ESP8266 Community Forum

- Fri Jun 26, 2015 5:00 am #21740 I've got a DHT22 running at https://data.sparkfun.com/streams/RMzp9ANyqzfjw4drzNnZ and at thingspeak https://thingspeak.com/channels/38247 so am watching with interest . My DHT is set at 15 and the Wire.setClock(400000);

It does not give much trouble until corrosion steps in as its outside and we are in rainy season and quite wet . A bit of INOX on the breadboard I have it stuck in and away it goes again or a bit of a wiggle. The BMP180 gives no trouble whatever .

I'd like to see what the circuit you end up with looks like if you can . Must be near Gatton with those temps but then its cold everywhere right now I think . Cheers

Get you modules here:

https://www.banggood.com/custlink/KvGGhGF4wG

- Fri Jun 26, 2015 11:34 am #21777

Then I discovered the readvdd33() system call .. why did nobody mention this ??

Actually readvdd33() is an internal symbol someone found in an old SDK. Since it was not intended to be used by end users, it might break or disappear without warning in newer releases. Recent SDK versions expose the following function:
uint16 system_get_vdd33(void);
See SDK documentation for usage.

- Fri Jun 26, 2015 12:02 pm #21780

I've got a DHT22 running at https://data.sparkfun.com/streams/RMzp9ANyqzfjw4drzNnZ and at thingspeak https://thingspeak.com/channels/38247 so am watching with interest . My DHT is set at 15 and the Wire.setClock(400000);

It does not give much trouble until corrosion steps in as its outside and we are in rainy season and quite wet . A bit of INOX on the breadboard I have it stuck in and away it goes again or a bit of a wiggle. The BMP180 gives no trouble whatever .

I'd like to see what the circuit you end up with looks like if you can . Must be near Gatton with those temps but then its cold everywhere right now I think . Cheers

i haven't tried the setClock with these guys yet, I had a look with a little saleae logic analyser and really all the esp sends is a single low pulse to start the sequence, the DHT then pauses for about 20mS while it does the conversion then spits out 40 data bits. short for 0 and long for 1.. so I guess the 15 or so is the threshold value and is also related to system clock.
Must have a look at the code ..

I am in horse country down the bottom of tambourine halfway out to canungra. So yeh .. a few degrees cooler than the coast especially on a clear night.

circuit so far is as above with the addition of a supercap across the supply leads and R1 and R2 removed.

I was bench testing without sending updates and calibrated the readvdd by subtracting 160mV from the mV reading.
As soon as I finished, I reenabled all the wifi side and stuck it back outside in its box .. now its reading 3.9V again ..
sigh

though I am taking the reading straight after a 1s boot delay (helps to reprogram it easily) I guess its auto connecting is screwing up the vdd reading.