The AM Forum

SDR kit, no sound card stuff, its built in.

http://ae9rb.com/index.php?main_page=product_info&products_id=8

Something play with at a low price.

I received it today and I am about 1/2 way done building it.
This is NOT an easy kit, lots of stuffing a lot of very fine wire into very small holes, and its almost all surface mount stuff, some VERY small stuff.

2 band (any 2) kit, but its supposed to work well.
 

I was going to order one but it was back ordered.

Let us know if that works with sdr-radio SW on RX?  It looks like it uses the si570 and softrock system so there is a good possibility it will work. The sdr-radio SW is real nice and easily allows remote operation but is RX only.

For an extra hunnert bux you can get the Afedri radio, which more or less emulates a SDR-IP and will do up to 2 MHz direct sampling bandwidth.  Interface over ethernet or USB, so you can plug it into a wireless router and have the radio available on any computer in the house.  This radio is pretty much independent of the sound card and it worked OK in a 7-year old laptop.

I just got the version with 2 front ends.  It works very well; the noise figure is a bit high but it does copy anything my other receivers can hear.

It comes out of Israel; took 4 days for delivery.

http://www.afedri-sdr.com/

Looks very interesting. The web page is a little confusing though. Do You have the

AFEDRI SDR Network Rev. 3.0a ?

Fine on it working over local Ethernet.
Will it work remotely with the SDR-radio without the use of a local computer?

Israel rig looks like little bro. to the QS1R ( real ADC) less a FPGA but with a pre amp.
Looks like a lot of fun for the money.

Yes, apparently.  I haven't done it yet.  Still messing around with it.

It acts as a DHCP host, or you can turn that feature off.  I think most anything you can do with an RFSpace SDR-IP you can do with this one although it's not 100% emulation.

The dude that makes them is a Russian living in Israel so his English isn't 100%.

Hi 'DTS and 'JN,

   How do you compare these two kits to the transceiver offered by FiveDash?

http://fivedash.com/index.php?main_page=product_info&cPath=1&products_id=7

I've been considering the Fivedash kit and at $90 it is attractive.

Thanks,
Ed

No idea.  Afredri isn't a kit nor is it a transceiver.

I didn't notice that the Peaberry was an xcvr, I thought it was a receiver.  Sorry for the confusion.

Its a 14 bit box, and I am not sure what software you can run with it.
The sdr-iq is a 14 bit radio, but it runs with 3 programs, which is nice.

Well, you need a good sound card for those to work into.

Looks like the AFEDRI will work with a bunch of different sw. From the web site:

The Peaburry V2 needs no sound card, and it has a 5 watt transmitter built in.

I got the receiver done today, and its ok, but not great.
Not easy to build at all, nor is the software easy to get working if you are not a geek.

The sound card is on a chip, and I suppose its not a fantastic sound card, which limits the spec's.

The radio works with hdsdr which is a good receiver program, and some have got it working with power sdr IQ. I think the hard part is getting the program to talk to the controller chip to work the LO and relays and so on. The peaburry uses the typical softrock kit chip so it should be easy, for a geek.

I thought I would try it for the money, but it does not work as well as my sdr-iq.

When it comes to sound card based radios,the sound card is real important, and a good sound card costs much more then this kit does...

As far as the kit goes, the parts are microscopic, many are unmarked, and there are transformers to build using very small cores, with small holes, that you need to stuff with very fine wire (#30).


 

I was wrong, its a 12 bit radio.
I do not see how a 12 bit radio can work very well.

Dynamic range must be very limited.
 

Why?

  Brett,

   Receivers all have a finite dynamic range. One way to overcome this is too have 1 or two stages of switched attenuators, and 1 or two stages of switched preamp. If each step was 20db, then we can shift the working range limits up or down 20 + 20 or 40 db. We then might have a problem when we need to receive that .5 uv signal that is 10 Khz from an S9+60 signal.

