CC110L: CC110L 433,92MHz range issues

Hi,

433 MHz is a more tricky frequency band when comparing to the 868/915 MHz band. The range distance can be affected more because of:

• for the same size antenna, the bandwidth is reduced and more sensitive to detuning
• higher environment noise floor that practically reduces the effective sensitive level
• lower output power compared to 868 MHz band

Are the passive components used in the front-end according to the CC110L 433 MHz reference design ?

Have you tuned the antenna when assembled in the final casing ?

Have you tested with a 433 MHz TI kit to compare the range distances ?

Have you measured the NF at 433 MHz with a spectrum analyzer set to the same bandwidth as the Rx bandwidth in the receiver ?

The NF varies significantly from country to country.

Regards,

   Richard

Hi,

For the antenna we use a 17cm rigid wire, and the passive components as specified in the datasheet reference design.

Also, we tuned the antenna and changed from a switched power supply to a linear regulator in order to minimize noise, as you can see in the picture below:

The yellow and teal signal graph are max holds. Values from yellow max hold is the device with a switched power supply, while the teal one is with a linear regulator. (This measurement has been carried out in the lab)

From here we took the device and tested the signal range in open field, in which we have measured that there is no external noise that could affect the test. (Average noise level = -90dBm)

The results obtained from the test are, that if the device at 433MHz has a switched power supply it has a range of ~67 meters aproximately and without it is around ~70 meters. As you can see, the noise from the switched power supply hardly affects the effective range.

About the 433 MHz TI kit, we do not have any to compare with. However, we have a CC1100DK 868/915 MHz development kit and CC1110-CC1111 kit, but I  do not think they are of any use for 433MHz.

I think that the best way to proceed, is to provide you with the equipments that we are talking about.

This allows you to review the situation, and also answers many of the questions that you may have.

Regards,

Felip

Hi Felip,

You could change the front-end passives on the CC1100 868/915 MHz EM to the BOM for 433 MHz.

If the average noise level is at -90 dBm then this will affect the range distance since this will set the practical sensitivity limit.

What is your datarate and output power level ?

Regards,

   Richard

Hi,

We have tested the radio range using the same settings::

Base Frequency: 433.919830 MHz

Data rate: 100 kBaud

Modulation: GFSK

Devation: 47 kHz

TX Power: 5 dBm

The result obtained from this test is that the max. range is around 40 meters, and from this point the communication starts to fail and data packages are lost.

This test has been carried out indoors. Comparing the data from our transmitter/receiver with the development kit (both in the same test conditions) range, we can say that both of them has the same range.

Do you have and suggestion on what else we could try?

Kind regards, 

Felip

The noise floor of -90 dBm is limiting the range but even with this high noise floor I would expect a greater range than 40 m.

Can you share the SA screen shot that the conducted output power level is around 5 dBm ?

  - measure the frequency offset at the same time.

Can you share the VNA screen shot that the antenna is tuned to 434 MHz.

I would expect about 120 m range with a 2 dB loss antenna and a NF of -90 dBm.

//Richard

Hi,

Using the same settings we got the following result:

As you can see it has a power level of 3.282 dBm and a frequency of 433.96MHz.

There is a difference of 1.718 dBm and 42.5 kHz.

And as you requested here is the antenna's VNA:

Felip

Hi,

Your frequency offset is large and this should be reduced.

Signal BW  = Data rate + 2 x frequency deviation = 194 kHz

Required RX filter bandwidth can be approximated as :Signal BW  + 4*ppm xtal*Frequency of operation = 194 kHz + 4*42.5 kHz = 364 kHz.

The frequency offset should be under 30 kHz otherwise you must increase the Rx bandwidth from 325 kHz.

Regards,

   Richard

Hi Richard,

I have done a range test following your suggestions and the range has improved up to 150 meters.

However, what I am looking for is to reach a range around 400 meters similar to 868MHz. For the time being, I am tuning the XTAL capacitors in order to lower the frequency offset, but it does not seem that this would help me to achieve my objective.

Do you have any other suggestions to improve the range any further?

Regards,

Felip

Hi,

Why not send a higher output power, e.g. 10 dBm ?

This should approx double your range from 150 m (+3 dBm) to 300 m (+10 dBm).

Which country / regulations are you targeting ? 

The data rate can be reduced as well, then the sensitivity will also be improved which will increase the range.

Regards,

   Richard

Hi Felip,

Any updates or should we close this post ?

Hi Richard,

The problem has not been solved yet.

We expected a similar range for the CC110L at both frequencies, 433MHz and 868MHz.

However, the effective range at 433MHz is much shorter, and needs a solution.

As you suggested, increasing the output power will increase the range. However, this is not what we are looking for.

The reason is, that the transmitter is powered by batteries, and more output power shortens the battery lifetime.

Some of our customers already complained about having to change the batteries too often.

Right now we are using the development kit to compare if the CC110L always works better at 868MHz than at 433MHz by testing conducted transmissions.

Should this be the case, than we may have to think of a compatible TI alternative for the CC110L that works better at 433MHz.

Can You recommend any modern TI alternative, similar in package, pinout and registers ?

We will keep you updated about the test results, and are looking forward to your reply on possible alternatives.

Regards,

Felip

Hi,

As mentioned previously, the noise floor at 433 MHz is higher than 868 MHz, therefore the range will be reduced compared to the 868 MHz band.

This is valid for all radio solutions that 433 MHz range will be lower than 868 MHz due to the noise floor level. This is not specific for CC110L.

We see exactly the same ourselves with range testing in Norway.

We have summarized the our range measurements v frequency in the figure shown below: