Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

SDR Radio Selection

Status
Not open for further replies.

jpanhalt

Well-Known Member
Most Helpful Member
This past several weeks, I have been playing with low-frequency RF and having a ball. I am in the market for an SDR radio. Its range must extend to 100 kHz and preferably to 60 kHz (WWVB). Not interested in frequencies much about 30 MHz.

Two cheap options I have identified are:

1) SDRPlay RSP (review: https://swling.com/blog/2015/07/a-review-of-the-sdrplay-rsp-software-defined-receiver/ ), about $150
2)AFEDRI (https://www.afedri-sdr.com/ ), about $250

Both of the above have decent reviews. Are there any others in the <$500 class? Any opinions about either of the above?

Regards, John

PS: I returned the Realistic DX-302 off eBay
 
1st looks good but VLF/LF requires good interference rejection on front end to pickup 10uV signals on WWVB .
This includes 10dB notch at 198KHz – to reduce level of BBC R4. Perhaps more is needed. You can simulate log spectrum with Falstad or similar.
**broken link removed**
http://www.g8jnj.net/softwaredefinedradio.htm

In 1974 when I was an EE student and also worked for Earth Sciences dept debugging some Seismic instruments designed by a grad. Physcis student . It used an addon WWVB 60kHZ board for binary clock on one channel of FM cassette recorder with other channel being FM Seismic waves. Dozen units would be deployed with remote explosive impulse generated 20 km to several hundred km away in water filled old mineshaft or below small lake and recordings were automated time delay with spacing in radial locations. so no 100m parallel cable had to be strung from a backpack. We used just a long wire to a branch for the antenna. Data was later uploaded to a PDP8 for data for synchronization and fourier analysis and 3D mapping to 30 miles deep.

Reception was always reliable in Northern central Canada but since you are on other side of pond your signal may vary, but global spreading of LF is good but still requires good front end filters and antenna. This is critical before SDR and can be customized for 60kHz with BPF and notch filters but good lightning signal rejection need to be added before high Q to prevent ringing and prevent momentary LOS from AGC. Global lightning signals are about 10~30 strokes per second.

Diurnal shift has an effect during sunrise/sunset on prop delay is probably not relevant for you unless you need sub-ms accuracy but is it needed for Doppler. WWVB has correction codes.
 
Last edited:
I am Thinking about the use of SDR's with other chips for a matrix of receivers with high resolution time stamps ( xxx ns) for Doppler triangulation for something like locating lightning noise but within a HV substation and then within a transformer tank with antenna. Although this spans the spectrum from audio to gamma rays, some bands are more useful for detection like 1~100MHz, 100~1000 MHz, IR, UV.

I've been toying the idea of advancing the state of the art in Power grid Insulation monitoring, but I can't do it alone.

There is a huge need to replace all old power grid distribution equipment every year continuously now and every generation with contracts in the 50 billion range for each location, but more intelligent monitoring and repairs might reduce overall c osts and extend useful life. Some infrastructure lasts 50 yrs others have early failures within 1 year from flaws in design, materials and installation in spite of improvements.

The best method for detecting HV insulation arcs which indicate flaws or contamination is called Partial Discharge (PD) which is like nano-sized lightning bolts internally that dont short out the grid but cause detonation, chemical reactions to create Hydrogen/Methane and decomposition of insulation from Pyrometry if excessive and repetitive per cycle and creates acetylene if high energy. All measurable by annual oil DGA tests but not soon enough for some equipment that often explodes. That's the risk of ignoring these events. There are over 10k thesis papers on this topic alone in Google or Microsoft Scholar, but most lack real word experience for commercialization unless guided by an OEM instrument manufacturer.

Since it is a trillion $ replacement liability , low cost commercial instruments can detect PD but not locate it unless you walk around with a portable antenna but not inside the unit for factory or field maintenance. 3D location instruments start around $10k and go over the cost of the unit. There is a huge need for low cost detection of these impulses and separating it from common corona which is more repetitive but lower energy.

I have tons of info if anyone wants to get on the bandwagon and collaborate which I can share in a Dropbox. Just suggest what area of interest in this project you have, SDR? Wifi data-collection, 3D location, User interface, business plan or whatever.
The scope can be defined to accommodate the needs of everyone with a plan once interest gains momentum.
 
@Post#1: Thanks for that advice on filters. I was playing with filters, including the MAX275, and that was what made me want a cheap spectrum analyzer. It would be really neat to use a very cheap DDS to scan and the spectrum analyzer to show the results. One of the attractive things about the SDRPlay was its 8 preselection filters; although, I didn't find anyone who commented on performance enhancement by including them. There are much cheaper SDR's based on the RTL2832/R820T chip set, but they generally don't go as low as I want.

@Post#2: I am sure you have heard of a piece of avionics called Stormscope. It is a lightning detector that can be used in lieu of or as an adjunct to radar. Its inventors found that turbulence was more related to lightning frequency than to rainfall intensity. It was cheaper than radar and more adaptable to single engine airplanes than radar was. It provided direction and ranging, but the manufacturer was pretty secretive about how it obtained ranging. Of course, triangulation will work, but it is not really practical for an individual pilot to do.

Thanks for the added info on high voltage arc chemistry. I was not aware high-tension lines were even insulated.

My interest in SDR was mainly for listening to amateur bands and messing with filters. Additionally, receiving WWVB would be a plus, but I just ordered a very cheap receiver for that signal.