  I am not a receiver guy or a mathematician, but 12 bits means 2^12 or A/D counts of 0-4095 range. That range is about 72 db since 20 * Log10 of 4096 = 72.... Is that an accurate way of looking at this? I might have that wrong... ::)

  If that is correct, then 72 db dynamic range could be limiting without at least a select-able front end attenuator. On the lower bands, the noise floor should be high enough such that MDS is not a big factor.

Jim
Wd5JKO

In the best case, it's about 6 db dynamic range per bit. In reality, it's usually less. For most hobby and amateur applications, no one will notice the difference between 12 and 14 bit converters. The AFEDRI has a voltage controlled amp in front of the ADC, so they are trying to do what JKO pointed out.

12 bits is very poor, the good ones are 16, and the govt has stuff much higher.

I am no expert, but the way I understand it is you sample the RF and encode it into digital, and the more steps
(bits) gives less noise and distortion, and a wider dynamic range.
If you have 12 bits, the steps will be much corser, and/or limited to a narrow range.

In digital encoding for voice (phone calls) its 8 bit, with the steps in the center (-13 dbm) finer then at the lower and higher levels to get lower distortion.

SDR encoding also drops or does not use a bit in some setups for some reason (I forget) so a 12 bit might be an 11 bit radio.

More is better, 14 is not bad, a little bit of noise and dynamic range limitations, preamps and attenuators can make up for it, but you have to change them for the situation.

The problem with that is some sdr's sample the entire spectrum, so if you want to pull out a weak signal on 15 meters and have a strong signal from an AM broadcast station close by, you need to attenuate that signal.

Some radios have band pass filters in the front end, but that can limit the band scope or skew the amplitude response.


I have the 14 bit sdr-iq, which has a preamp and attenuators, plus band pass filters, and it works very well, but I would like a 16 bit radio that runs the software I like.
For ham radio, I like sdr-console, and hdsdr is also good (nice display).
psdr is also good, but sdrmaxv was not very good for me.

A 16 bit radio can get close to the best analog receivers, and the filtering in almost all the sdr's is much better then any analog receiver....with no limits in the fidelity....

My homebrew receiver is a little bit better then the sdr-iq in the noise department, giving better cleaner audio out on weak signals, that is the 14 bit resolution I suppose. That varies a little with the sdr program I use on the sdr-iq.

The dynamic range of the Afedri is fine, using it in the congested SWBC bands.  There's some bible beater on 60M that peaks at -12dBm on the Afedri and 20 KHz away I can copy Radio Rebelde FINE BUSINESS OM which is between -50 to -70 dBm, or much weaker signals about 50 Khz away with no cross-mod detectable.

Its one weakness is it has no bandpass filters so if you crank up the front end gain you end up getting a bunch of crap.  The FE gain is adjustable from -10 to +34 dB.... any untuned 34 dB gain amp is gonna have issues.

I think the LO in the Afredri probably isn't the best, there's still some hiss in the audio even on strong stations.

In any event it's an inexpensive SDR that has many of the features of those 3X the price.  Helps you can skype or email the designer and get an immediate response on questions.  The documentation isn't great but that's true of ANY SDR (except perhaps RFSpace products), particularly those TAPR HPSDR projects that pre-suppose a huge vocabulary of funky board names and other jargon.  Perhaps those TAPR radios are great, I don't know, I can't even begin to understand the BS and vernacular.

Provided the total, broadband signal (i.e. the sum of all the RF signals that fall within the passband of the front end) at the input of the A/D converter is not larger than the maximum voltage the A/D converter can accommodate without overload... the effective number of bits per sample in the A/D conversion process is equal to the number of bits per sample (11 bits in this case, if you don't count the +/- sign bit) + 0.5 log base 2 [one half the sampling rate / the bandwidth of the signal you are trying to extract]

For example, if you are trying to listen to (i.e. to extract using digital filtering) a signal whose bandwidth is 10kHz (i.e. +/- 5kHz around some carrier frequency), and if the sampling rate is 80 MSPS... then the effective number of bits per sample... as it impacts on the quantization noise in that 10kHz band that is caused by the sampling process... is 11 bits per sample + 0.5 x log base 2 [0.5 x 80MHz / 10 kHz] bits per sample = 11 bits per sample + 0.5 x log base 2 [4000] bits per sample = 11 bits per sample + 6 bits per sample = 17 bits per sample (not counting the +/- sign bit), or 18 bits per sample, including the +/- sign bit.

