Airspy SDR by SDR# author

Status
Not open for further replies.

KC1UA

Scan New England Janitor/Maintenance
Database Admin
Joined
Oct 27, 2002
Messages
2,051
Location
Marstons Mills, Cape Cod, Massachusetts
Here's my current status:

Arrived at Sort Facility HONG KONG - HONG KONG Thursday, November 20, 2014 at 00:52 Origin Service Area: SHENZHEN - SHENZHEN - CHINA, PEOPLES REPUBLIC Destination Service Area: BOSTON, MA - BOSTON - USA

Interestingly I don't have a delivery date on mine. I did this directly from the dhl.com website itself.

Edit: Now I see a date of 11/21, but via the iTead order status. Cool!
 

mtindor

OH/WV DB Admin
Database Admin
Joined
Dec 5, 2006
Messages
10,368
Location
Carroll Co OH / EN90LN
Here's my current status:

Arrived at Sort Facility HONG KONG - HONG KONG Thursday, November 20, 2014 at 00:52 Origin Service Area: SHENZHEN - SHENZHEN - CHINA, PEOPLES REPUBLIC Destination Service Area: BOSTON, MA - BOSTON - USA

Interestingly I don't have a delivery date on mine. I did this directly from the dhl.com website itself.

Edit: Now I see a date of 11/21, but via the iTead order status. Cool!

Must be nice. I essentially have gotten through the same tracking steps as yours, however both DHL and Itead show the 24th -- IN TRANSIT: 24 NOV.

If you're arrives on the 21st I'll be supershocked. And of course I'll be pissed if I then have to wait over the weekend. Apparently it'll arrive up in the NE somewhere. That's the only way I could account for the difference in delivery dates. If you're scheduled on a Friday, it only takes one more day to screw up my delivery date and force me to Monday.

I'm not holding my breath on any delivery date until I actually see a DHL waypoint inside the states.

At any rate, I'm not reall upset. I'm just anxious. When it gets here I'll have fun with it for sure.

Mike
 

KC1UA

Scan New England Janitor/Maintenance
Database Admin
Joined
Oct 27, 2002
Messages
2,051
Location
Marstons Mills, Cape Cod, Massachusetts
Maybe it's because I ordered so early. It'll be nice if it gets here on 11/21 but if I have to wait until next week that's fine too. I was more concerned in actually being notified of its shipping and having a tracking number than with the actual day it would arrive here. Not to mention, I'm going to be busier than a one-legged man in a butt kicking contest this weekend so I may not have a lot of time to play with it in any event! :D

I'll post whether or not it arrives on Friday. Given that it's only Wednesday and it's about to get on a plane to Boston, Friday seems feasible.
 

Flatliner

Member
Joined
Aug 10, 2014
Messages
391
Location
UK
Well it was an overall weak signal issue, but I think the primary issue may have been me! I'm having much better fortune with it now. It's doing a pretty good job of what I want it to do so it's probably a keeper.

That said, I'm looking forward to seeing how Airspy stacks up against it. This comparison should be a good topic of conversation going forward although they are in many ways apples vs. oranges.

For me, the HackRF is the "near" ultimate all rounder, with its1-6000MHz range and the 20MHz bandwidth, which is also great for seeing an entire WiFi signal. Being able to use it as a stupidly wide range signal injector also... the Airspy cannot touch any of this, and the HackRF wasn't much more than the Aipspy either. Hopefully Airspy will find its niche though, but it's coming a bit late now, also especially as the R820T2 is appearing in dongles at a tenth of the price of the Airspy.

Glad to hear that you sussed that HackRF finally. What was the issue in the end? By the way, if you haven't seen it, the RF Analyzer is available for Android. I can now stick the HackRF on the back on a Galaxy Note. I used it today, portable. Very impressed to far.
 

KC1UA

Scan New England Janitor/Maintenance
Database Admin
Joined
Oct 27, 2002
Messages
2,051
Location
Marstons Mills, Cape Cod, Massachusetts
For me, the HackRF is the "near" ultimate all rounder, with its1-6000MHz range and the 20MHz bandwidth, which is also great for seeing an entire WiFi signal. Being able to use it as a stupidly wide range signal injector also... the Airspy cannot touch any of this, and the HackRF wasn't much more than the Aipspy either. Hopefully Airspy will find its niche though, but it's coming a bit late now, also especially as the R820T2 is appearing in dongles at a tenth of the price of the Airspy.

