Just take a chance on this one as the PIC list is a remarkable place in terms of experience and knowledge.
I got a development kit with the Nordic Semiconductor nrf9E5 8051 based 868 MHz radio transceiver (transceiver and uP in same chip) chip (http://www.nvlsi.no/index.cfm?obj=product&act=display&pro=82) early this year. The kit contains two boards that have a antennas, serial ports, temp. sensor some jumpers, connectors for I/O and A/D and some other extras.
My intention was to implement the VSCP (Very Simple Control Protocol, http://www.vscp.org) on this hardware as a free reference implemention. But I have never got the radio to work. There is a several samples included both precompiled and in source form and binary form and I can get everything to work except the radio.
There is one sample that light a LED on the received board when a button is pressed in the transmitter board. Does not work here.
Another demo is rangetest. In this demo a led is lit as long as packets reach the received board. Always led off here.
After the usual double checks, replacing received unit with transmitter unit, checking and changing power supplies etc Nordic Semiconductors replaced the boards. My problems are the same with the new boards though. Nor The support guy at Nordic Semiconductor or I have a clue on what can be wrong.
Does anyone have experience with nRF9E5 and can give me a suggestion on something to investigate or is there some general things that can go wrong or should be thought of when working with RF (first time for me). I have for instance a cellular station not far from my house. Can that station cause problems?
>
>Does anyone have experience with nRF9E5 and can give me a
>suggestion on something to investigate or is there some general
>things that can go wrong or should be thought of when working with
>RF (first time for me). I have for instance a cellular station not
>far from my house. Can that station cause problems?
What band do they operate in? Here, 860-ish is analog cell band.
Could be severe desense.
I've used the 915 products, and 2400, and they work nicely.
A bit interesting to talk to, but other than that...
>
>>
>> Does anyone have experience with nRF9E5 and can give me a suggestion
>> on something to investigate or is there some general things that can
>> go wrong or should be thought of when working with RF (first time for
>> me). I have for instance a cellular station not far from my house.
>> Can that station cause problems?
>
>
> What band do they operate in? Here, 860-ish is analog cell band.
> Could be severe desense.
>
> I've used the 915 products, and 2400, and they work nicely.
> A bit interesting to talk to, but other than that...
>
>
They can work at 433 / 868 / 915 MHz. My boards are at 868 MHz (IMS band in Europe).
The code I used so far has been from Nordic Semiconductor and included with the dev. kit. so I have assumed that the code is OK. Also loading of it is OK as I can get other things to work (sending data through serial port and such.).
>They can work at 433 / 868 / 915 MHz. My boards are at 868 MHz (IMS
>band in Europe).
>
>The code I used so far has been from Nordic Semiconductor and
>included with the dev. kit. so I have assumed that the code is OK.
>Also loading of it is OK as I can get other things to work (sending
>data through serial port and such.).
If you have the bits, try coupling them together, though say 60-100dB
of attenuation.
That should eliminate any RFI possibilities.
I didn't have any trouble with their demo apps, but I wasn't using
the 8051 version.
>
>> They can work at 433 / 868 / 915 MHz. My boards are at 868 MHz (IMS
>> band in Europe).
>>
>> The code I used so far has been from Nordic Semiconductor and
>> included with the dev. kit. so I have assumed that the code is OK.
>> Also loading of it is OK as I can get other things to work (sending
>> data through serial port and such.).
>
>
> If you have the bits, try coupling them together, though say 60-100dB
> of attenuation.
> That should eliminate any RFI possibilities.
>
> I didn't have any trouble with their demo apps, but I wasn't using the
> 8051 version.
>
Just o make sure I understand. You suggest that I should lower the transmission power. I have just tried full power so far.
>Just o make sure I understand. You suggest that I should lower the
>transmission power. I have just tried full power so far.
No, I mean connecting the two units with cables rather than an
antenna, with significant attenuation inbetween.
That would preclude any on-air interference from bothering you.
If you can't do that, then you need to maybe move your operating
frequency a ways off and see if that helps.
It's easier with a spectrum analyzer handy, just look and see if you
have any significant junk near or on where you are operating.
>
>> Just o make sure I understand. You suggest that I should lower the
>> transmission power. I have just tried full power so far.
>
>
>
> No, I mean connecting the two units with cables rather than an
> antenna, with significant attenuation inbetween.
> That would preclude any on-air interference from bothering you.
>
> If you can't do that, then you need to maybe move your operating
> frequency a ways off and see if that helps.
> It's easier with a spectrum analyzer handy, just look and see if you
> have any significant junk near or on where you are operating.
>
>
>
OK now I understand. Will try that.
> There is one sample that light a LED on the received board when a
> button is pressed in the transmitter board. Does not work here.
>
> Another demo is rangetest. In this demo a led is lit as long as
> packets reach the received board. Always led off here.
Suggestion:
Make a screened box and try the simplest test possible with RX & TX in
the box. IF that works you can suspect local interference. Even
something like a steel (old technology) cake tin with a peep hole
would do.
Taking them somewhere else a substantial distance from where you are
now and from cell sites may help as a test.
Did your returned units work OK when they got them back?
At 09:00 PM 4/22/2005, Russell McMahon wrote:
>>There is one sample that light a LED on the received board when a
>>button is pressed in the transmitter board. Does not work here.
>>
>>Another demo is rangetest. In this demo a led is lit as long as
>>packets reach the received board. Always led off here.
>
>Suggestion:
>
>Make a screened box and try the simplest test possible with RX & TX
>in the box. IF that works you can suspect local interference. Even
>something like a steel (old technology) cake tin with a peep hole would do.
On those bands, making a screening box isn't trivial.
I did one, it took most of a day.
Copper PCB material, finger stock for the lid, and mosfoam for absorber plates.
>Taking them somewhere else a substantial distance from where you are
>now and from cell sites may help as a test.
Good idea. I will try this. Will any metal box do for this?
I sent the development kit back to the application engineer I have contact with as requested but the support person I speak to is in another place. He told he was trying to get hold of the returned kit but he never come back on that. Guess he have other things too do and is a bit fed up with all my questions now... But I still hope to eventually get this info.
I very much appreciate you guys taking time trying to help.
>> There is one sample that light a LED on the received board when a
>> button is pressed in the transmitter board. Does not work here.
>>
>> Another demo is rangetest. In this demo a led is lit as long as
>> packets reach the received board. Always led off here.
>
>
> Suggestion:
>
> Make a screened box and try the simplest test possible with RX & TX in
> the box. IF that works you can suspect local interference. Even
> something like a steel (old technology) cake tin with a peep hole
> would do.
>
> Taking them somewhere else a substantial distance from where you are
> now and from cell sites may help as a test.
>
> Did your returned units work OK when they got them back?
>
>
> RM
>
At 03:03 AM 4/23/2005, Ake Hedman wrote:
>Good idea. I will try this. Will any metal box do for this?
>
>I sent the development kit back to the application engineer I have
>contact with as requested but the support person I speak to is in
>another place. He told he was trying to get hold of the returned kit
>but he never come back on that. Guess he have other things too do and
>is a bit fed up with all my questions now... But I still hope to
>eventually get this info.
It depends on the amount of isolation that you need, and we don't know that.
A microwave oven might prove to have a useful screening effect, but I
don't know how effective that design would be at lower frequencies.
Some of them depend on a 1/4 wave choke around the door.
