Group Actions
Engagement
  • Author Author: ralphjy
  • Date Created: Date Created
  • Views 5545 views
  • Likes 12 likes
  • Comments 5 comments
Related
Recommended

Bluetooth LE Transmit Testing with the FPC1500

ralphjy

Table of Contents
Blog 1: Bluetooth LE Transmit Testing with the FPC1500
Blog 2: Getting Started
Blog 3: BLE Development Boards
Blog 4: Test Development
Blog 5: ESP32 Measurements
Blog 6: nRF52 Measurements
Blog 7: CYW43-measurements
Blog 8: Summary

This intro blog describes a special project that I am doing for Element14. The project will use a Rohde & Schwarz FPC1500 Spectrum Analyzer  to test the Bluetooth Low Energy (BLE) Transmitter Radio Frequency Physical Layer (RFPHY) on IoT Development boards that I use.

Engineering design validation and manufacturing compliance testing require a comprehensive tester that can provide communication with and control of the EUT to allow automatic testing of both the transmitter and receiver of the RFPHY. The Rohde & Schwarz CMW family of radio communication testers are good examples of that type of tester.

For user level testing, these comprehensive testers tend to be overly complicated and cost prohibitive. To address the need for a simple lower cost test solution for BLE transmit signals, Rohde & Schwarz developed an application solution using the FPC Spectrum Analyzer with the FPC-K7 modulation analysis option. This allows the verification of transmitter characteristics such as the packet structure, output power and modulation. This is described in the following Application Card: The Smarter Way to Test BLE Transmit Signals

I have a large number of IoT Development boards with BLE communication capability. These represent a wide selection of different board and BLE device manufacturers. And a variety of antenna configurations - PCB, chip, and external (via U.FL connector). I've wanted to do quantitative measurements of these boards and configurations, but I haven't had the opportunity or capability until now. I'll admit that I've just accepted the performance that I've gotten and have not even tried adjusting the firmware (the transmitter parameters in the BLE library that I happen to be using). I'm excited to see how well this works and what types of issues that I encounter. Hopefully, this will help me improve BLE performance in future projects.

Ideally, I'd like to have a simple setup that I can use with all of my board types. I'll use a common reference antenna as the input to the FPC1500 and a fixed set of separation distances to measure power/performance loss. I do have some concerns about the ambient RF noise in my environment and what effect it will have on my measurements, so that is something that I'll need to check. My main concern is about the difficulty in programming the various boards with an equivalent test program for comparison purposes. I'm hoping that I can use a common broadcasting function (TBD) that is available in the various BLE libraries and development environments that I use. Direct Test Mode would provide me with the most control and flexibility, but I'm having some difficulty implementing that firmware for some of my device configurations.

As a sanity check, I started with a Sparkfun ESP32 Thing Plus programmed with iBeacon advertising firmware. This particular version of the board has a U.FL connector for an external antenna, so that allows me to try different antennas and also provides the fallback of doing a direct connection to the FPC1500 in case I encounter ambient noise issues (I have a lot of stuff on 2.4GHz - some would be problematic to shut down).

Here is a picture of my setup doing a close proximity test using the default 3dBm transmit power:

image

And the corresponding screen capture using R&S InstrumentView on my PC:

image

So, a promising start with a lot of work ahead…

  • in reply to ralphjy

    Agree, semi-rigid cable is so much easier to self-assemble! : ) I've occasionally bought scrap off-cuts of it from ebay, it makes for awesome cables.

  • in reply to shabaz

    Thanks for that link.  Using the base part number (415-0033), I can find them in US distribution.  It's been a long time since I assembled any SMA cables.

    There was a time when I did a lot of work with semi-rigid - where the only tool needed was an X-acto knife Smile.

  • in reply to michaelkellett

    I've been using these cables, they are double-shielded (RG316DS), and I can't assemble them for a lower cost than buying these. Also some SMA-to-BNC connectors Farnell or RS so that I can use the same cables with 'scope or any other equipment too, since I don't have many BNC-to-BNC cables. For prototyped circuits, I use a SMA connector, mostly from LCSC, they are no name I recongnise, but seem to be of much better quality than ebay etc., so I've got about 60 of them, half in vertical, half in horizontal, for lots of prototyping. For instance I've been using this vertical connector, it seems fine, and others there may be just as good too.

    A reel of thinner coax RG-178 is nice for prototyping too, but it is extremely hard to assemble to a straight SMA connector, but very easy to right-angle connectors. I've been using right-angle connectors from Farnell, but the price is too high (£5+) for more than a few.

  • in reply to michaelkellett

    The particular set of adapters and cables that I am using are from Amazon.  I needed them quickly and the prices were reasonable.  The quality has been okay, but I know that is a risk with these cheaper components. 

    I do not have the R&S calibrator.  That would be nice to have.  I haven't tried the VNA functionality yet.  I'll need to go back and check out the roadtests to see how to manually calibrate and apply the calibration values.  For now I am using a NanoVNA to do a quick check of my cables/adapters and external antennas.

    That's a nice TEM cell.  I'll watch for your posts if you get one.  

    Ralph

  • I look forward to seeing how you get on with this.

    I have an FPC1500 but not used it greatly yet.

    |'m thinking of getting one of these for susceptibility testing and possible emmited RF noise measuring:

    siglent.co.uk/.../

    Where are you buying your RF adapters and cables. There is a huge range of prices for these parts, from really cheap from Amazon and Ebay to crazy expensive (like £50+ for BNC to N adapter). Most of mine are from CPC or Amazon.

    And do you have the R & S calibrator.

    MK