Glad to hear that you sussed that HackRF finally. What was the issue in the end? By the way, if you haven't seen it, the RF Analyzer is available for Android. I can now stick the HackRF on the back on a Galaxy Note. I used it today, portable. Very impressed to far.

I think I "just got it" with the HackRF after a while, and I'd guess this began when I finally took the plunge and did the firmware update via Pentoo, which was actually quite easy. My main issue was likely just finding the right balance of the various controls involved. I find that it works far better for me in SDR-Console than it does in SDR#. I do love both software packages though, and of course out of the gate Airspy will be limited to SDR# for Windows users until Simon gets his and fits it in.

I'm hoping the Airspy will be a good receiver and that the aliasing/imaging claim will hold true. It is 12 bit vs. HackRF's 8 bit, and while the aliasing and imaging does seem to be reduced in the latest HackRF firmware there are still issues. It's nice to have 20 MHz wide to view, but in all honesty I find myself using that large of an FFT rarely. The Airspy is also built by someone more interested in RF reception and less interested in hacking, as is the primary use for the HackRF. Finally, I really have no interest in anything above 2 GHz right now anyway, so the Airspy's coverage for me is fine. Oddly enough the odd SDR out in my case may be my bladeRF. It frustrates the crap out of me with aliasing and imaging and will be the only one of the three that does not have a protective case that houses both the main board and add-on transverter. If Airspy stacks up as I hope it does, the bladeRF will get sold at a "nice price" so it has a happy home. :D
 

jdobbs2001

Member
Joined
Apr 20, 2005
Messages
221
Curious what makes this airspy better than the nooelec mini2 since both have the same reciever?

Other than the bigger baseband, what is important for me is how sensitive, dynamic range and noisefloor.

The 820Tv2 is the same on the 26 dollar noelec. So the radio side should offer the same quality.
 

PACNWDude

Member
Joined
Oct 15, 2012
Messages
1,336
SDR# and dongle game changer for me.

I'm waiting for both it and HackRF to become available. It will be nice to be able to see 10 MHz and 20 MHz of spectrum at a time with more robust devices. All of these devices are game changers and are definitely a large part of the future of this hobby.

SDR# and the dongles that use it have been total game changers for my peers. I work in an emergency response communications role, where many of us have send thousands of dollars to Uniden over the years. A year or so ago, we began sending money to SDR# instead.

Vetting out a MotoTrbo system that was about to be put into place, finding signals in the local area, demodulating practically everything, has become easier as a result.

I am able to use an Asus EEE PC 701 and SDR# with a USB dongle to act as a scanner, while having internet access with a Verizon dongle.

+1 on it being a game changer.
 

Flatliner

Member
Joined
Aug 10, 2014
Messages
391
Location
UK
Curious what makes this airspy better than the nooelec mini2 since both have the same reciever?

Other than the bigger baseband, what is important for me is how sensitive, dynamic range and noisefloor.

The 820Tv2 is the same on the 26 dollar noelec. So the radio side should offer the same quality.

The other main chip, the ADC, which converts to the analogue signal to digital, does it with 12 bits, rather than 8 bit. This should increase the dynamic range, the SNR, hence deal better with strong signals a little better. Also the device will use the full frequency width of the R820T2. The dongles don't utilise it, in the debug mode that is uses to get I/Q.
 

woodpecker

Member
Joined
Aug 7, 2005
Messages
729
For me, the HackRF is the "near" ultimate all rounder, with its1-6000MHz range and the 20MHz bandwidth, which is also great for seeing an entire WiFi signal. Being able to use it as a stupidly wide range signal injector also... the Airspy cannot touch any of this, and the HackRF wasn't much more than the Aipspy either. Hopefully Airspy will find its niche though, but it's coming a bit late now, also especially as the R820T2 is appearing in dongles at a tenth of the price of the Airspy.

