Extending range of regular nRF24L01+

[Yveaux]

I ran into this article: http://www.instructables.com/id/Enhanced-NRF24L01/?ALLSTEPS
It shows how to improve range of a regular nRF24L01+ module by adding a simpel DIY dipole antenna to it:

This guy uses an even simpler solution:

Cheap DIY NRF24L01 Antenna Modification – 02:48
— Pete B

I haven't tried it myself, but it might come in handy for those of you experiencing range issues.

[sundberg84]

Hihi isnt this "other guy " @petewill ?

http://forum.mysensors.org/topic/775/8-lamp-outlet-smart-plug-module/64#

[Yveaux]

@sundberg84 Hehe, good finding Or he might have identical hands as 'the other guy' :metal:
I ran into this video on Hackaday and thought
it would be nice to share it with you guys...

[NeverDie]

That last comment on Hackaday was interesting: "There is another way to extend radio range, and you don’t need additional hardware, just some heavy error correction code. We implemented a LDPC (Low density parity check code) on a cortex m3 with a 2.4 Ghz ISM band chip in one of our projects, with very good results."

Anyone try that?

[Yveaux]

@NeverDie then you need to implement error recovery instead of detection only.
Any errors corrupting the header of the nrf packet will not be recoverable anyhow (unless you continuously run it in some sort of sniffer mode, but you don't want that...)

[NeverDie]

Regarding the antenna mod, then, does it somehow increase efficiency and thereby increase the effective power of the transmission, or does it merely make it more directional?

[Yveaux]

@NeverDie poddibly both, and make it more sensitive when receiving.

[tbowmo]

The antenna will have a certain gain, depending on how it's made, this will be added in both directions (TX/RX). The better antenna you make, the better radio coverage you'll get.

One caveat to making them yourself, is that it has to match the impedance of the radio input, otherwise you'll get into troubles with standing waves that will "destroy" your transmitter / receiver.

[petewill]

@sundberg84 said:

Hihi isnt this "other guy " @petewill ?

Haha! Well, I've definitely been called worse things than "other guy"...

Cool, my video made it on to hackaday! I use this cheap hack all over my house with definite improvement. Like the site says, not the most "technically sound" but it definitely works for me.

[NeverDie]

@petewill said:

@sundberg84 said:

Hihi isnt this "other guy " @petewill ?

Haha! Well, I've definitely been called worse things than "other guy"...

Cool, my video made it on to hackaday! I use this cheap hack all over my house with definite improvement. Like the site says, not the most "technically sound" but it definitely works for me.

Roughly how much improvement would you say? 2x the range, or more like or 50% or less improvement?

[petewill]

@NeverDie Hard to say really. I haven't done any sort of range testing other than placing nodes around my house. I had nodes that were maybe 50 ft and 6 walls away from the gateway that wouldn't communicate at all and when I put in the antenna they communicate every time. I usually suggest that if you're having issues you give this a try since it's easy, cheap and reversible if it doesn't work.

Maybe some day I'll get around to testing the range but for now I'm using all my spare time to make new sensors

[ssrini2]

petewill:
Do we need cat5? Will any cable do?
Also do i need to have this antenna at both the tx and rx? Thanks for your cool video! Will try it and report my observations. We've got concrete walls in India.. I am hoping the range is still alright

[petewill]

@ssrini2 I have only used cat5/cat6 cable but other wire may work. The stuff I use is solid copper (not stranded). There are many different examples out there of what can help but I have always used the method from my video. I would do some experimenting to see what works best for you. With concrete walls you may find that you want to turn on the repeating functionality on some of your MySensors devices.

[ssrini2]

It worked quite well with solid copper wire across concrete. Much appreciated. Thanks. I am now trying to see if there is a smaller length that could suffice for my experiments. Any thoughts on that? Is there a smaller wavelength wire that could suffice? Anway I will post in this forum the lengths that worked for me. Cheers!

[petewill]

@ssrini2 There are specific lengths that will work. I used an antenna calculator to find the length I used. I can't remember where I found it but I think a quick google search should work.

[gohan]

Hi guys, I tried the single wire mod but without any noticeable improvement: with a single wall in between I can't go more than 5 mt or I start getting NACK messages. I'll have to try the dipole or my radio modules are crap (as I always suspected)

[Nca78]

@gohan said in Extending range of regular nRF24L01+:

Hi guys, I tried the single wire mod but without any noticeable improvement: with a single wall in between I can't go more than 5 mt or I start getting NACK messages. I'll have to try the dipole or my radio modules are crap (as I always suspected)

Have you first checked your power source ?
Did you put a capacitor big enough and as close as possible to the VCC/GND pins of the radio module ?

[gohan]

Yes, I'm using both adapters with voltage regulator and caps or 3v from the mega 2560 plus 100uf cap. I suspect the modules because they were very cheap. I'm waiting for the PA to arrive and meanwhile I was just trying to play around with the ones I got.

[Nca78]

@gohan yes then it sounds like bad modules, that extra dollar for a lot of 10 modules often make a lot of difference in quality

[gohan]

The problem is being lucky enough to get working ones. Let's see with dipole if something changes.

[Tmaster]

@gohan said in Extending range of regular nRF24L01+:

Hi guys, I tried the single wire mod but without any noticeable improvement: with a single wall in between I can't go more than 5 mt or I start getting NACK messages. I'll have to try the dipole or my radio modules are crap (as I always suspected)

Tooooooo short! real range of printed antena modules are around 20 to 30m with one wall at midle.
You should have something doing noise,like router on same freq our other 2.4 ghz device on same frequency.
on my case i live ouside city so my router its the only one on 2.4ghz but never cause me problems .
Its really bad luck all your modules come with problems

some of my modules have and laptop antena instead of pcb antena,but only increase range in +/- 5m though walls
there's one increment that i read here from other member did on amplified version; cover the module with aluminium foil ,and leave antenna outside aluminium,i think that avoid signal propagation before antena do this job.

(

[gohan]

I put them on channel 10, that should be away from wifi networks. I leave in the countryside so I don't think I have any other interference. I suspect they are just crappy modules. Hope for the shielded PA to arrive soon (of course all the interesting parts I have ordered are arriving much later than all the other junk )

[petewill]

@Tmaster said in Extending range of regular nRF24L01+:

cover the module with aluminium foil ,and leave antenna outside aluminium

This is very good advice if using the PA version. One of my PA radios was so bad it was actually took down my z-wave network network (yes, z-wave is 900MHz). After wrapping with aluminium foil (and grounding that) it works beautifully.

[gohan]

I am not using the PA version, just a common clone with trace antenna. After rating them 1 star on Amazon, the seller offered a refund in exchange for a change in review.... Sorry but I am only going to add the seller has been kind to offer a refund, but if the modules are junk I can't change my opinion

[Tmaster]

so if you have nothing to lose ,cut the antena trace near last smd component and solder an external antena and try again. should increase signal.

[gohan]

I tried with dipole mod but didn't gain anything.

[mfalkvidd]

@gohan are you aware that you need to align the antennas if you use a dipole?

I am not sure how to align a pcb antenna with a dipole, but two dipoles need to be aligned so they are in parallel. Otherwise the received signal can have lower strength than when using the pcb antenna.

[gohan]

I know but I am not going to waste more time on them