Wireless systems

Go To Last Post
13 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi there,

I would like to build a circuit supplied by a non-rechargeable battery 3V lithium / 350mAh. This circuit must be able to be connected to devices/sensors been connected from the same battery.
All the devices comunicate with the master and maybe some of them or the most percentage can't comunicate with master directly, so for this reason each device must play the role of a trasceiver and a live broadcaster simultaneously.
The whole product wireless and sensor must consumpt not more than 20uA. From the sensor's side there is no problem because it is easy to have < 5uA. The sensor is something that exists and its micro have the appropriate pins to be connected with a USART, SPI or I2C.
Also any time there is a new event in the network because of a sensor or button or anything else, hen the master must know the event as soon as possible (in 10sec max).
I took a look at the ZigBee but as it seems the consupmtion is more than 2mA. I also found an article that compares the ZigBee with another protocol called ANT (from www.nordicsemi.com). Reading I found that ANT consumpts ~50uA rms (~which is a little bit more than my specs, but ok) but lower wireless comunication range.

So guys, if you would be able, I would like to know your ideas and knowings about this issue.
My musts are:
1. low rms current consumption (<50uA and lower if it is possible)
2. Distance 30 - 60 meters
3. Fast receiver enable
4. Wake up on receive

Ok, I know that the ZigBee has open stuck, so it is easy to buid your applycation without wondering about the whole comunication protocol, but unfortunately consumpts enough current.

What can I do???

Michael

Michael.

User of:
IAR Embedded Workbench C/C++ Compiler
Altium Designer

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The crux of your problem is the <150uW power and 30-60 meter range. Look into atomic clock receivers. They receive a seriously faint signal. If your receiver is good enough to pull out the signal from the noise then the transmitter can be very low power. The rest of your problem is all software. I have no idea what environment you plan on using this device but take that into consideration as well.

I have no idea why rechargeable batteries are being excluded since there is little difference between a non-rechargeable and rechargeable battery. I don't think your circuit could tell the difference at least.

Go electric!
Happy electric car owner / builder

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

sgomes wrote:

I have no idea why rechargeable batteries are being excluded since there is little difference between a non-rechargeable and rechargeable battery. I don't think your circuit could tell the difference at least.

You know rechargeable batteries need a charger so an external power supply. Actually the reason of using batteries is not for working when the external supply fails, but for functioning without the need of any external supply. Also Non-rechargeable batteries have more energy density in the same package and also less leckage current, so less energy loss as the time passed.

Michael.

Michael.

User of:
IAR Embedded Workbench C/C++ Compiler
Altium Designer

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Micheal -

I think you misunderstand how ZigBee works. It normally sleeps with very low current and "wakes up" to test whether or not a signal is being received. The AVERAGE current is or can be very low.

For the simple sort of thing you are talking about, look at XBee. It does not have the mesh networking but you can implement your own simple network if you choose. It does have a master/slave mode as well as peer-to-peer.

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

icarus1 wrote:
sgomes wrote:

I have no idea why rechargeable batteries are being excluded since there is little difference between a non-rechargeable and rechargeable battery. I don't think your circuit could tell the difference at least.

You know rechargeable batteries need a charger so an external power supply.

No I didn't know they need an external power supply! Wow. I guess you learn something new everyday. So inside all my kids toys are little power supplies that magically make the rechargeable batteries work?

Quote:
Also Non-rechargeable batteries have more energy density in the same package and also less leckage current, so less energy loss as the time passed.

Ah but you see you did not include that spec in your OP! So how long does this device have to operate? And why is density an issue? Do these devices need to be small? This could be a problem if you want to pick up tiny signals.

Go electric!
Happy electric car owner / builder

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

For a small number of nodes or point to point, don't use ZigBee. Just use the 802.15.4 MAC layer and instead of ZigBee, use wireless serial port extension software from most vendors like Digi/Maxstream. At PCs, you can connect a '15.4 radio to USB or serial on the PC and use a virtual serial port driver (free from vendor) so the PC sees COMn: .

ZigBee is the OPTIONAL complex network routing layer software that uses 802.15.4 as the wireless media.

The '15.4 radio can sleep 90% of the time, like wake up several times a second or minute to check for pending traffic.

many of the '15.4 radio modules consume only 22mA @ 3.3V with the receiver on. Don't start with chips, start with plug and play modules.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

XBee is a lot easier than ZigBee. It is also 802.15.4 based. Mostly agree with stevech.

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ka7ehk wrote:

I think you misunderstand how ZigBee works. It normally sleeps with very low current and "wakes up" to test whether or not a signal is being received. The AVERAGE current is or can be very low.

Jim

So Jim, reading your reply I understand that a ZigBee receiver may be at sleep mode (very low consumption) and each time it wakes, then proceed it's stored received data to a micro. But how this is possible. How the receiver functions and consumpts nothing while the control is in sleep mode???

I am trying to find which is the best way for my project. I have some experience in wireless systems, but nothing on this issue called ZigBee, XBee or anything else.