Glad to hear that you sussed that HackRF finally. What was the issue in the end? By the way, if you haven't seen it, the RF Analyzer is available for Android. I can now stick the HackRF on the back on a Galaxy Note. I used it today, portable. Very impressed to far.

The HackRF seems to have pretty poor reviews and a very mediocre spec the only thing it has over the Airspy is a wider bandwidth and frequency range but if that bandwidth is full of spurii and clock spikes then its useless.

I think your personal issue with Yousef has clouded your judgement, I think the Yousef needs to brush up on his communication skills but he's written a damn fine piece of software and hopefully by this time next week I'll be reporting he's designed some fine hardware to complement it.
 

Flatliner

Member
Joined
Aug 10, 2014
Messages
391
Location
UK
The HackRF seems to have pretty poor reviews and a very mediocre spec the only thing it has over the Airspy is a wider bandwidth and frequency range but if that bandwidth is full of spurii and clock spikes then its useless.

I think your personal issue with Yousef has clouded your judgement, I think the Yousef needs to brush up on his communication skills but he's written a damn fine piece of software and hopefully by this time next week I'll be reporting he's designed some fine hardware to complement it.



My opinion of Youssef hasn't clouded my judgement, and it's far from only my opinion either. Whereas SDR# is a useful program, and hopefully Airspy will be a good radio, Youssef himself, is rather boorishness and crass. He's no different than a couple of the "specialists" who my clients have asked to leave site because of their inability to communicate with the staff.

What HackRF needs is preselection. When I use filters with it, and work within its limits, I've found the device to be very usable indeed. And the "only things" do mean that regardless, the HackRF can go places that that Airspy simply misses, 2.4GHz and 5GHz. That's WiFi, Bluetooth, Zigbee....

I will probably get an Airspy eventually, but I'm not going to be the reason for another sale.
 
Last edited:

SCPD

QRT
Joined
Feb 24, 2001
Messages
0
Location
Virginia
Curious what makes this airspy better than the nooelec mini2 since both have the same reciever?

The tuner chip is the same. Both devices require a USB 2.0 port. That's the extent of similarity.

The RTL2832U baseband chip has an 8051 microcontroller core that runs at 28.8 Mhz. The AirSpy uses a Coretex M4 processor running at 204 Mhz.

The RTL2832U converts the IF to 8 bit I and Q samples by running dual ADCs side-by-side. Slight differences in the ADCs result in the DC spike at zero hertz that burns a hole in your waterfall.

AirSpy uses a completely different mechanism to derive an analytic IQ signal from the raw IF. It runs a single 12 bit ADC at 5 or 20 msps. This single in-phase only stream of samples is passed on to the host PC. The host PC driver converts the in-phase only signal to IQ sample pairs at half the raw sample rate (eg. 10 or 2.5 msps). Anti-aliasing filters limit this to about 9 Mhz or 2.25 Mhz of usable spectrum.

Since the IQ is derived in software - there's no ADC mismatch that causes the zero hertz DC spike and no IQ correction is needed.

Most of the RTL2832U sticks out there use a consumer grade clock reference that is off by 50 ppm (or worse) and may drift several ppm as room temperature changes. After-market resellers offer upgraded RTLs with a 2 ppm of better TXCO for $90 USD. That's almost half the price of an AirSpy (and less than a third of the bandwidth and two-thirds the sample depth).

The AirSpy uses a clock reference rated for 1.5 ppm. I don't remember what the exact range is for thermal drift but it is very small.
 
Last edited:

Flatliner

Member
Joined
Aug 10, 2014
Messages
391
Location
UK
The Nooelec RFI screened version is only $70 with a 0.5ppm TCXO.

I solved the drift issue with a TC heater on each crystal and have several dongles in a screened case.
 

mancow

Member
Database Admin
Joined
Feb 19, 2003
Messages
6,879
Location
N.E. Kansas
Mine shows it just arrived in Cincinnati. I hope it arrives tomorrow and lowband is hot again. The higher sensitivity will be fun to play with.
 

prog

Member
Premium Subscriber
Joined
Nov 18, 2014
Messages
73
The tuner chip is the same. Both devices require a USB 2.0 port. That's the extent of similarity.