John
 
@There are much cheaper SDR's based on the RTL2832/R820T chip set, but they generally don't go as low as I want.

John

Nooelec has an upconverter for cheap SDR with a low end of 100KHz (easily modified to a lower range)

I have the original 100MHz version.
https://www.nooelec.com/store/ham-it-up.html

Noise generator addition to latest board.
 
Last edited:
nsaspook
Thanks.. All I need is a good noise source for my pass band filter. Kind of brought back old memories of doing corrected fluorescent emission spectra. We did not have flat excitation sources and had to take a whole box of Hollerith cards to the computer center to get the final spectrum printed (circa 1966).

John
 
... Hollerith cards...
Ah, the old "Do not fold, spindle or mutilate" days. Had to use those at Purdue (mid to late '60s) and then later in the NAV. Pain.In.The.Butt.

Bet you didn't know that in paper recycling circles (at least in the early '70s), used punch cards were the most valuable.
 
Hmmm...Purdue 1960's. You just missed my thesis adviser, Emil H. White. He did explosives research and nitrogen chemistry, and he had the scares to prove it. Went from Purdue to Yale to my university.

We had a whole room of "card punchers" in the early 70's. Two people for each station. One entered and the second person re-entered the data for confirmation. Very few instruments were on-line even as recently as 1976.

John
 
OT, but...

Remember the Purdue quarterback Bob Griese? Went on to the Miami Dolphins?

First (and only) time Purdue went to the Rose Bowl (1966, I think) and only because Michigan State, which was #1 (Purdue was #2) in the Big Ten (now Twelve), had gone the year before. And you couldn't repeat the very next year. Sports Illustrated had a front page pic of a player in White and Gold. Inside, the first sentence of the article on Purdue started with: "Would you want your daughter to marry a Boilermaker?"

Nice...
 
@Post#2: I am sure you have heard of a piece of avionics called Stormscope. It is a lightning detector that can be used in lieu of or as an adjunct to radar.
John

Not heard of Stormscope but from seeing how it works, it would use a bandwidth around 1MHz rather than up to 1GHz so the resolution is not enough for short distances.

I understand how it works, and have to give the guy full marks for marketability, and huge profits.

The radar data is rather misleading if one were to compare with a satellite representation. I would steer at least 25 miles away from it.

Basically he uses amplitude of the impulse noise to determine distance and directional antennae sample and hold the levels while the software bins them. Location of (+) on the radar scope are irrelevant and just show the frequency of discharges per time interval, which could be a minute or more and thus if traveling at 5 miles per minute or 300MPH in a jet would be offset by at least 5 miles. He shows 7 directions in front which were stated as passive antenna with S&H. So basically it a glorified impulse counter with range and direct determined by signal strength. Effective for detecting lighting after it strikes.

Single events are ignored unless close range.
 
I assumed incorrectly that Stormscope was something you may have heard of and that its technology might have application to your proposed project. It has around for at least 30 years, and it is incorporated into some pretty sophisticated aircraft along with radar. Of course, pilots used the ADF indicator long before Stormscope to get a feeling for the direction of lightning strikes. I had radar in my last airplane, and it has limitations too.

John
 
Can't speak to the sophistication of the Stormscope's electronics, but...
... I would steer at least 25 miles away from it. ...

Flew many times in a Lake amphibious (formerly owned by Jimmy Buffet). It had a stormscope and, at least in this case, was accurate enough to keep us out of the storm clouds that worried our pilot (and as a result terrified me... :nailbiting:). The 'scope allowed us to maintain about a 5 mile skirt, which always seemed adequate.

BTW, it was pretty cool flying around in a plane with an anchor and a bilge pump, but the prop wash BANGED the fuselage like a drum.
 
Last edited:
upload_2015-12-14_19-3-34.png

I read about 2 plane crashes where the pilot flew into bad weather data displayed which was 8 minutes old and showed clear display.
 
One thing I noticed on the SDRPlay receiver is that the band switching is done with some Peregrine CMOS switches PE42641 (See Image). The issue I see is that the switch is only specified down to 100 MHz, but the first 3 bands are < 100 MHz. I don't know if this is a big deal, but thought it was worth mentioning.

So I am not to familiar with SDR, is it like a set of basic RF receiver building blocks like LO, mixers, gain, etc. All user programmable, or basically do you define the parameters such as, IF frequency and BW, IF gain, demodulation format?
Just curious?

rfswitch.PNG
 
Sort of depends, Mike. Most of them are MEMS and have software selectable, individual mechanical oscillators, resonators and filters, plus switches and the like. And (to a lesser and greater degree) concurrently they incorporat solid state amps, mixers, etc.. Their like a whole slew of crystals and LC circuits that you can mix and match to tune the front end of the rcvr to achieve a particular "band" sensitivity.

As to the frequency range, the PE42641 seems to be designed to work best at 500MHz and up.

But the reviews of the SDRPlay clearly indicate it works adequately at the lower (<100MHz) ranges: https://www.eham.net/reviews/detail/12358. Generally, it's considered an "entry level" rcvr platform (no good for SSB, for instance).
 
Martin Ossmann of Germany had designed and published one such design, linked below.
**broken link removed**
please check whether that suits you needs, japanhalt
all the best
 
Thank you Sarma.

That is really quite a nice and complete article.

Regards, John
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top