Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: [1]   Go Down
  Print  
Author Topic: SDR Choice  (Read 3695 times)
VA2PBJ
Member

Posts: 201




Ignore
« on: June 27, 2013, 11:44:33 PM »

Why is there an abundance of SDR receivers and not matching transmitters / or transceivers? What is the point?

Do people just not want to transmit?

Why are we calling devices like the "SDR Cube" a SDR radio? It has firmware not software and has knobs where computer ports should be. The firmware makes it closer to the big 3's with DSPs.

I want the general SDR market to grow, supporting non propriety software. I would like to know in 10 years that I can "fix" my radio (more likely to get it to work with whatever current os). I like what flex has done but I would not want to be stuck if they go under. It is just not much choices out there, if you want to do more than listen.

I am looking at SunSDR2, and they are offering more of what I would think of the norm that should be out there. It's under powered to be a main brick but it's in the right direction. They open sourced the code so I wouldn't get stuck in the upcoming years (I am a programmer). For all the receivers out there, why are there not more of these?
Logged

7 3 Peter VA2PBJ
SWL2002
Member

Posts: 374




Ignore
« Reply #1 on: June 28, 2013, 03:23:45 AM »

Why is there an abundance of SDR receivers and not matching transmitters / or transceivers? What is the point?

Do people just not want to transmit?

Why are we calling devices like the "SDR Cube" a SDR radio? It has firmware not software and has knobs where computer ports should be. The firmware makes it closer to the big 3's with DSPs.

I want the general SDR market to grow, supporting non propriety software. I would like to know in 10 years that I can "fix" my radio (more likely to get it to work with whatever current os). I like what flex has done but I would not want to be stuck if they go under. It is just not much choices out there, if you want to do more than listen.

I am looking at SunSDR2, and they are offering more of what I would think of the norm that should be out there. It's under powered to be a main brick but it's in the right direction. They open sourced the code so I wouldn't get stuck in the upcoming years (I am a programmer). For all the receivers out there, why are there not more of these?

There are many other users of SDRs than do not have any interest in transmitting.  You don't understand because you see everything through Ham colored glasses.

SunSDR2's software is not open source.  They published the source to their old software because they were illegally using open source code in their software, but not publishing their derived source.  They were called on it and they reluctantly made the source available.  So please don't hold this russian company out as a good example of open source!

Haven't you heard of HPSDR or the ANAN10/100/D that has been discussed recently right here in the very forum you are posting your naive question to?  How about the ELAD or the ADAT or the HiQSDR?
« Last Edit: June 28, 2013, 03:45:48 AM by SWL2002 » Logged
ZENKI
Member

Posts: 997




Ignore
« Reply #2 on: August 02, 2013, 07:31:05 AM »

A lot of the SDR makers typically dont want to be involved with the ham radio market. A good example is Winradio.
They have a   excellent product  the G31DDC which I use  as TX monitoring  spectrum analyzer.

The software that comes with the radio is more intended for MW DX'ers  and casual SW listening.
There are very few ham friendly features in the software.

The fist clue that tells you that a SDR receiver maker has no interest in the ham radio market is when they dont include a mute for RX protection.
The next step after this would be to include a complete RX/TX switching system that enable one to integrate your product with a ham transceiver switching system.
Not one SDR receiver maker offers this feature which is what is required to use their products safely in a  ham station. Who wants to blow up a 1000 dollar DSP radio?

The next issue is that that they  dont  seem to understand the potential uses for their products besides tuning the ham bands.
For example a SDR receiver with its very linear S-meter could make a nice EMC receiver.
All this would  require is the manufacturers to  offer 9khz standard EMC bandwidth and a quasi peak  detector. It would also require the ability to
set an antenna factor correction in the receiver. The way they measure signal levels will also will have to change. Rather than sampling the signal strength  to the resolution bandwidth they need to measure the signal strength  density across the whole 9khz EMC bandwidth.