The RTL2832U baseband chip has an 8051 microcontroller core that runs at 28.8 Mhz. The AirSpy uses a Coretex M4 processor running at 204 Mhz.

The RTL2832U converts the IF to 8 bit I and Q samples by running dual ADCs side-by-side. Slight differences in the ADCs result in the DC spike at zero hertz that burns a hole in your waterfall.

AirSpy uses a completely different mechanism to derive an analytic IQ signal from the raw IF. It runs a single 12 bit ADC at 5 or 20 msps. This single in-phase only stream of samples is passed on to the host PC. The host PC driver converts the in-phase only signal to IQ sample pairs at half the raw sample rate (eg. 10 or 2.5 msps). Anti-aliasing filters limit this to about 9 Mhz or 2.25 Mhz of usable spectrum.

Since the IQ is derived in software - there's no ADC mismatch that causes the zero hertz DC spike and no IQ correction is needed.

Most of the RTL2832U sticks out there use a consumer grade clock reference that is off by 50 ppm (or worse) and may drift several ppm as room temperature changes. After-market resellers offer upgraded RTLs with a 2 ppm of better TXCO for $90 USD. That's almost half the price of an AirSpy (and less than a third of the bandwidth and two-thirds the sample depth).

The AirSpy uses a clock reference rated for 1.5 ppm. I don't remember what the exact range is for thermal drift but it is very small.

I'd add the RTL chip resamples the IQ stream like a hog by dropping samples irregularly. This allows for more samplerate flexibility but also causes some extra phase noise even if the input clock is clean. Given the very wide DVB-T signals it's supposed to handle, this is acceptable but it's a whole different story when tuning narrow band signals. This aspect is shared with the SDR Play RSP also and the only way to do it right is using a DDC in a more complex ASIC like the AD chips or in a FPGA and none are cheap. It ain't no free lunch.
In the other hand, Airspy uses a high quality DDC in the host (PC) to generate the base band IQ signal. This gives the opportunity for proper decimation and resampling, adding more resolution at every stage. So, the problem is shifted to the PC side and cracked there.
Bottom line, Airspy is an hybrid approach that takes advantage of the heterodyne architecture, a DDC and some high quality hardware, adding more S to SDR in the way.
 
Last edited:

KC1UA

Scan New England Janitor/Maintenance
Database Admin
Joined
Oct 27, 2002
Messages
2,051
Location
Marstons Mills, Cape Cod, Massachusetts
Mine shows it just arrived in Cincinnati. I hope it arrives tomorrow and lowband is hot again. The higher sensitivity will be fun to play with.

Mine shows in Cincinnati as of 0058 Eastern Time this morning. I'd be surprised if it made it here today,but it still says 11/21 delivery, so we'll see.
 

KC1UA

Scan New England Janitor/Maintenance
Database Admin
Joined
Oct 27, 2002
Messages
2,051
Location
Marstons Mills, Cape Cod, Massachusetts
OK, so I'm surprised. Mine just arrived at work! Report to follow asap!

Edit: A few unboxing photos.
 

Attachments

  • WP_20141121_002.jpg
    WP_20141121_002.jpg
    40.5 KB · Views: 1,074
  • WP_20141121_003.jpg
    WP_20141121_003.jpg
    29.8 KB · Views: 1,749
  • WP_20141121_005.jpg
    WP_20141121_005.jpg
    40.5 KB · Views: 1,042
  • WP_20141121_006.jpg
    WP_20141121_006.jpg
    22.9 KB · Views: 1,059
Last edited:

mancow

Member
Database Admin
Joined
Feb 19, 2003
Messages
6,879
Location
N.E. Kansas
A marked improvement in sensitivity. I'm listening to lowband right now and had previously been using the HackRF. The Airspy's noise floor shows right about -75. Signals are very strong and I haven't found any false peaks or mirrored images yet. So far it's working extremely well.
 
Last edited:

garys

Member
Premium Subscriber
Joined
Jun 13, 2002
Messages
6,069
That looks good. It even comes with a real antenna. I just hope they do a second production run.

OK, so I'm surprised. Mine just arrived at work! Report to follow asap!

Edit: A few unboxing photos.
 
Status
Not open for further replies.
Top