[werk_ag]

Hello Zdenek

Just a note that perhaps you may use to update your spreadsheet: The SHT21 and the HTU21 aren't equivalent sensors. They have almost the same specs, both use I2C to communicate, but each one requires it's own code to be read.

Regards
Werk_AG

[hornychz]

Hello Zdenek
Just a note that perhaps you may use to update your spreadsheet: The SHT21 and the HTU21 aren't equivalent sensors. They have almost the same specs, both use I2C to communicate, but each one requires it's own code to be read.
Regards
Werk_AG

Hi Werk_AG,

And thank you for your note - now I've already corrected it. (That I really did not know.) Thank you again.

Regards

[AllyCat]

Hi,

Yes that's a very interesting and useful spreadsheet, but I think the "Protocol" column might be a little misleading. Of course it's a complex issue and some sensors may use (or be available in) more than one format. In particular, I believe (from the Sandaysoft forum) that Davis use a "legacy" version of the SRT31. Furthermore, a simple address discrepancy may prevent a sensor being recognised, or a particular "library" might use the bus in an "unexpected" way.

"One Wire" is normally considered as a very specific protocol developed by Dallas/Maxim to carry bi-directional data and (optionally) power on a single wire. Actually of course, a second "Earth Return" wire is normally required to complete the circuit. It uses ingenious (and complex) "arbitration" methods to allow multiple devices (sensors) to be identified and read/controlled over the single pin/conductor.

Some "dumb" sensors may periodically "spew out" a stream of serial data on a single pin/conductor, but this would not normally be defined as "One Wire". Similarly, the second group of three sensors in your table appear to have "Data" and "Clock" pins and thus are unlikely to be "One Wire". A very common data interface is called "SPI" (Serial Peripheral Interface) but this usually has separate "Clock", "Data Out", and "Data In" (sometimes an "Enable" pin as well). So I'm not sure what the "Clock" plus bidirectional "Data" line used by Sensiron might be called.

Similarly, the I2C bus (developed by Philips) would normally be identified by the specific pin names SDA (Serial DAta) and SCL (Serial CLock). But a few manufacturers do use the I2C protocol but choose to "hide" its origin, due to possible trademark/copyright issues, or simply "pride" (the Philips patent will have expired many years ago). So I'm also "suspicious" about the protocol name used for the SHT10-15 (also one is marked STH) but I haven't checked their data sheets.

Cheers, Alan.

[hornychz]

Hi Alan,

certainly you're right - the table can only help as the initial view when deciding what sensor to use, what is its sensitivity, accuracy, stability, and also the approximate price. But I consider very important for my remark about "insight into the datasheet."

(E.g. accuracy is usually dependent on the size/range of the measured values ​​and often power. The table is a simplification for rapid orientation in the field.)

I also know that the aforementioned protocols 1Wire and I2C are originally defined differently, but so are the ways to connect identified in those datasheets and especially as to distinguish between two basic ways to connect ...

Spreadsheet I initially prepared for themselves, sensing, because I am not proficient, their parameters can not remember, and I still did not have to just look. But now I thought it might just as well serve someone else, so I will have transliterated into English.

Error in labeling (SHT -> STH) I corrected, thank you for your note.

Regards

[werk_ag]

Hi Zdenek

Maybe you also want to change the protocol name for the SHT10, SHT11 and SHT15 sensors. Those sensors don't use the I2C protocol (SDA, SCL).

Regards

[hornychz]

Maybe you also want to change the protocol name for the SHT10, SHT11 and SHT15 sensors. Those sensors don't use the I2C protocol (SDA, SCL).

Hi Werk_AG,

Thank you for comment, I've already corrected it ... Maybe it would be better so ...

From:

Embedded Lab



Regards

[werk_ag]

Surely!

[JT118]

Accuracy & Precision; I have various SHT15 and SHT21 sensors running. All are housed in a single radiation shield about 10mm apart. They seldom show the same though the difference is fairly constant suggesting precision is OK but accuracy may be suspect. Only calibration can indicate which is accurate.

I have an SHT31 which I must get working in next few days.

   

[Zappie]

For my TX i purchased these sht21 sensor. It has a stainless steel housing.



Extended the cable to 3 meters and still working fine.


2 weeks ago i ordered another one with 5 meters of cable in a different housing. Think a plastic housing will do better in a radiation shield and i had to remove a part of the pcb sealing.



Greetz, Zappie

[werk_ag]

Hi Zappie,

Thanks for that info.
Where do you found the last one with the plastic cover? I liked it.
Have you already tested the version with 5 meters of cable?