Re: [Discuss-gnuradio] Just wondering why?

[Cinaed Simson]

On 11/01/2017 01:55 PM, Adrian Hodgson wrote:
> Thanks for responding I have posted responses within the text.
> 
> On Wednesday, 1 November 2017 20:04:57 GMT Cinaed Simson wrote:
>> On 10/30/2017 04:04 PM, Adrian Hodgson wrote:
>>> Quite new to this so perhaps a dumb question.
>>>
>>> I have the following flowchart Narrow_band_QT.grc and screenshot attached,
>>> just something I have been playing with.
>>
>> Note, you should set the RF again to 0 when receiving. There's frontend
>> on the hackrf and you run risk of smoking the receiving RF amplifier if
>> there's a strong RF transmission nearby.
>> Also, the RF gain is either 0 or 1 - on or off.
>>
>> See the hackrf tutorials greatscottgadgets.com website.
>>
> Not a problem for me most of my testing is done with my test set, the RF gain 
> set at 14 dB means I can receive signals down to -115 dBm, I tend to use 
> signals at around -80 dBm so I am far away from the -5 dBm maximum for the rf 
> amp.
> I did have a play with my wireless doorbell following on from the smart 
> gadgets tutorials and eventually got it to work, but my main use would be 
> monitoring the local amateur repeaters, being a licensed radio ham!
> 
>>> There are a few variables I use one is band_width.  Now the strange thing
>>> is if I start with a default of 100KHz band width when running I can
>>> scroll down to 10 KHz bandwidth and then go back up again, but if I start
>>> with the default set to 10 KHz bandwidth it will not scroll up to 100KHz.
>>> It is more noticeable when looking at the bandwidth tab which is directly
>>> after the rational resampler. one can see the pass-band chance width as i
>>> change the bandwidth.
>>
>> The bandwidth is equal the sampling rate - in your case - 2M.
> 
> I think I did not explain myself correctly
> 
> I did find two issues with my flow chart, one was that I used one variable to 
> feed another variable, so removed that issue.
> 
> The second was the QT freq display sample rate after the resampler should 
> have 
> been at 240kHz not 2M.  When set to 2450K then the frequency display matches 
> the true frequency.

240 KHz is the output sampling rate of the resampler - which matches the
quadrature rate - or input sampling rate of the nbfm block.

> 
> But that still did not resolve the issue I am having.  If I set the default 
> band_width variable to 100e3 and run the flow graph, if you go to the 
> band_width tab and scroll on the bandwidth number you will see the width of 
> passband signal reduce and go back up as you scroll.
> 
> But set the default band_width to 10e3 and run the flow graph and it does not 
> work.

Okay, I think I know what you're doing - I didn't look inside the
resampler block the last time.

If you scale the interpolation in the resampler block, you change the
output sampling rate of the resampler block - which no longer matches
the quadrature rate - or the input sampling rate of the nbfm block.

In short, the plumbing is screwed up.

You need to scale the quadrature rate - or the input sampling rate of
the nbfm block so it matches the output sampling rate of the resampler
block after you re-scale the interpolation.

-- Cinaed


> 
> Attached are modified files.
> 
> The grc should run with a rtl2833 as well as a hackrf one, at least I believe 
> so as I use the osmocom source block.
> 
>>
>> Look at with hackrf WBFM tutorial greatscottgadgets.com website.
>>
> Went through that.
> 
>> The channel width of the NBFM is roughly 12-16 kHz - or even 20 kHz
>> depending on which part of the RF spectrum your working.
> 
> Granted but the demod is working all the way up to 100KHz bandwidth or 20kHz 
> mod, the paramters change as I change the  the counters.
>>
>> The Power Squelch may not work on the hackrf - I tried it awhile back
>> and I couldn't get it to work. But things have changed since then so
>> your mileage may vary.
> 
> Yes it does work for me.
>>
>> The only parameter you may be able to change reliably using a runtime
>> slider on the hackrf may be the center and offset frequencies - provided
>> the offset frequency is within the center frequency +- sampling rate/2.
>>
>> I did play with sliders on the frequency and sampling rate but I had
>> trouble - and since I didn't need it - I didn't pursue it. That was a
>> couple of years ago so thing may have changed. And if I recall correctly
>> it was changing the runtime sampling rate which gave me trouble.
>>
>> You should post your hackrf questions on the
>>
>>   address@hidden
> 
> Will do.
> 
>>
>>> I am using it with hackrf one, to which I will have several questions
>>> about
>>> but will leave that to a later post.
>>> I am lucky in that I have things like a spectrum analyzer and radio test
>>> set to work things through with.  I am slowly working through the
>>> tutorials and a book called Practical Signal processing, perhaps my age
>>> but finding it hard going especially when it comes to Python, cmake, and
>>> some of the mathematics!
>>>
>>> Anyway if anyone can let me know why I get the results I do I would
>>> appreciate it.
>>> OS is SUSE Leap 42.3, GNU radio is 3.7.11, just in case that makes a
>>> difference.
>>> Cheers
>>> Adrian
> 
> 
> One of the next steps will be to get the dsd blocked installed if I figure 
> that 
> out, one of the local repeaters is either analogue or digital I believe.
> 
> Thanks
> 
> Adrian
> 
> 
> 
> _______________________________________________
> Discuss-gnuradio mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
>