[Hackrf-dev] HackRF settling time

[Quentin Smith]

I haven't tried it myself, but my guess is that hackrf_transfer is the 
wrong tool for this - most of the time is probably due to the startup of 
the HackRF, which happens each time you invoke hackrf_transfer, instead of 
just due to tuning.

You should write an app (maybe just a simple modification of 
hackrf_transfer) that opens the device once and then tunes different 
frequencies at fixed intervals, and I bet you'll find a much shorter 
settling time on all subsequent tunings.

--Quentin

On Thu, 9 Oct 2014, Ibrahim Basaran wrote:

> We are trying to sweep all ~ 6GHz with HackRF. In our tests we have seen that it takes 2.5ms to 5ms to settle for a configuration. 
> Our test is simply run hackrf_transfer, record for a second, look at the data in time domain and note the point where it seems settled
> visually. 
> My questions are : 
>  1. Are our settling time values correct?
>  2. Is there any better way to test or has anybody done such tests?
>  3. Which part of the hardware causing this time? Our guess is the clock source.
>  4. Is there any way to make it faster? 
> 
> İbrahim BAŞARAN
> Senior Researcher / Uzman Araştırmacı
> TÜBİTAK BİLGEM M070
> 41470 GEBZE KOCAELİ / TURKEY
> ibrahim.basaran at tubitak.gov.tr
> T +90 262 648 1268 
> F +90 262 648 1100
> 
> www.bilgem.tubitak.gov.tr
> Sorumluluk Reddi
> 
> 
> 
> 
>