[Hackrf-dev] Antennas, Cables and Connectors [Important Questions on RF Performance]

[Michael Ossmann]

On Thu, Sep 04, 2014 at 08:16:34AM -0400, Sohil Shah wrote:
>
> He made the switch from RP-SMA in the Jawbreaker to SMA

Jawbreaker had SMA connectors, not RP-SMA.

> but will using the Right Angled Connector as opposed to the straight
> connector make a difference

No.

> Mike was there any particular reason you used the straight connector
> for your prototype backers as opposed to the right angled one?

Yes.  The reason I used vertical connectors was that I have a drawer
full of vertical connectors.  :-)

> Q2. I also wanted to know if these products are high quality products
> based on the spec sheets I provided

They look fine to me.  The main specifications you should look for in
cables and connectors, in order of importance (in my opinion), are:

1. Do they fit?
2. Are they 50 ohm?
3. Do they claim to operate over the entire frequency range you intend
to use?
4. What is the insertion loss?

Most good connectors will have specifications that allow you to answer
at least some of these questions without having to test them yourself.

> I want to use this configuration as a sacrificial connector as
> mentioned by Mike in
> https://www.kickstarter.com/projects/mossmann/hackrf-an-open-source-sdr-platform/posts/758974#comments
> so as to extend the life of the SMA Female Connector on the HackRF
> over 500 cycles(Mike mentioned you can get well over 500 cycles but
> this just increases the life even more). Will this create too much
> loss of signal?

No.  (Well, it possibly could but only for very rare applications.)

> Q4. I see some of the Antenna manufacturers creating these graphs of
> signal strength over frequency as seen in the Spec sheet for the LPA
> http://www.wa5vjb.com/pcb-pdfs/LP8565.pdf on page 2 and even in Mike's
> Presentation https://www.youtube.com/watch?v=4Lgdtr7ylNY Jump to
> [20:10] Is there a reliable way across different software platforms to
> create these graphs.

No.  With Software Defined Radio, you can't change the software and
expect the same performance.  To make graphs such as the spectrum
analyzer plot in my presentation, you need calibrated RF test equipment.
To compare the performance of a separate HackRF One, you must use the
same software that was used with the first HackRF One.

> I would really like to test the RF performance of the HackRF with
> different combinations of these connectors, cables and antennas. Mike
> would you consider making a video that shows us how to do that.

The right way to do that is to use calibrated RF test equipment.  I'm
sure you can find existing videos online about how to use RF test
equipment to measure cables, connectors and antennas.

I consider HackRF One to be uncalibrated RF test equipment.  If you go
through calibration of a particular software+hardware setup, then you
can use HackRF One for certain calibrated measurements.  However, that
calibration step requires calibrated equipment.

It is possible to make relative measurements with uncalibrated HackRF
Ones.  For example, you can use one HackRF One to transmit through a
cable and attenuator to another HackRF One.  You can use such a setup to
compare the performance of multiple cables against one another.  At some
point I might be willing to do a video on doing such relative
measurements, but it wouldn't be any time soon; I'm trying to produce a
series on SDR, not on RF testing.

> Q5. Mike this is an impressive graphic though from a different project
> https://www.kickstarter.com/projects/1085541682/bladerf-usb-30-software-defined-radio/posts/491106
> would you cover something similar to help demystify the use and
> creation of such graphs to help us understand Radio and calibration of
> our devices better?

My video series will cover QPSK and other modulations in a way I hope
will demystify such impressive graphics.

Mike