LimeSDR Now Backed by the European Space Agency | Hacker News Reader

LimeSDR Now Backed by the European Space Agency

rtl-sdr.com

122 pointswalterbell38 comments

LimeSDR vs mini: http://qrznow.com/limesdr-mini/

”the LimeSDR Mini is a smaller, less expensive version of the original LimeSDR ... uses the same LMS7002M radio transceiver as its big sibling. The Mini has two channels instead of four, and, by popular demand, SMA connectors instead of micro U.FL connectors.”

A positive review (2016): https://myriadrf.org/blog/first-tests-limesdr-gqrx/

A negative review (2017): http://sv3exp.blogspot.com/2017/03/limesdr-review.html?m=1

I bought one (the LimeSDR) and based on that experience declined to buy the mini. If I had to summarize the mismatch it was that MyriadRF, the folks who are the actual developers behind the systems, had a decidedly windows/microwave focus. If you wanted to build an LTE simulator and use PothOS on Windows there were people to talk to, but gnuradio and Linux support was hit or miss. That was made particularly challenging by the LimeSDR folks modifying and building tools in their own PPAs which had incompatible changes/libraries with the 'mainline' sources. As a result I could build a working gnuradio setup that only had LimeSDR support (through SoapySDR) or one that supported the other three SDRs I had simultaneously. That was really painful.

The 'microwave' focus was particularly apparent when, by default, there was no reasonable way to listen in on things below about 200Mhz. (all the FM band for example). They later offered an HF mod which was to remove an inductor(coil) from the front end selector on one of the antenna inputs to give it better frequency response.

Even with the change I found a number of systems with better performance in the low bands. It isn't a deal killer but it means if you're going to be exploring the low bands you probably want an upconverter.

>Even with the change I found a number of systems with better performance in the low bands.

Such as? I'm not aware of anything in the price arena of the LimeSDR that can Tx.

If you're transmitting the RSP2 (mentioned earlier) is about $100 cheaper and can transmit in all of the lower spectrum. HackRF works well as a transmitter in this spectrum as well where its dynamic range isn't the limitation like it is when receiving.

You can add a down converter of course and shift spectrum where the SDR is comfortable playing into the lower bands, but if you're going to do that you might as well get the PLUTO-ADALM from Analog devices which is only $150 (I got mine on the intro deal of $99 but still).

That said the LimeSDR excels at MIMO type applications (as that really is its design target).

It's important to point out that the two reviews listed from 2016 and 2017 are for the previous version of the LimeSDR, not the LimeSDR Mini which was just recently released.

What hw/sw changes were made between versions?

It's a very different architecture. The original LimeSDR was able to receive from 100kHz - 3.8GHz. The LimeSDR Mini can only do 10MHz - 3.5GHz. It also only supports about 1/2 the bandwidth and sample rate.

The official page highlights more differences: https://www.crowdsupply.com/lime-micro/limesdr-mini

The original LimeSDR was able to receive from 100kHz - 3.8GHz.

This is incorrect, and it is the source of much pain. The chip that provides the ADCs and DACs, the LMS7002, has an input range of 100kHZ to 3.8Ghz but the design of the LimeSDR front end filter banks made its minimum usable frequency with decent SNR about 200Mhz[1]. Initially you couldn't even pass a tuning frequency below that through the API.

Even with the vendor/factory supplied "fix" the LimeSDR has a hard time getting below 100Mhz.

A lot of people read the chip spec and expected the card implementation to be able to do the same, they were unpleasantly surprised to find out that was not the case. Hence a lot of griping early on. On the plus side, since it was deaf to the FM band it was more sensitive to higher frequency signals. Other SDRs like the SDRPlay RSP2[2] are actually quite usable down even into the AM range (500Khz - 2Mhz) and have a notch filter for the FM band to keep it from swamping the input and eating up all the dynamic range.