Saying this another way:

The ratio of the amplitude of the largest broadband signal that the A/D converter can accommodate (in volts) to the rms quantization noise in a 10kHz bandwidth at the output of a digital filter that selects out that 10kHz bandwidth is 10.79dB + 20 log [ 2**17] dB = 10.79dB + 20 x 17 x log [2] dB =10.79dB + 17 x 6dB = 113dB. The extra 10.79dB corresponds to -10 log [1/(12)]... which takes into account the fact that the difference between the actual value of the signal being sampled and the nearest sampling level will be randomly distributed between +/- one half the spacing between the levels.

Yet another way to see this is as follows:

If you sample a 10kHz bandwidth signal at 80 MSPS (4000 x the Nyquist rate), then you can, in effect, average groups of 4000 samples to reduce the effective quantizing noise by 10 log(4000) = 36dB compared to an A/D converter which samples at the Nyquist rate of 20 kSPS. A 36dB reduction in quantization noise is equivalent to having 6 more bits per sample.

Stu

I have no idea what Stu said, but I know more bits gives less noise and distortion, and a wider dynamic range.

I have had a flex 5000, two of them actually, a 3000, a qs1r, the sdr-iq, and the Peaberry V2.
The 5000 is a very good receiver, as is the qs1r.
The 3000 was also quite good, the sdr-iq is ok, and the Peaberry is not great.
(I got the TX part of the peaberry done tonight)

My test is to unplug the antenna with all the gain and agc stuff set normal and listen to the noise.
The homebrew is silent. The flex stuff and the qs1r is quiet with some noise, the sdr-iq sounds like the antenna is connected but with no signals on the band, the peaberry is louder then that.
Both the sdr-iq and the peaberry have some noise on moderate signals, the sdr-iq its slight, the peaberry its louder.

Update:
I corrected an issue with noise from the laptop power supply getting into the radio, and it works VERY well indeed.
If you get the kit, be sure to check that you do not need to ground things, like the laptop/computer, the negative power wire, and use snap on noise filters on the USB cable.
Doing that got my noise level 20 db lower, to where its very quiet and sensitive, it picks up lots of signals without the antenna switched in!

I was fooled because the sdr-iq does not pick up the noise, its USB powered, and in a metal box, so maybe that has something to do with it.
A metal box might be a good idea if used in the ham shack...

Outstanding performance for the price!
 

He said when you over sample, you effectively get more bits. This is how many CD players used a 1-bit A/D and still achieved something like 100 dB dynamic range.

Thanks for the translation!

In professional recording, 16 bit was the standard (is) and the CD is known as the red book standard. Most of the playback systems we have used for digital audio for the past 25 years is based on 16 bit resolution. This is also known as word length. The difference between 16 bit (CD) and 24 bit (DVD) is night and day. The standard sample rate is 44.1 khz for CD, and 96 khz for DVD. For my own use in the studio, I use 96/24. In this environment, the audio is much more natural sounding and has more openness, better dynamic range. To make things even better, the quality of the clock is of utmost importance... big time. All of this is analogous to this discussion in terms of audio. The general rule for all of this is no matter how high your sampling rate, the word length will dictate the audio quality.
Higher sampling rates require more processing power too which is kind of a running joke in the industry...
"Oh yeah, let's record at 192khz @ 24 bit and then insult the process by down converting to 44.1/16 bit CD".
What I am saying here, is that no matter your sampling rate, your word length is the most important part of the equation. 8, 12 or 14 bit audio sucks and is useless to me. You guys with better knowledge of Digital RF design can chime in with how this works as I related here. Perhaps explain how this works by the time it hits our ears. Audio is audio and it matters how you "slice it".
73 de Billy N6YW

Food for thought on the subject of D/A and A/D conversion:

Consider an inkjet printer which has only a black ink cartridge. It can print black dots (ink) and white dots (no ink). So it is basically a 1-bit D/A converter.

