Hey guys,
I have built a few devices using the popular nRF24L01+ chips. These are RF communication modules that communicate in the 2.4Mhz bandwidth.
Here is a picture of one, and my questions are below:

And here is the circuit schematic of the nRF24L01+:

So the modules are working great, my only problem is that the range I am getting is not good. I am only getting roughly 25 feet / 7 meters in an office/house setting, and I need at an absolute minimum to double that range.

I have been experimenting with the different datarates. Supposedly the lower the data rate, the more reliable and the more range you get. My options are 250 kb/s, 1mb/s, and 2mb/s. From my experimenting, I got the best reliability and range from the 1mb/s.... for some unknown reason.

I have also been experimenting with using different channels, which did not seem to make any difference.

I wanted to note that the amount of data being transmitted is very small, so date rate is not important to me. I also wanted to note that these devices are plugged in so power consumption is not something I am worried about, I just want these to work.

Questions:
1. Is it possible to modify the nRF24L01+ module so that it produces a stronger/amplified signal? E.g. desoldering some components on the module and replacing them with something else, or a further modification if necessary.
2. Can anyone offer any advice on what I might try to increase this range?

Thanks and any help is greatly appreciated!

 

What output power did you program your device for?

 

What output power did you program your device for?

Choices are -18dBm, -12dBm, -6dBm, and 0dBm. I have it set to the maximum output power of 0dBm.

 

Some of these modules that I've seen can go as high as 5dBm.

What voltage are you using?
What is the current usage during transmit?
What are you using as a receiver?

 

Some of these modules that I've seen can go as high as 5dBm.

What voltage are you using?
What is the current usage during transmit?
What are you using as a receiver?

Is there something I can do to boost it up to 5 or 10dBm? Sorry if this is a trivial question.

It is being powered from 3.3V.
Per the datasheet, its supposedly using 11.3mA during transmissions at 0dBm (I don't really have a good way of measuring this currently)
I am using the exact same module for both transmitting and receiving.

 

Measuring current is one quick way of determining if your module is actually transmitting at 0dBm. You can't break Vcc and put an ammeter in line to measure current?

 

Measuring current is one quick way of determining if your module is actually transmitting at 0dBm. You can't break Vcc and put an ammeter in line to measure current?

Well... its on a PCB so I would have to sever the trace. Are you thinking that I have defective nRF24L01's?

 

Did you build the board or buy it?

 

Did you build the board or buy it?

Are you talking about the nRF24L01+ module board? Or the board that the nRF24L01+ is connected to? I bought the nRF24L01+ modules (addicore) the SMD version, then I made a PCB that the module is soldered to.
Here is the link to the nRF24L01+ modules that I am using:
http://www.ebay.com/itm/171967250024?_trksid=p2060353.m2749.l2649&ssPageName=STRK:MEBIDX:IT

 

Can we see a photo of how the two boards are orientated?

 

Well... its on a PCB so I would have to sever the trace. Are you thinking that I have defective nRF24L01's?

No, you can measure the current going into the +5 V pin on the connector. You'll need to jury rig a power cable temporarily. You'll get one reading when transmitting, and another when not transmitting. The difference is related to the transmit power.

Also, in with these types of antennas, polarization can be important. You'll get max range if all the antennas are oriented in the same way.

Lastly, I know these chips are advertised with much more range, but your experience seems typical.

 

Measuring current is one quick way of determining if your module is actually transmitting at 0dBm. You can't break Vcc and put an ammeter in line to measure current?

Can we see a photo of how the two boards are orientated?

No, you can measure the current going into the +5 V pin on the connector. You'll need to jury rig a power cable temporarily. You'll get one reading when transmitting, and another when not transmitting. The difference is related to the transmit power.

Also, in with these types of antennas, polarization can be important. You'll get max range if all the antennas are oriented in the same way.

Lastly, I know these chips are advertised with much more range, but your experience seems typical.

Here is a photo of how they are orientated on the PCB's:

 

Thank you. Just from what I see, I would say your bottom board is sucking a large part of your signal.

Another photo please. One of entire board assembly and one of cover or cabinet.

Also, where is unit installed relative to room volume? i.e. high or low?(ceiling or floor), next to interior or exterior wall?

