nRF52840 Coded Phy Range Problem

RE: nRF52840 Coded Phy Range Problem

Simonr — Tue, 23 Jul 2019 07:35:38 GMT

Sorry that I can't be of more help at the moment, but please keep me updated on your advancements in this case. If there are any questions, I'm happy to answer!

Best regards,

Simon

RE: nRF52840 Coded Phy Range Problem

Aamir — Tue, 23 Jul 2019 05:57:30 GMT

Thanks for your reply,

Yes we are in talks with HolyIOT and we'll also be doing the tests again in a more suitable environment. we're also in talks for the new DKs. so, we'll have a reference at all times.

RE: nRF52840 Coded Phy Range Problem

Simonr — Mon, 22 Jul 2019 14:18:06 GMT

Very well, then your application code is off the board, which means the problem is either with your test environment or the custom board. You could try contacting HolyIoT to see if they've done any long-range tests with their module to see if they've gotten any better results, or try conducting your test in a more suitable environment for long-range communication.

Best regards,

Simon

RE: nRF52840 Coded Phy Range Problem

Aamir — Mon, 22 Jul 2019 13:42:59 GMT

Hi Simonr,

I've tested the long-range demo code today, what I got was a slight improvement over my code. This test was done in the same environment as my original test.

Tx Power: 8 dbm PHY: Coded PHY

My Code: ~270 meters

Long Range demo code: ~320 meters

So, there was an improvement but not by any significant margin and I think this improvement should mainly be attributed to the advertising interval, as in the long range code, the advertising interval is 50 ms whereas in my code the minimum I've tested was 100 ms.

RE: nRF52840 Coded Phy Range Problem

Aamir — Fri, 19 Jul 2019 09:18:27 GMT

Hi Simonr,

Thanks for your reply,

we'll have to make changes to our custom board in-order to test the long-range demo. Unfortunately, it can only be done after the weekends. I'll let you know the results after we have tested the code.

RE: nRF52840 Coded Phy Range Problem

Simonr — Fri, 19 Jul 2019 05:43:52 GMT

I don't think there are, because I can't think of what should be the limiting factor in the code, but to confirm it for sure, you should test one of the long-range demos on the boards you have (in the same environment), to see if you're able to achieve a similar range.

Best regards,

Simon

RE: nRF52840 Coded Phy Range Problem

Aamir — Thu, 18 Jul 2019 13:50:22 GMT

I have basically modified the blinky examples of both the Central and Peripheral to test the range. (The code I posted above)

could there be a problem with my code, because as far as I understand, the changes I made to code to enable long range should give me similar results as your experiments, shouldn't it ?

yes, environment could be a factor for degradation in range and we are in talks to get our hands on the new DKs

but, that is only possible if we are confident that there is nothing wrong with the code.

So, please let me know if there could be any bottlenecks or limiting factor in the code, then we can move on to the hardware upgrade.

RE: nRF52840 Coded Phy Range Problem

Simonr — Thu, 18 Jul 2019 12:57:11 GMT

Thanks for the clarification. Are you using one of our example projects when testing this? If not, could you try using one of our demos (Testing long range or nRF52840 long range demo) to test the range as well? You should also check out the links awneil posted to see if the environment you're testing it in could be victim to various interference sources. I'd also recommend trying to get your hands on a couple of DKs for testing, as that would be the easiest way to find out if it's the environment or the hardware that's the problem.

Best regards,

Simon

RE: nRF52840 Coded Phy Range Problem

awneil — Thu, 18 Jul 2019 09:26:14 GMT

[quote userid="75734" url="~/f/nordic-q-a/49926/nrf52840-coded-phy-range-problem/198993"]In what environment have you tested this setup? It should be tested outside where the chips are in field of view of each other and without too many interference sources, or else you can experience notable range loss.[/quote]

Indeed.

Height above ground, and proximity to other surfaces also has an effect.

http://www.idc-online.com/technical_references/pdfs/electronic_engineering/Fresnel_Zone.pdf

https://www.fanstel.com/testreport

RE: nRF52840 Coded Phy Range Problem

Aamir — Thu, 18 Jul 2019 09:11:20 GMT

Here is the datasheet for the SoC

devzone.nordicsemi.com/.../HolyIOT-nRF52840.pdf

RE: nRF52840 Coded Phy Range Problem

Aamir — Thu, 18 Jul 2019 09:01:58 GMT

Hi Simon,

Thanks for your support.

Also I only have Preview DKs with me and they are not supporting advertising over Coded Phy,

https://devzone.nordicsemi.com/f/nordic-q-a/49916/nrf52840-coded-phy-peripheral-problem

secondly, I want to clarify, we buy the SOC Chips from HolyIOT, it comes with the chip mounted and the antenna as well. we only make rest of the pcb design and mount the SOC on our board.

So, are you asking about our PCB design or the SOC ?

RE: nRF52840 Coded Phy Range Problem

Simonr — Thu, 18 Jul 2019 05:45:34 GMT

Hi Aamir

Are you able to test using two nRF52840DKs to confirm whether or not the issue is due to the custom boards? I can't see a reason why your code would fail. Could you upload the gerber files of your chip as well as attach a link to the antennas product page/datasheet? If you don't want to share this with the community please just say so and I will set this case to Private mode so that only you and Nordic engineers will be able to see this case.

Best regards,

Simon

RE: nRF52840 Coded Phy Range Problem

Aamir — Wed, 17 Jul 2019 14:03:42 GMT

devzone.nordicsemi.com/.../Peripheral.c devzone.nordicsemi.com/.../Central.c

These are my code files

RE: nRF52840 Coded Phy Range Problem

Aamir — Wed, 17 Jul 2019 13:49:02 GMT

Hi Simonr,

Yes my advertising type is BLE_GAP_ADV_TYPE_EXTENDED_CONNECTABLE_NONSCANNABLE_UNDIRECTED

and the flag is set to BLE_GAP_ADV_FLAGS_LE_ONLY_GENERAL_DISC_MODE

Yes, we did this test outside and both the device were in field of view of each other, but, I can't be sure about interference as there were many Wi-Fi devices around in range.

We have CERAMIC PATCH ANTENNA with dimensions 8 mm x 1 mm

the antenna is identical on both the Central and the Peripheral device.

Even though you could see the pcb under the antenna, but it is just a layer of pcb and does not have ground or power plane or any kind of track beneath it.

RE: nRF52840 Coded Phy Range Problem

Simonr — Wed, 17 Jul 2019 13:09:47 GMT

Have you set the advertising type to extended (BLE_GAP_ADV_TYPE_EXTENDED_CONNECTABLE_NONSCANNABLE_UNDIRECTED)? What is your advertising flag set as? This needs to be "BLE_GAP_ADV_FLAGS_LE_ONLY_GENERAL_DISC_MODE". Both these parameters are needed for long-range mode to work.

In what environment have you tested this setup? It should be tested outside where the chips are in field of view of each other and without too many interference sources, or else you can experience notable range loss.

Seeing as this is a custom chip, it could also be that your antenna is not very well suited for long-range applications. You could test if this is the case by using two nRF52840 DKs instead to see if the range changes. What antenna have you used, and how does your matching network look?

Due to the summer holidays in Norway, our support team is understaffed this week, and delayed replies must be expected. Sorry for the inconvenience!

Best regards,

Simon

RE: nRF52840 Coded Phy Range Problem

Aamir — Wed, 17 Jul 2019 12:54:30 GMT

I forgot to mention, this is all measured over line of sight