Another feature that would be nice to have is the ability to set a MASK for pass  or fail for whatever you want to measure. This would be a great way for station to measure their own TX quality and see when their splatter falls  outside of acceptable levels. It also could be used for things like PSK IMD testing or even amplifier/radio IMD testing.

The next feature would be the ability to user the receiver to log signal level over a long period of time. Another feature would be  the ability to do a polar plot using the SDR receiver with some PC software. If the SDR manufacturers dont want to speak to the ham community about how this is done or what software is available these  kind of features are never going to appear. What gets me is that the shortwave market is dying and  the ham market is still very strong. Besides these kind of features are not only used by hams.

Besides transmitting there are some nifty stuff that you can do  using SDR receivers like beam steering and  direction finding.  It would be easy  for most hams to set
up a 3 or 4 channel correlative interferometer or watson watt direction finding system. This kind of system could give hams even on small lots to have good low noise receiving ability. Thats before we start to talk about QRM elimination  which can be done faster and better than the simple phase controllers we now use.

As for the DUC transmitters its just a matter of time. DDC/DUC  will be the cheapest way of producing a  transmitter. The ham manufacturers are just using their existing design skill set while they learn all the design secrets of DDC/DUC radios. Icom will probably be the first company to produce a DDC/duc transceiver. A radio like the 7800 could be offered for  half the price if it was using DDC/DUC architecture with Icoms  huge market share. Its just a matter of time.

The ADAT is a very nice radio. Its just not widely available and the company seems to be resourced stretched or is not interested in going big globally.

The real problem in the SDR  market at the moment for the ham manufacturers is which platform to choose that will be a winner and find one that has wide market acceptance. While hams are anal about receiver performance at all other costs its going to be hard to tell hams that a  direct sampling transceiver has performance that is good enough for the 99% of hams out there. Those that require very high contest dynamic range numbers for CW contesting will be the minority that will cry  about how bad  the receiver performance of SDR  direct  sampling radios are, which is  mostly a urban myth propagated by people living in the past.

You cannot match the signal purity and low inband RX IMD of a direct sampling radio. They just sound so sooth and  low noise. You wonder why anyone would want to put up with a raspy sounding high performance number down conversion radio  which is very fatiguing?  Then  again those who live and die by the numbers are still the ones transmitting the most horrendous SSB IMD  and keyclicks that negates  the performance of their "high IMD number" radios. Most hams have not listened to a ultra linear receiver whose  receiver  has no RX IMD. When they do they will soon realize what they are searching for is number high dynamic range numbers but low distortion and less inband RX IMD. Direct sampling radios in this area is  the clear winner. The more I use SDR direct sampling receivers the less and less I want to use  my high performance SDR downconversion transceiver. Its very tiring listening to this  kind of technology.

SDR transmitters are even simpler to  design that a analog transmitter. It just needs a lot of attention payed to thinks like spurious products, IMD, ALC and eliminating other TX design faults. This is before we start talking about the phase noise performance of the both the TX and the RX. The big laugh today is watching everyone brag about their transceiver numbers of their favorite toys, yet when we study the phase noise of the RX and TX we soon realize that they dont really understand the receiver performance equation very well.  We have many radios that have inadequate RX phase noise performance to achieve  the spectacular dynamic range numbers that they so dream about. This unhealthy obsession with receiver performance needs to stop and emphasis needs to placed back on fixing poor transmitter design issues and other simple things like making sure the radios are a joy to use and listen too.

I can pull out a old R2 DC  receiver and every time I turn this receiver on I always ask myself how i spent so much money on so many crap sounding receivers full of IMD and distortion. Lately every time  i turn on a direct sampling receiver its like relaxing in front of the fire with a good  wine versus using equipment that sounds like I am standing at the back of a crowd in rodeo with all the noise and  distortion. Hams  are missing out on lot by chasing pointless receiver numbers in favor of technology that could make the enjoyment of their hobby  much better.







Why is there an abundance of SDR receivers and not matching transmitters / or transceivers? What is the point?

Do people just not want to transmit?

Why are we calling devices like the "SDR Cube" a SDR radio? It has firmware not software and has knobs where computer ports should be. The firmware makes it closer to the big 3's with DSPs.

