Is DHT11 useful? I think not!

[hmmpic]

I have just got some (5pcs.) DHT11.

All measure about 25-32% humidity but there are about 58% in the real world?
The temperature show about +2, that is ok.

The DHT11 sensor is completely useless, but cheap ...

Is there something that works, and costs less than 1$/1000pcs.

[asmboy]

i don't know your code , circuit or methodology. so no way to tell if your basic assumption/conclusion is even true.

but i am an ardent believer in this:

you get what you pay for - or slightly less

[temtronic]

Actually have a 'silly' application for them so I downloded the datasheet....
Specs are pretty good when you consider the price. One thing does stand out is to 10 seconds per reading, which may be causing you to misread ?
Without seeing your code or envirinmental containr, it's hard to guess where your problem is.
There is a LOT of hits from Google about it being OK though...

Jay

[hmmpic]

Yes i have been on google, and lot other have low accurate too. Maybe my ebay sample is not the real stuff, but some china clone one?
The code is ok, and hardware too. I know you doubt.
The problem can be the read speed I reread every 8 second. In my datasheet there are nothing about it. Or maybe it is the 2 minutes they talk about as maximum read cycle? It is china datasheet real bad.
Do you have a link to your datasheet? I have my from "Aosong".

[PCM programmer]

This website compares several different humdity sensor chips:
http://www.kandrsmith.org/RJS/Misc/hygrometers.html
It has 8 links in the bulleted list at the bottom, with detailed information.

[Ttelmah]

Thing you have to be very careful about is how you place the sensor. It is not something that you can put in the middle of a board with other electronics. This applies to all humidity sensors. You are measuring RH, and this depends massively on temperature. If (for instance) the electronics on your board are raising the temperature round the sensor by 5 degrees, and your other sensors are running at room temperature (say 20C), then the sensor will only read about 34. The amount of actual water in the air is the same, but with higher temperature the amount of water the air can hold shoots up, so the relative humidity falls.

You need to have the sensor out in a nicely vented pod well away from power supplies, chips etc..

The graphs here:
<http://www.engineeringtoolbox.com/relative-humidity-air-d_687.html>

Give the relationship to temperature.

[hmmpic]

Hi

Thanks for all the nice reply. I have solved the problem and the solution is surprising me.

After I have manually checked the data plan on a logic analyzer and may code and the environment the sensor is mounted in, and the data sheet for to check RH/C. I was blank.

Then I posted a picture to "Aosong" asking for them to verify that the sensor is the real stuff.

And my sensor and close to all from ebay is fake sensors, they work but real bad. Do not use sensor with label only with laser text!

***

Here are what Aosong wrote:
I am sorry this product is not manufactured by us. This may be from some local supplier copy us. They offer low price and with
very bad quality.

All our product are with Laser tag.

[Ttelmah]

There is a remarkable number of such problems around.
SD cards.
PICs.
Lots of chips in general.

They are obviously basically the same, but presumably lack the calibration. I'd suspect they might even be manufacturing rejects, scavenged and badged (this was happening with PICs a couple of years ago...).

[hmmpic]

It is very annoying. I have now ordered directly from the factory.
Thanks for all the suggestion and help.

[newguy]

The guy I replaced ordered qty 2000 Si9979 3 phase BLDC motor control ICs from a shady broker. Upon power-up, they'd all smoke, then eventually they'd actually blow up.

I've been able to get authentic chips (as recently as December), but I've used reputable brokers.

This same guy also got a shipment of counterfeit 68376 processors from the same vendor. ...Think he would have learned.