I had an idea today to test and use the rtl-sdr V3 that I've had for awhile but neglected in the past year. I've last used the dongle for monitoring the local VHF police comms. It worked well, but since I bought a scanner, the V3 has been gathering dust.I had remembered that the V3 had a "direct sampling" mode that could be activated in software. Normally, the V3 is useable down to roughly 25mhz, but operating in direct mode it was claimed to work down to the lower HF bands. I had read preliminary reports that mentioned many spurious signals plaguing reception, so I dismissed that mode and didn't give it a try. Since the weather has been cold here, boredom set in being confined to the home, I thought I'd do something different with radio. I dusted off the dongle, attached it to my all-band e-probe, and fired up the HDSDR software program on the pc... setting the V3 into the direct sample mode.
I tried the unit on all my favorite "haunts" from Upper HF down to lower HF. I was really pleasantly surprised at the performance of it. I didn't experience the spurious responses others reported... the HF band actually looked pretty clean. I had good reception of the HF broadcasters, parked on some of the lower HF bands and uploaded spots to wsprnet on 160m and 40m. The V3 even decoded a DL2 on 40m on the first decode, along with many stateside. I moved lower into the AM BCB. Had no problem with intermixing... all the channels looked as they should with clean copy. I parked on 530 khz and listened to Radio Enciclopedia from Cuba for awhile. I was impressed!
O.k., now the "acid test"... will it work below 500 khz? I saw no 'net reports about reception that low, but I had to give it a try. I set up for wspr decoding on 630m and by golly it was doing the job! Moving lower, I saw a few beacons and sure enough I copied a few. The lowest beacon freq I copied was 250 khz.
Forgot to mention... below 500 khz, I had to use my LPF inline to cut off >500 khz. I had a lot of "junk" all over the MF-LF spectrum without it.
My opinion on the negative net reports... I believe they come from ops that most likely are overdriving the device. My e-probe doesn't deliver too much signal on its output, which seems to be just fine for the V3's operation.
Below HF, the sensitivity does degrade so a preamp would probably give better results. Of course this doesn't affect the BCB at all since those stations need no amplification. I noted that the stations on 630m that I decoded were down in db level from what I normally copy them at with my other sdr's. Besides the amp, one would need to keep the LPF inline, and a band-pass filter may be needed also. Of course, it's just easier to use a better sdr but it would be interesting seeing what this very neat inexpensive device would be capable of under the right conditions.
Over-all I'm very impressed to see this minimal package with such a broad useable frequency range do so well. Btw, I also previously used it to copy ADBS aircraft at >1 GHZ. I consider the time to set the V3 up and test it out well spent.
Ray