I want the general SDR market to grow, supporting non propriety software. I would like to know in 10 years that I can "fix" my radio (more likely to get it to work with whatever current os). I like what flex has done but I would not want to be stuck if they go under. It is just not much choices out there, if you want to do more than listen.

I am looking at SunSDR2, and they are offering more of what I would think of the norm that should be out there. It's under powered to be a main brick but it's in the right direction. They open sourced the code so I wouldn't get stuck in the upcoming years (I am a programmer). For all the receivers out there, why are there not more of these?
Logged
VA2PBJ
Member

Posts: 201




Ignore
« Reply #3 on: August 02, 2013, 07:42:58 AM »

That was a useful answer. Thanks for the info.
Logged

7 3 Peter VA2PBJ
M0HCN
Member

Posts: 473




Ignore
« Reply #4 on: August 05, 2013, 12:22:30 PM »

A DUC transmitter is easy, easy, easy, once you have the FPGA with the cordic core and cic filters working for the receiver.

Take one TxDac part or its equivalent from other manufacturers, add a little filtering and a pre driver (Current feedback opamp or two is the easy way), a couple of relays to provide a switched attenuator for when you wish to go low power, a driver and a final, some more filtering and a mess of protection electronics, job done.

Do the modulation limiting in I/Q space on the control processor rather then with an ALC loop, that way you avoid all the overshoot related pain in the arse, (RF envelope is strictly limited by +-sqrt (I^2 + Q^2) where I and Q are the baseband IQ pair.

Mind you, transmitters always were easier then receivers.

Ref measuring receivers, as long as you can set a wide enough bandwidth for the output of the CIC chain the measurement function is merely a 'small matter of software'.

One other thing that can be done with a multi channel sample synchronous transceiver is a dandy network analyzer, including being able to measure s21 and 22 if 4 rx channels and two tx channels are available, you can even do load pull testing and the like. 

Incidentally most of the LT2208 based receivers can also do 2M & 70cm if you butcher the LPF in the receive path and add suitable bandpass filters (By subsampling, the S/H bandwidth is ~700Mhz).

I have a half designed 4 channel synchronous sampling transceiver using the LTC2208 * 4 and a rather large cyclone 5, I like the notion of a multi operator single radio station where each operator can have the lobe off the 4 square pointed a different way.

TX side will be a two output DAC so I can experiment with quadrature PAs and some simple beam forming on both TX and RX.

It is going to be 'fun' to homebrew (700 odd ball FBGA), but I will find a way.

I suspect that part of the reason we have not seen much action this way is that VHDL is not a common skill among radio types, and that the cost of the sand is still rather high (About £1,000 for my homebrew effort if you include the board as one off (4 * ADCs @~£100 each, FPGA @ £150, 8 layer pcb, low noise reference clock, lots of power conversion junk (1.2V for the core at lots of amps)), and thats assuming the layout works first time!
 
Regards, Dan.
Logged
N0YXB
Member

Posts: 331




Ignore
« Reply #5 on: August 06, 2013, 02:17:40 PM »

I have a half designed 4 channel synchronous sampling transceiver using the LTC2208 * 4 and a rather large cyclone 5, I like the notion of a multi operator single radio station where each operator can have the lobe off the 4 square pointed a different way.

TX side will be a two output DAC so I can experiment with quadrature PAs and some simple beam forming on both TX and RX.

It is going to be 'fun' to homebrew (700 odd ball FBGA), but I will find a way.

I suspect that part of the reason we have not seen much action this way is that VHDL is not a common skill among radio types, and that the cost of the sand is still rather high (About £1,000 for my homebrew effort if you include the board as one off (4 * ADCs @~£100 each, FPGA @ £150, 8 layer pcb, low noise reference clock, lots of power conversion junk (1.2V for the core at lots of amps)), and thats assuming the layout works first time!
 
Regards, Dan.

Sounds very interesting.  Good luck and I hope you will keep us posted.
Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!