Date   
#firmware #hardware #firmware #hardware

 

Changing sample rate

A 10MHz sample rate is somewhat odd in that it makes it difficult to decimate down to a standard audio sample rate for the I/Q channels. Is it possible to change the programming of the clock generator to get an ADC clock that is a multiple of 48kHz? For example, 9984kHz is 48kHz * 208. That is close to 10MHz. I notice that the reference input says, "10MHz to 100MHz." How do I tell the Airspy what the reference clock is? If the reference is assumed to be 10MHz and I inject 9984kHz with the Airspy lock to that? I would think that 16kHz would be too far to pull the internal oscillator.

Thank you.

Brian Lloyd
brian@...

Re: Optimizing R2 for FM DX

Ian DXer
 

Paul> …is with Vasilis IF processor plugin, the asymmetric filtering is a real bonus and has made the difference between IDing something or not so many times I've lost count.

Same results here J

Also many months ago it dawned on me I could probably use the notch filter on a local/semi local FM BCB station that was just transmitting an unmodulated carrier to ‘uncover’ & hear the much weaker dx station on the same frequency. All I had to do was create a notch for the centre frequency & another two notches for each of the 19kHz pilot carrier sidebands.
Result was clear reception of the music station underneath.
Back in the ‘old-days’ I would’ve needed to use a phasing unit, attenuators/amplifier & two antennas to try and achieve the same result - in this example.


Paul> In some situations careful use of the notch filter allows you to pull RDS off a station that is buried under a stronger signal which is a useful tool.

That’s something I never thought of trying. I’ll give it a try next time an opportunity presents itself. Thanks Paul.

Re: Optimizing R2 for FM DX

Ian DXer
 

Thanks Pat & Youssef for feedback

Yssf> If there is enough demand, then maybe I can design a small program with all these automagic features. "DX+" looks like a good name ;-)

Well I’m definitely interested, so please add me to the list :-)

As an additional feature request for ‘DX+ or SDR#’; I’m very interested in the ‘Splatter Removal Tool’ (SRT) that Leif (SM5BSZ) describes on his Youtube channel here:

https://www.youtube.com/watch?v=V8_0XIbKCzk

and longer 43min version here: https://www.youtube.com/watch?v=DhBI7YLKTxE


A simple(er) SRT GUI version for inclusion within “DX+” software that is easier for users to implement & use would be an amazingly useful FM DX tool for usage with the Airspy family of SDRs for Windows PCs.

I’m a little disappointed that there hasn’t been more interest in Leif’s ‘Splatter Removal Tool’ (SRT).
I suspect the problem is:-

1.    1. Lack of widespread publicity

2.    2. Some dxers have been slow to migrate from standalone receivers to the new SDR technologies & some may not have sufficient technical knowledge or confidence to attempt to install or use SRT/Linrad as described.

3.    3. Difficult concept for users to follow given Leif’s description on Youtube.

In regard to point 3, I think, if I was to promote & better describe what is occurring here I would initially produce a video using a combination of whiteboard instruction beginning with block diagrams & verbal instruction of setup & methodology & also what we are trying to achieve combined with nil-error concise video instruction of how to implement, along with existing examples etc.

I love watching the genius Leif at work with his videos & following his work via several mediums.
I’d really love to see his initial hard work with the Splatter Removal Technique developed further.

73s
Ian


Re: Optimizing R2 for FM DX

prog
 

On Fri, Feb 1, 2019 at 03:09 PM, Ian DXer wrote:
He provides some helpful hints on getting the best out of the HF+ for FM DX & makes some software development suggestions that Youssef & friends might care to consider in future. Konrad suggests further software development (SDR# or similar) might result in ever better performance on FM BCB with HF+.

Very nice review, indeed! The observations are consistent with what we know about both devices.

The main purpose of the HF+ project is to design an inexpensive *HF * receiver with the performance of the high-end segment. The good FM front end is literally a plus :-)

If there is enough demand, then maybe I can design a small program with all these automagic features. "DX+" looks like a good name ;-)

If there is even more demand, I can also design a tabletop receiver with audio output and all this automagic embedded into the DSP. Our friends at ST already developed these goodies for the rich cousins of the STA709 yet to be released. The silicon has the same front end, more processing power and a high quality DAC. Anyone?

Re: Optimizing R2 for FM DX

Patrick
 

What a detailed review !
Thanks Ian for the link.

Pat (HF+ owner and FM DXer)


Le ven. 1 févr. 2019 à 15:09, Ian DXer <baxterinoz@...> a écrit :
Reading Paul's & Youssef's comments I was wondering how the HF+ would perform against the  XDR-F1HD as a FM DX receiver & had wondered if anyone had indeed performed a comparative test/review.
To my pleasant surprise a Google search resulted in this find (below); just published 2-3 days ago & makes for recommended reading.