I need to have a network using up to 1000 slaves and a master. Each slaves current consumption must be as less as possible, because they will be supllied from a battery each one. The master will be supplied by external power supply, so there will be no low consuption issue for this.
The master will poll the network for check if all slaves are present. Of course there will be slaves out of the master's comunication range, so they must receive the master's polling via other slaves that are near them. I think that this is one of the ZigBee characteristics.
Each time there is an allarm or a fault in the network the master must know this in a time < 10sec.

What is the best thing for my applycation.

Thanks,

Michael.

Michael.

User of:
IAR Embedded Workbench C/C++ Compiler
Altium Designer

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ZigBee (and XBee) devices can handle a master with 65,000 slaves.

The receiver can wake up for a few milliseconds and go to sleep for 10s to 100s of milliseconds. It CAN detect the presence of a signal and wake up a companion micro.

The power consumption is low enough that it can last "6 months to 2 years" on a set of batteries, depending on the message rate.

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Jim,

I hope that you are right and I believe that you are, because where is the use of a wireless protocol. The need is to have low consumption because the more percentage of wireless systems is to work without the need of external supply.
I red an article in a Europian magazine (EETIMES) that compares the ZigBee with another one called ANT (from www.nordicsemi.com). Tommorow I will scan the pages and I will post the article. As I see in the article the ZigBee has enough current consumption ~2mA(rms).
MAybe I am a little bit confused because I am trying to find the right method or protocol to use before I begin with the design.

Very thanks for your time,

I will post the article,

Michael

Michael.

User of:
IAR Embedded Workbench C/C++ Compiler
Altium Designer

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

802.15.4 systems don't work by keeping the receiver enabled all the
time. If it's just a sensor, why do you need the receiver at
all (except for receiving the acknowledgment shortly after the
transmission, of course)? After all, sensors tend to collect data,
and then send them off at whatever interval to some kind of hub (or
coordinator). In that case, only the hub has to keep its receiver
enabled, and is then typically some device with a stronger power
supply behind.

If that's not applicable to your scenario, the trick is to find a good
scheme about how to synchronize the device's wakeups so it won't have
to keep the receiver going for too long (thus wasting energy), yet still
not miss the transmissions from whomever is going to send something to
that particular end-node. Even ZigBee and IEEE 802.15.4 are still not
quite optimal in their protocols here, albeit vendor proprietary
extensions to that do already exist.

You won't find a receiver that is on one hand sensitive enough in the
frequency range of interest here (which is probably one out of either
433, 868/910, or 2450 MHz) yet still only consumes a few microampere
of current when being run in continuous operation. Either you're
getting something that doesn't draw current but requires a lightning
stroke to open the receiver :), or it will draw quite some amount of
current just for listening (between about 10 to 25 mA, depending on the
manufacturer, technology, and frequency band) but has a good enough
sensitivity for your desired transmission range.

Also, keep in mind that the higher the frequency, the smaller your
antennas will get, but the more current the circuitry will consume
(as a general rule of thumb), and the more critical any kind of obstacle
or reflective medium within the propagation path will be. So while a
couple of boards equipped with Atmel's AT86RF230 802.15.4 transceivers
(using the PCB antenna described in one of the appnotes) made it between
700 and 1000 m of transmission path in open air, the usual range you
could achieve within an office building is more like about 50 m, and
even there you'll already find local "dead spots", points where radio
interference will cause the transmission to fail, yet if you move on by
some 10 cm, everything's back up fine again.

Jörg Wunsch

Please don't send me PMs, use email if you want to approach me personally.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

icarus1 wrote:
Jim,

I hope that you are right and I believe that you are, because where is the use of a wireless protocol. The need is to have low consumption because the more percentage of wireless systems is to work without the need of external supply.
I red an article in a Europian magazine (EETIMES) that compares the ZigBee with another one called ANT (from www.nordicsemi.com).

Michael

As we said earlier in this thread, keep "ZigBee" separate from "802.15.4". XBee is the name of Digi/Maxstream's product, not the name of a protocol. If you don't need mesh routing, then you don't need ZigBee atop 802.15.4. The beauty of 802.15.4 OEM modules from many vendors is that it's not proprietary and the API into the MAC layer services is defined by IEEE 802.15.4.

If you choose something sole-source and proprietary, you will regret it if your product/project has a multi-year life.

May I suggest you do a bit of old fashioned engineering study and reading about 802.15.4 (you can download the IEEE spec) and don't rely on junk written in the trade journals.

You won't do much better than 802.15.4 modules that are FCC (and international) type certified as unlicensed. In Japan and France, for example, the max radiated power is 10mW; in the US it's higher.. So beware the scope of your project.

On the legal front: beware that if you start with other than a type-certified OEM module, YOU or your customer will pay $$$$ to test and prove/file compliance tests for FCC Part 15 and other countries' equivalents. That's why there's a market for these already-certified OEM modules.

we can help you a bit.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

As I promissed yesterday, I post the EETIMES article that compares the ZigBee with the ANT.

Attachment(s): 

Michael.

User of:
IAR Embedded Workbench C/C++ Compiler
Altium Designer