[1] Firstly, we would like to highlight the fact that it is not possible to achieve optimum performance across the entire operating range of LimeSDR — all the way from 100kHz to 3.8GHz — with only a single matching network. -- https://myriadrf.org/blog/optimising-limesdr-matching-hf/

[2] https://www.sdrplay.com/rsp2/

Interesting I wasn't aware of the filter issue with the original LimeSDR. The lower range was one reason why it was on the top of my list of something to pick up for when I've got a proper reason to need a full featured SDR, rather than just the mostly toy RTLSDR based stuff. That said I think it's still up there for the high input bandwidth (60MHz woo!) for me. Unlikely to ever need real transmit ability but being able to receive that much is a nice feature.

Agreed, the bandwidth is nice, but most of the signals you would play with in the lower bands have much lower bandwidth. Digital TV (like DVB_T) for example is only about 8Mhz. Generally in the lower frequencies taking up more than a few megahertz of bandwidth really messes with the available spectrum. Up in the microwave region (2Ghz+) that is not as much of a concern.

Both devices use the same transceiver chip so I don't think the architecture is very different at all.

Same transciever, different FPGA, different ADCs, no extra ram, and some other differences. Different architectures on the digital sides of things.

Not quite sure how unbiased of a source the positive review is. Myriad is owned and fully controlled by Lime.

Anyway, I'm not quite sure what ESA will be able to do that Canonical couldn't. This whole app store idea for radio seems very farfetched.

Other than the basic SDR projects like tracking aircraft and decodimg garage door fobs, what cool projects are people using this hardware for?

The thing to note about the Ettis and Lime devices is that they're good for so much more other than just moving I/Q to your computer. They have powerful FPGAs that make it possible to do filtering and decoding and such in hardware.

If you're just looking to play around with radio, you don't need a device with these features -- not until you're much further along the learning curve. For any practical usage of these devices, it's assumed you have the wherewithal and radio background to build the pieces that it lacks; filtering, amplifiers, antennas and feedlines, and so on.

This is very much a professional device, and it's not meant to replace the common usage of $20 rtlsdr dongles within the hobbyist radio community.

Funny you say that - those are literally the two thing I've done with my SDR!

Tracking aircraft is pretty fascinating though, if you live near a reasonably busy airport (as I do). Decoding random remotes is also useful - I bought a couple of "Emergency Alarm Buttons", which are very simple battery-powered buttons. I can detect their press and my PC can then stop music, shut off lights, etc.

I've recently learned that the local displays on bus/tram-stops are updated via radio, so my next project is attempt to receive and decode this data. Getting real-time information about buses & trams arriving at the stops near my house. (I can already do this, and have a toy project with an ESP8266 device + LCD screen in my hallway, but that uses the Web-API. I like the idea of using radio-alone.)

Came here to ask this. I'm an extra-class ham but I've never done any SDR. Are these suitable for regular ham use for voice and digital modes? (assume you'd want a big amp)

Yes. See my video here: https://www.youtube.com/watch?v=W_o8sOZHoeE&feature=youtu.be Software is http://qradiolink.org built with GNU radio.

For an extra class ham interested in HF more than UHF/VHF I'd recommend the Airspy HF+, it's RX only, but quite good at RX.

the limesdr is not good for HF though. other sdr's are better.

I used LimeSDR at work. We were nervous to buy from Crowd Supply, but they are really a good value add. Compared with something like Kickstarter they inventory the products and handle returns- something a tiny startup might not deal with well.

LimeSDR was convenient for us because the LMS7002M has two radios, each with I & Q plus 3:1 input switch (so 6 inputs total).

Question: can I use LimeSDR and some SDR software to save on hardware and not get a Yaesu or Icom ham radio?

I wanted to play with ham radio for a while but I am not sure I want to spend 750+ Euros for a Yaesu station.

It would be a good proxy for those radios, but if you're licensed and are simply looking for an inexpensive transceiver, Baofeng UV-5R radios can be found on Amazon for ~$20 and will get you working right out of the box.

If you try to mock up a station using a TX/RX SDR system, it would probably much more expensive and require significantly more hardware.

He's probably looking for an HF rig -- baofengs definitely don't do HF. As someone else said, the LimeSDR would work well, just need an amplifier because the power out is on the order of milliwatts.

> He's probably looking for an HF rig

Indeed.