I haven't (yet) watched the reviews videos, but it is evident from reading the review that Konrad is very familiar with both receivers & doesn't hold back on providing an extensive review from a FM DXers perspective. He provides some helpful hints on getting the best out of the HF+ for FM DX & makes some software development suggestions that Youssef & friends might care to consider in future. Konrad suggests further software development (SDR# or similar) might result in ever better performance on FM BCB with HF+.

Would welcome comments.

Ian
(HF+ owner)

Re: R2 on RPi3b+

SkipF, NT1G
 

Be sure to opt for the eMMC memory card, too.
Much quicker than the SD card!

Re: Optimizing R2 for FM DX

Ian DXer
 

Reading Paul's & Youssef's comments I was wondering how the HF+ would perform against the  XDR-F1HD as a FM DX receiver & had wondered if anyone had indeed performed a comparative test/review.
To my pleasant surprise a Google search resulted in this find (below); just published 2-3 days ago & makes for recommended reading.


I haven't (yet) watched the reviews videos, but it is evident from reading the review that Konrad is very familiar with both receivers & doesn't hold back on providing an extensive review from a FM DXers perspective. He provides some helpful hints on getting the best out of the HF+ for FM DX & makes some software development suggestions that Youssef & friends might care to consider in future. Konrad suggests further software development (SDR# or similar) might result in ever better performance on FM BCB with HF+.

Would welcome comments.

Ian
(HF+ owner)

Re: R2 on RPi3b+

Martin Smith
 

On Thu, Jan 31, 2019 at 04:35 PM, Bryon NF6M wrote:


a faster CPU (1.6Ghz)
From this page ( http://www.orangepi.org/orangepiplus2/ ) the CPU is a Allwinner H3.
And looking up information about the Allwinner H3 CPU here ( http://linux-sunxi.org/H3 ) says "It's rated to run at 1.296GHz, anything higher is overclocking (and usually overvolting). If you run it without heatsink, fan and proper dvfs settings, you risk overheating.". The Allwinner H3 CPU has 1x 2.0 OTG and 3x Host 2.0 USB ports all of which support High Speed, which is 3 more real USB 2.0 HS ports than all the RPi CPU's.

Looking at the board that OrangePi are selling, it has no fan or heatsink, so it may be able to function at 1.6GHz for short periods of time until thermal overload forces it to use a lower clock speed, but that will shorten the useful lifetime of the product (but you can always use "cpufreq-set" to limit the maximum clock rate and extend the lifetime of the part). The soldered down 16GB eMMc (CPU supports eMMc up to rev 4.41) BGA flash, will eventually fail after 100k writes per block, but as long as the device can boot from the TransFlash/SD slot that should not be an issue. And most Linux distributions with enough knowledge can be modified to work from a readonly filesystems which would extend the useful lifetime of the onboard eMMc to be longer than the useful lifetime of the product. This is never done by default for many reasons, it is a really bad idea unless you know exactly what you are doing and all the future security ramifications.

They are using one USB 2.0 HS port for SATA 2.0 (via GL830 USB-to-SATA bridge chip).
The Realtek RTL8189ETV, IEEE 802.11 b/g/n is on a module that use SDIO, so not using a USB 2.0 HS port.
The USB 2.0 OTG HS port is exposed.
And four USB ports are created through a FE1.1S usb hub chip (1-to-4 USB hub)
So it actually looks to me like one potential USB 2.0 HS port is not routed from the CPU and exposed through any external connectors. So if I was using that board I'd probably use the OTG port for SDR hardware.


On paper the Orange Pi Plus 2E might be a better option, same CPU, same RAM, same 16GB eMMC Flash, cheaper price, with a RTL8189FTV chip soldered to the PCB instead of on a module and it actually exposes all three USB 2.0 HS host ports (no internal USB hub and no USB-to-SATA bridge), the USB 2.0 OTG port has a microUSB connector.

Disclaimer: I don't own either board, but I usually look for potential flaws or gotchas with products before I spend money. Either board should, at least on paper, perform better than a RPi because they have a dedicated GMAC(Gigabit Media Access Controller) built into the CPU and are not using a USB hub chip with integrated USB NIC(Network interface controller).

Re: R2 on RPi3b+

BryonB
 

Has anyone looked at the new Orange Pi Plus2? 

It is priced similarly to Raspberry Pi 3, but has 2G RAM, a faster CPU (1.6Ghz), and comes with 16G of somewhat faster flash storage built in (but still with SD slot for expansion). I'm wondering if it might be "just right" - between performance of Raspberry Pi3 and oDroid. It claims to onky need 2A supply - slightly less than Raspberry Pi.

--Bryon, NF6M



On Thu, 31 Jan 2019 4:01 am prog <info@... wrote:
On Thu, Jan 31, 2019 at 12:59 PM, Mike Gould wrote:
I'll put my odroid order in today then
Get their AC adapter as well to avoid any problems.

Re: Optimizing R2 for FM DX

prog
 

For single channel DX, you better look at the HF+. The XDR uses an older generation tuner from NXP that is now a few years old, and not quite as good as modern tuners like the one we are using in the HF+, although they share some common aspects (The base technology was designed by a joint venture between ST and NXP, then every company developed it further).
For wide band operation, I think you probably already have the best option.

