low power design and wifi

On Sun, 17 Nov 2013 14:11:23 +0100, en2 Gave us:

I saw this but if i connect some USB wifi adapter i will end up in the same situation as before or maybe worse. I dont know, maybe am wrong...

On Sun, 17 Nov 2013 16:20:26 +0100, en2 Gave us:

Two models have built in wifi and bt

See my other post announcing that I bought one for links.

Only obvious thing I can add is high gain directional antennas to keep the Wifi power down.

Mikek

On Sun, 17 Nov 2013 09:28:49 -0600, amdx Gave us:

I wish there were a way to get two spdif jacks on them, and then use them as comm links.

Then, one could daisy chain a bunch of them together and make a nice emission free cluster.

But you could can the cluster and the router and use less power in the links. But that is my cluster, not his scenario.

------------------------------

So, the model with build in wifi costs 60$ without VAT and shipping. End prize would be around $90 and that's quite expensive.

Raspberry has lower prize but 5W consumption so this is also out.

I have to use external antenna and it can be directional one so i have to have some external jack to connect this antenna to.

It looks like that only option iw some built in wifi enabled device, but i dont know if this can fit into budget range.

What about some ultra low power WIFI module ? Maybe i can use this module with external antenna to get this 50 meter range...

It's a very common wish list (or wishful thinking), transmitting a few data bytes over WiFi with battery device. But WiFi was not designed for low po wer. The connect/disconnect process is complicated and power consuming. Th e reasoning is that you save on the internet gateway, since WiFi routers ar e everywhere. Very often, the project get killed by the battery.

We use Asymmetric Network Transmission (ANT). The portable device TX at 30 mA, but the gateway TX at 100mA. This is for a range of around 20 meters. For 50 meters, you probably need 200mA from the gateway. In between packe ts, the transmitters are off to save power.

It's a very common wish list (or wishful thinking), transmitting a few data bytes over WiFi with battery device.

But WiFi was not designed for low power. The connect/disconnect process is complicated and power consuming.

The reasoning is that you save on the internet gateway, since WiFi routers are everywhere. Very often, the project get killed by the battery.

We use Asymmetric Network Transmission (ANT). The portable device TX at

This is for a range of around 20 meters. For 50 meters, you probably need 200mA from the gateway. In between packets, the transmitters are off to save power.

-------------- signing every word :-)

I would "stop" transmission for half minute till i get full buffer.

ANT:

- can you share some link ?

What about other possibilities. Some ISM FSK band transceiver.

So device would be some kind of basic ISM tranciever and it would not have some big power consumption. On the other side, 50 meters away, i can set some serial to IP converter. I know it's long shoot but..

I think you are missing the point.

No matter how much data you send, you will need to connect to the router every time.

The power use just to connect may be higher then the power to send the data.

The smaller the data packet the worse your power usage.

There are lots of these type of devices. TI has many to chose from, pick one and start from there.

Good Luck

PS: If you can find one of these on ebay:

----------------------------- I agree with you and that's why i want to use buffer and send data when i must. I was looking at CC300 but the problem is that it uses 275mA on TX line and this can not work on battery powered system.

Interesting, the last product I did that had built-in Wifi: We could only obtain compliance *IF* we tested the system WITH the proposed fixed-connection antenna and if we needed to make a disconnect for the antenna [which might inadvertantly enable a user to 'modify' our system], we had to use a 'strange' RF connector, like one with left-handed threads or such. Supposed to be unavailable to 'normal' consumers.

Has compliance changed?

It could take minutes to reconnect WiFi. You also have to enter the passcode in the device somehow.

Look up 802.15.4 without ZigBee. Ant is not Bee. Plus you have to actively manage power budget for the Radio and Processor. There are no magic module to do it for you.

On Sun, 17 Nov 2013 13:31:32 -0700, RobertMacy Gave us:

These tiny boards have little tiny coax connectors on them.

On Sun, 17 Nov 2013 12:40:21 -0800 (PST), snipped-for-privacy@gmail.com Gave us:

Easily automated. And safely so as well.

Power consumption is influenced by the WAP too. It needs to have reasonable beacon and DTIM intervals, and enough buffering to not drop packets while your radio is sleeping.

-------------------- What about : RTX4100

If it's that little 'button' type, I think I have the part number for that connector, ...somewhere around here.

Doesn't this product have to meet compliance? Be tested as a system? It is true that buying a module and sticking it into a system 'sometimes' gains you automatic compliance meeting "intentional radiator" specs, but not often. You still need to meet Class B "unintentional radiator" limits and test for that. If you're lucky [or did it perfectly] with the addition of the Wifi module; you won't have to test for the 'extra' radiation part, which can easily be another $3k in test lab time. [from memory]

Check to make certain you comply with EVERY restriction listed in the modules registration. If ONE item varies, you start over. Get each modules documentation, that may also help in making your selection.

Footnote, if you place two like Wifi and zigbee in the same product you'll probably end up with the two antennas closer than allowed in their registration and you have to do the WHOLE testing series for compliance. However, the FCC made a mistake on one of the zigbee modules and lists NO restrictions on it being near another antenna. - major oversight! ah, but the other one catches you.

Page 19 of:

TX: 300mA or 1W at 3.3V for 100 seconds (minimum to reconnect and send)

vs.

Page 142 of

TX: 23mA or 80mW at 3.3V for 1 second (minimum to send)

----------------- wow this is huge difference ! Let me check the whole datasheet