Absolutely, but you'll probably need an amplifier as these things tend to be ~1mW-10mW output power.

LimeSDR is not that great for HF work - the chip is really designed for cellular frequencies. Although you can technically tune it to lower HF freqs, it really struggles at that range - there is lots of evidence on the forums that demonstrate this.

Yeah, something like uBitx[1] + amp may be a better combo.

If you're interested in digital modes than 10w is usually enough for PSK, JT* and FT8.

[1] http://www.hfsignals.com/index.php/ubitx/

Check out the SDRPlay RSP2[1]. It is less expensive and covers 1khz to 2Ghz. It comes with nice radio package (SDRuno) as well.

[1] https://www.sdrplay.com/rsp2/

I've been dreaming about a project where a full-duplex amateur satellite LEO ground station could be based around a limesdr with a daughterboard, but that'd be for 1.2G / UHF / VHF and I lack the time, energy, money, and expertise to bring it to reality. Depending on what the IC-9700 price ends up being, I think it could actually be cheaper to roll-your-own radio for this purpose than to buy something off the shelf.

However, for HF (and at this point in the solar cycle, where anything less than 50W is wasting time), it's almost certainly going to be a cheaper and better experience to pair a used 100W transceiver with something like a RSP1A [1] instead of trying to roll your own. This is what I do; I've tapped the IF on my FT-950 and pair it with a rtlsdr for the purpose of having a waterfall. Being able to "see" the bands makes a huge difference.

---

[1]: https://www.sdrplay.com/rsp1a/

> However, for HF (and at this point in the solar cycle, where anything less than 50W is wasting time)

QRP is still very much possible, I don't have any problems with CW, FT8 or PSK31 on my KX3(10W) with a simple end-fed halfwave antenna. It's all about using the right mode for the band conditions.

For voice 100W is still nice to hav but I wouldn't call not having 50W wasting your time.

Of course! I should choose my words more carefully. My point wasn't to say that QRP is pointless, most of the time I spend on radio nowadays is <25W. It's just a struggle, but the reason that you and I seek out that struggle is because we've probably both made plenty of contacts, and we're in it for the sake of being the best operator possible and seeing what we're capable of with limited resources, not for chasing DX with big beams and legal-limit power. For us, a 5W contact with less distance is worth more than a 100W contact that goes halfway around the globe.

That's not how I felt when I first got into radio, though, and I wouldn't recommend that new hams go straight for QRP operation. You're just going to get frustrated.

(PS: How do you like your EFHW? My father has one and we have been really underwhelmed, noise is just unbearable compared to his vertical. For portable operation it might be okay, but I think he's going to replace it with an off-center fed dipole. I'm jealous of your KX3, I've been eying that new Xiegu X5105 and the KX2. Something portable opens up a lot of fun operating scenarios.)

-- 73 DE W9NLS

That's the thing is that everyone has a different idea for ham radio. I went straight to QRP and the digital modes and rarely touch voice(except to chase some SOTA activations).

The EFHW does well enough for me but I'm out in the country so I'm probably fighting a lot less noise. 20m is usually S3 and 40m S1-S2. Love the kx3, if I didn't want to do SDR stuff with the IQ output I could have probably done fine with a kx2.

You'd also probably need an upconverter like the HamItUp (which works for the HackRF).

If you're just using a standard RTL-SDR, then yes, an upconverter like HamItUp or SpyVerter would work.

However, the LimeSDR Mini supports down to 10MHz out of the box, which is "good enough".

If you really need to get the 160, 80, 60, and 40 Meter bands then the converter would be needed, or you can just use the previous-generation the (full size) LimeSDR, which can receive down to 100KHz.

Anyone know how the LimeSDR Mini compares to the LimeSDR or HackRF?

The most important difference is that that HackRF has a huge ecosystem of established software now than it's been a handful of years since the kickstarter.

With crowdfunded SDR devices expect a 1.5-3 year lag between hardware release and the software/firmware being there to do what it advertises.

The main CrowdSupply page has a very good comparison table: https://www.crowdsupply.com/lime-micro/limesdr-mini

Reply on news.ycombinator.com