Re: Optimizing R2 for FM DX

Paul <aceblaggard@...>
 

I'm guessing you want to have as much bandwidth as possible, there really isn't much point in getting an R2 for any other purpose on B2 if you already use an XDR. Sensitivity can match the XDR although if there's anything on B2 remotely high power transmitting locally or in the direction of the beam then you'll struggle. I have about 50° to my SE where I can increase gains so the performance at 8MHz view is the same as the XDR but outside of these boundaries the R2 just overloads until the gain is backed down. Once I get within about 40° of a 5 x 100W site a few KM away the R2 is hopeless and no amount of decimation and gain reduction will allow me to get anywhere near the sensitivity of the XDR.

My optimal settings for sensitivity are in free mode 5/15/12, that gives better performance than either of the more simple sensitivity and linearity controls and one reason why I've stuck with SDR Sharp because the last time I checked none of the other programmes had the free option. Once I start beaming towards local masts IF gain has to be reduced to 4 then start lowering the LNA gain.

I've found selectivity to be on a par with the XDR (depending on the order value but it's a trade off if you're trying for RDS) although mine is modified with external PC control so I'm able to manually adjust the bandwidth to make better comparisons. One advantage of the R2 and Sharp is with Vasilis IF processor plugin, the asymmetric filtering is a real bonus and has made the difference between IDing something or not so many times I've lost count. In some situations careful use of the notch filter allows you to pull RDS off a station that is buried under a stronger signal which is a useful tool.

Paul

Re: R2 on RPi3b+

prog
 

On Thu, Jan 31, 2019 at 12:59 PM, Mike Gould wrote:
I'll put my odroid order in today then
Get their AC adapter as well to avoid any problems.

Re: R2 on RPi3b+

Mike Gould <mikeyg123@...>
 

I'll put my odroid order in today then 
Thanks for the info everyone 😃

On Wed, 30 Jan 2019 at 14:11, Michał Frątczak <f.michal@...> wrote:
My XU4 runs fine on 5.15V

Re: R2 on RPi3b+

Michał Frątczak
 

My XU4 runs fine on 5.15V

Re: R2 on RPi3b+

Christof Proft
 

Hi,

I have an Odroid XU4. This computer has a problem with its power requirements, it will only work with 5,8 to 6 V.
With 5 to 5,5V the Odroid will not even boot...
So I will replace it with a RPI 3B+, because I dont want to have a fourth alimentation voltage on SWL camps.

vy73
chris

Re: R2 on RPi3b+

Andrew Jones
 

All,

Thought it worth asking what Linux SBC's do people recommend/and or find work well. We know the RPi can do an 'okay' job, but has some technical limitations, whilst the Odroid-XU4 can perform very well (perhaps it is even overkill for a lot of users?), but for double the price. Has anybody else had any experience with some of the other boards out there? Tinkerboard, BananaPi series, OrangePi Plus, etc...

Andy

Re: R2 on RPi3b+

f1hdi
 

Folks,

On an odroid xu4 board , you can run max four clients (safely 3 without any other comsuming process) @8MHz bandwith, it's a little bit short for a daily job server but it works.

Be carefull with odroid boards, they are very sensitive to static.

73

Le 30/01/2019 à 00:56, prog a écrit :
On Wed, Jan 30, 2019 at 12:49 AM, Michał Frątczak wrote:
Can rPI really pull 10MSPS from an AirSpy? Because last time I tried it could not, I'll verify again this evening.
OTOH, odroid XU4 can do this with no problem.
Technically, one can more or less stream the data at 20 MSPS Real, but the RPi's CPU cannot do anything useful with it for general purpose radio.
Even the conversion of that data from Real to IQ will choke.

Re: R2 on RPi3b+

prog
 

On Wed, Jan 30, 2019 at 12:49 AM, Michał Frątczak wrote:
Can rPI really pull 10MSPS from an AirSpy? Because last time I tried it could not, I'll verify again this evening.
OTOH, odroid XU4 can do this with no problem.
Technically, one can more or less stream the data at 20 MSPS Real, but the RPi's CPU cannot do anything useful with it for general purpose radio.
Even the conversion of that data from Real to IQ will choke.

Re: Optimizing R2 for FM DX

prog
 

On Wed, Jan 30, 2019 at 12:51 AM, Paul Ormandy wrote:
Would appreciate any tips to elevate its performance to somewhere near the XDR (if that's possible).
Use the decimation to narrow the IF bandwidth to the strict minimum.

Re: Airspy Website SSL Certificate 2019 Expired

prog
 

On Wed, Jan 30, 2019 at 12:51 AM, Andrew Peek wrote:

To airspy.com website admin, airspy.com SSL certificate has expired today,

 

Expired: 29 January 2019, 19:15:02 GMT

The auto-renew didn't work for some reason. Now fixed.