If I couldn't mount rf board perpendicular to main board.....I would cut out area of main board under antenna section of rf board. If you can cut out section(I can not see other side), leave small strip at top for existing components.

These frequencies are great for low power line of sight. But they do lose range if they have to bounce a lot.

Depending on the units location and structure layout, you might need a repeater for a reliable link.

If a repeater is not available, and source and receiver are stationary, a directional antenna can be used. The antennas won't necessarily point at each other. I have used external structures(skyscrapers, bridges, tank farms) to aim antennas at, establishing reliable links around and not through a stubborn structure.

Just some suggestions, on what information I have.

 

Thank you. Just from what I see, I would say your bottom board is sucking a large part of your signal.

Another photo please. One of entire board assembly and one of cover or cabinet.

Also, where is unit installed relative to room volume? i.e. high or low?(ceiling or floor), next to interior or exterior wall?

If I couldn't mount rf board perpendicular to main board.....I would cut out area of main board under antenna section of rf board. If you can cut out section(I can not see other side), leave small strip at top for existing components.

These frequencies are great for low power line of sight. But they do lose range if they have to bounce a lot.

Depending on the units location and structure layout, you might need a repeater for a reliable link.

If a repeater is not available, and source and receiver are stationary, a directional antenna can be used. The antennas won't necessarily point at each other. I have used external structures(skyscrapers, bridges, tank farms) to aim antennas at, establishing reliable links around and not through a stubborn structure.

Just some suggestions, on what information I have.

This devices mounts to a wall like a picture frame and is inside a plastic enclosure. There are multiple other sensors that are throughout the office/house that are mounted in a similar orientation (all the way down to the floor or shoulder level).

I placed a "keep out" area underneath the antenna area of the nRF24L01+ so the ground plane on the top and bottom layers have been removed. Are you saying that is not good enough and it should be routed out instead? In that case should I just mount this nRF24L01+ module upright (perpindicular to the PCB)?

 

I placed a "keep out" area underneath the antenna area of the nRF24L01+ so the ground plane on the top and bottom layers have been removed. Are you saying that is not good enough and it should be routed out instead? In that case should I just mount this nRF24L01+ module upright (perpindicular to the PCB)?

Well, you "kept out" components and traces, but left the dielectric in. As BR-549 said, you should cut out the area under the antenna. If that is not possible, then use a pin strip to mount the BT board as high above the PCB as you can.

 

As a general rule, if you have the room, the rf board should be mounted perpendicular to your main board.

If that's not possible, mount rf board so that the antenna part of it, is sticking out from and free and clear of other components. Physically isolate that section.

If you have to leave it as pictured, as SLK001 says, cut out the main board.

If possible, use both vertical and horizontal polarization at source. If only one antenna at source, slant it.

If this is for security, consider dividing sensors between two separate located sources.

The cost is nothing and it gives some redundancy.

 

So I ended up doing a test and soldered the RF boards perpendicular to the PCB's and sure enough the range and reliability opened up significantly. I also tested out some other modules with the power amplification circuit and I am now able to get roughly 150 yards!! I redesigned my PCB's to have the module either hang off the side, or to have a cutout so the ground plane (or the FR4 board) is not directly underneath and this seems to be working equivalently to having them perpendicular.

Question:
I am getting 5% dropped payloads with the auto-acknowledgement feature turned on (up to 15 retries 4000uS apart).
I am getting 60-80% dropped payloads with the auto-acknowledgement feature turned off. This makes sense since 0.80 ^ 13 = roughly 5% so that correlates with the auto-acknowledgement feature turned on.

Is that a reasonable/typical drop rate? It seemed like that was a bit high, but I have nothing to base this on. I am still using the 256kb/s setting which is supposedly the most reliable. Any feedback is greatly appreciated, thanks!

 

Has anyone using these modules (or something similar) gotten this kind of drop rate before?

 

If you went from a range of 25 feet to 150 yards, you should expect some degradation of service. Does the percentage of dropped payloads drop as you decrease the range? Frankly, given what I have read about these modules, I am surprised they are working as well as you describe. I have several of the modules that I bought to experiment with, but have not done anything with them. I would be interested in your application.

ETA: It's fairly easy and cheap to build directional antennas for 2.4MHz, which would improve your coverage in selected directions.