Now let's make the dots very small in diameter compared to what someone looking at a printed page from a normal reading distance can resolve with his eye (a high pixel/inch number... which is analogous to a high sampling clock rate)

Next, in a given area on the page, print out (alternately on a dot-by-dot basis) 50% black dots (ink) and 50% white dots (no ink). When viewed from a normal distance (i.e. without magnification), that area will look grey. By changing the percentage of black dots between 0%, 12.5%, 25%, 37.5%, 50%, 62.5%, 75%, 87.5%, and 100% you can print 8 shades of grey (including white and black).

Therefore, you can achieve an effective D/A conversion of 3 bits per pixel with a 1 bit per pixel printer... provided the printer prints dots that are 1/8th the width (same number of lines per inch) vs. what you need for normal viewing resolution.

Likewise, a 12 bit D/A converter (with a suitable low pass filter at its output) could be used to produce an analog output signal that is equivalent to what a 16bit D/A converter would produce... if the sample rate of the 12 bit D/A converter is 16 (i.e. 2**4) times the sampling rate of the 16 bit D/A converter.

With respect to A/D conversion, the use of "dithering" can produce the same effect. I.e. sampling a printed page with a sensor that can only distinguish between two light levels (1 bit A/D)... but with a very high number of samples per inch... can be used to produce a higher equivalent number of bits of amplitude resolution by adding an optical "dithering" signal to the light being produced by the material on the page.

By analogy... the out of band portion of the total broadband signal falling on the high speed A/D converter corresponds to a "dithering signal" that adds to the narrowband signal we are trying to receive... at the input of the high speed A/D converter.

Stu

You can increase the dynamic range of a system by oversampling and dithering even if the A/D does not have the dynamic range, this was used with radar 10 to 20 years ago when the best high speed A/D converters were 8 bits.    Usually the signal riding on noise provides a natural dithering and the higher speed of the A/D provides for the oversampling or signal integration.   When coherently integrating I and Q channels the noise is reduced by the square of the number of samples integrated.    It really works.

The Peaberry is of a design that uses a mixer to convert the RF down to audio signals (I and Q), then processes it in the computer, and outputs it to the sound card.

The A to D conversion is done in the sound card chip, and I think its fixed in how it works.

In the setups where they do the A to D conversion at RF, I think they are limited in what chips are avalable (from the cell phone world), plus if you oversample, the cpu load goes way up in whatever is doing the processing.

I wonder where the tradeoff is between a 12 bit and 16 bit samples and oversample speeds...
If you want to listen to 30 MHz and have 4 samples, it has to run at 120 MHz sample rate...or is it 240 MHz?

To add to the situation, the pipe to move data in and out of the computer is a limitation with USB, less with firewire, and likely best with gig Ethernet.

There is lots of room for confusion regarding the required minimum sampling rate. ["For every complex question... there is a simple answer... that is wrong"]

Generally, the minimum sampling rate needed to sample an analog signal whose frequency content is limited (by analog filters) to 0-30MHz would be 2x the highest frequency... i.e. 60M samples per second.

Therefore to "over sample" this entire 30MHz wide signal, at 4x the minimum sampling rate, you would need a 240M sample per second sampler. However, the number of bits of accuracy per sample that can be obtained, in a sampler that can operate at 240M samples per second, is less than the number of bits of accuracy per sample that can be obtained in a sampler in the same general price range, that employs the same technology (e.g. CMOS silicon integrated circuits made from transistors having a specific gate length), and which can operate at 60M samples per second.  

