Page 1 of 1

Upconverter vs. Direct Samling?

Posted: Mon Mar 09, 2020 8:06 pm
by hotpaw2
Has anyone done any comparison testing (noise, sensitivity, etc.) of using an upconverter, such as the new Ham-It-Up, versus using RTL-SDR V.3 direct sampling for HF?

Re: Upconverter vs. Direct Samling?

Posted: Mon Mar 09, 2020 9:43 pm
by snn47
Hotpaw I would say it depends on how you define the direct sampling setup, you use e.g. 8, 10, 12 ... Bit AD, what type of preamp if you use one, are filters used and which type HP,LPW, BP vs.
the performance specs of the upconverter and by the SDR after the upconverter, you assume 8, 10, 12 ... Bit AD sampling sampling, what type of filters are used (HP,LPW, BP).

Assuming you had the same AD circuit in use in both
- the noise figure would have to be higher for upconversion due to losses in mixer and additional noise generated in components and oscillator.
- the IM would be limited by the upconverter and if this is good enough then by the SDR
- the type of filter employed, the filter bandwidth and slope of attenuation will define how much of the spectrum and how much energy the SDR has to process. With filter beeing easyer to design at frequenices above 100 MHz an upconverter, could, despite higher noise figure still provide better reception due to better filter performance.

There is another option not mentioned by you, which is if the oscillator of the upconverter is tuned instead of the SDR.
Basically most non SDR receivers are upconverters today if they use a frequency tuned oscillator to tune the receive frequency as long as the receiver frequency is below the IF used in the receiver opposed to using a fixed oscillator frequency as used in the upconverter you mention and tune the receive frequency in the SDR. Many SW receiver have for better performance a high first IF, e.g. >45 MHz, compared to older designs that used 9 MHz IF.

Re: Upconverter vs. Direct Samling?

Posted: Tue Mar 10, 2020 8:17 am
by alanzfq
From what I've read the direct sampling of the V3 works surprisingly well.
I say surprisingly because the first attempts at direct conversion were often nearly useless.
But in my opinion it will never equal the more costly upconverter method.

Re: Upconverter vs. Direct Samling?

Posted: Thu Mar 12, 2020 3:54 am
by rtlsdrblog
Assuming V3 direct sampling vs an upconverter:

It's hard to say which is better as it entirely depends on your RF environment and antennas.

With direct sampling if you're getting extremely strong signals then it is easier to overload and cause intermodulation compared to using an upconverter. In those cases it is recommended to use filtering for the band of interest. With weak-moderate signals coming in it should work fine. But you also need to be aware that it images with signals folding around 14.4 MHz

An upconverter can get you better results if you have a good HF antenna setup, and strong signals coming in. It does result in more conversion losses though. At the same time, an upconverter costs $50 + $20 for the dongle. So you're paying $70 + shipping costs for an HF setup. If you're truly interested in high performance HF it would be better to look at an Airspy HF+ Discovery, or an SDRplay RSP1A unit.

My recommendation is:

1) Get a cheap RTL-SDR Blog V3 and try out HF direct sampling. If you're technically inclined, build a few filters if you experience overload. Even if you don't end up using it as your daily HF rig, you could still set it up as a dedicated WSPR/FT8 etc monitor.

2) If direct sampling is insufficient and your truly interested in the HF bands, save up your cash and look at the Airspy HF+ DIscovery and SDRplay units rather than an upconverter, as the step up in performance with be much more significant.