - I'm quite happy to test it but I'm hoping that one of the more knowledgeable members will be along to help me with a pinout so I don't immediately wreck it by hooking it up incorrectly. I have a couple of stations here so I can easily do a side-by-side comparison to test accuracy.

[johnd]

I guess you need to explain why you might want to use SHT21 in place of the SHT31 that's readily available from Davis as a drop-in replacement for SHT11? OK, you need to apply the prescribed temperature offset to the console temp, but that's easily done. Maybe cost is an alternative factor, but the SHT31 board cost isn't exorbitant.

[dataworx]

I think there are a few factors, one of them being price ($6 versus $40-$50, which is a considerable difference). The primary reason though is that I'm an inveterate tinkerer. With Scottish blood. So if I can find a cheaper, effective solution I'm all for it, especially for stations that are a decade old and need to be kept in service.

[jerryg]

Well unless something has changed it won't work with the davis software it needs to be the ls series and i don't think they make it in that version. I tried one and it would not work, i believe it is i2c only.

[dataworx]

Thanks jerryg, I was fairly sure someone on the forum would have tested this sensor at some point. Perhaps something has changed - since I already have the sensor I'd like to test it anyway. Are you able to help me with the pinout?

[jerryg]

No i don't have the pin out anymore because it won't work with Davis station so saw no need to keep any info.

[dataworx]

Thanks jerryg

I have a dead Davis sensor here, and I have a schematic of the sensor itself, so I'll see if I can strip the protective coating and establish the pinout.

 [ You are not allowed to view attachments ]

[belfryboy]

As jerryg points out the sht21 uses i2c serial communication, but the Davis vantage pro 2 uses legacy sensibus communication. The two are not compatible. Although you probably won't damage your equipment your results will be disappointing. If you get the two data sheets side by side it should become clear.

[dataworx]

Thank you belfryboy, that's the sort of information I was looking for. With my rudimentary knowledge I've been unable to determine whether there would be a communications protocol issue. I have to assume then that nobody on this board has yet found a drop-in, non-Davis replacement, whether SHT11, 15, 21 or whatever?

EDIT: sorry to waste everyone's time, I just found this thread from last year, looks like this is an avenue that's been well explored:

https://www.wxforum.net/index.php?topic=29500.0

[dataworx]

Pardon the resurrection - I'm still digging around looking for a replacement to the Davis SHT11 chip/board. I have 7 of these original sensors here, all with varying faults.

Has anyone here yet come up with a suitable replacement at a sensible price?

[johnd]

The answer really hasn't changed in the past 6 years, for all the reasons previously stated. For a VP2 T/H sensor to operate correctly then it needs to use the Sensibus protocol. And the only known source of Sensibus-using Sensirion T/H sensors is Davis themselves.  To my knowledge, no-one has ever been able to find an alternative source, in part because - Davis-aside - Sensibus seems to be EOL.

Presumably one day Davis will need to redesign the ISS SIM board input to use eg a standard I2C input for the T/H sensor, but I can understand why they hold off doing so for as long as they can source the Sensibus part and can preserve full backwards compatibility for all VP2 stations with digital sensors back to 2006.