Alternatively, if you use digital processing techniques to extract (filter out) a slice of a 30MHz wide analog signal that is being sampled at 60M samples per second... and (for example) the slice is only 10 kHz wide... then you are (in effect) oversampling the 10 kHz slice of the total 30MHz wide signal by the ratio of 3000:1.

Some types of A/D converters (depending upon the design details) can sample at very high sampling rates, but only with a modest number of bits of accuracy per sample. Some types of A/D converters (depending upon the design details) can sample at only modest sampling rates (e.g. 48k samples per second), but with a high number of bits per sample (e.g. 24 bits per sample). Some types of A/D converter designs allow you to trade off higher sampling rates against a lower number of bits of accuracy per sample.

When using a very high sampling rate... you typically incorporate some type of high speed digital processor (e.g. a specialized digital signal processor) on the same board as the A/D converter... in order to reduce the data rate required between the board and the general purpose computer it connects to.


Note that SDRs that use analog techniques to produce a pair of bandlimited I and Q outputs... that feed a stereo sound card... are using 2 A/D converters (one for the I channel and one for the Q channel). With certain designs (but not all designs), each of the two A/D converters can operate at a sampling rate of only 1x (not 2x) the highest frequency contained in the analog I and Q signals.

Stu

Looking at the chip, it looks like there is a lot of flexability in it, and it seems like a good one...

http://www.ti.com/lit/ds/slas533b/slas533b.pdf

The Afedri's max A/D bandwidth is 2 MHz; the RFSpacde SDR-IP is 1.333333MHz.  Believe the clock is 80 MSPS so the dynamic range is adequate.

Stu,

Your post is very interesting. Thought provoking. I freely admit not enough mathematical foundation to fully work with these issues. I see the idea clearly. Makes perfect sense. Excellent explanation. One you've used before, perhaps in a class?

The idea of using a higher sample rate to create an effectively higher bit depth is interesting.
Not sure it quite works in a practical application, although I suspect that the so-called "delta-sigma" converters might be employing something along these lines.

Seems like the idea is akin to Pulse Density Modulation (PDM)?

However at least in the audio world higher sample rates and oversampling were usually employed to move the HF limit up in frequency, not to increase bit depth.

I'm wondering what the reason(s) might be that this seemingly useful and simple technique seems not to have been employed, at least in audio applications (maybe it is used in industrial applications)? And, do you have any links to papers where this technique is used, or to actual applications?

(I understand the need for the DAC to be able to function at a higher sample rate, and that this might be the sole limitation)

              _-_-

http://www.linear.com/product/LTC2208 (http://www.linear.com/product/LTC2208)

The ADC in my DDC  sdR followed by an Altera Cyclone III Ep3C25 FPGA.

http://www.altera.com/devices/fpga/cyclone3/cy3-index.jsp (http://www.altera.com/devices/fpga/cyclone3/cy3-index.jsp)

With a 125 MHZ clock and suitable band pass filter, This combination enables sampling of any entire  60Mhz range up through  300MHz.  

Reading the literature of these devices and similar material may help clarify some stuff.

Over sampling has been done in CD players since they came out I think.

Not since they came out, but regardless the purpose is to extend the effective bandwidth so that artifacts are out of the bandpass... not to increase the number of bits.

                   _-_-

I found some spec's on the Peaberry:

From the ARRL report:

Noise floor (MDS), 400 Hz DSP filter BW:
10.1 and 14.2 MHz, -115 dBm
18.1 and 21.2 MHz, -114 dBm

Gain compression, 400 Hz DSP filter BW:
14 MHz, 20/5/2 kHz offset:  105/105/105 dB
Reciprocal mixing: -105/-93/-87 dBc
Two tone, 3rd-order DR: 99/99/99

Note that this latter reading would put the Peaberry on the Sherwood list at #5, tied with the Perseus, and just below the Flex 5000A.

For $149.