With USB or serial control I guess you will never reach the necessary accuracy in switching time. I see the only possibility in some hard wired connection to the USRP.
Ralph. From: USRP-users [mailto:usrp-users-boun...@lists.ettus.com] On Behalf Of Luke Hough Sent: Thursday, January 23, 2014 3:22 PM To: Adeel Anwar Cc: usrp-us...@lists.ettus.com; discuss-gnuradio@gnu.org Subject: Re: [USRP-users] [Discuss-gnuradio] B200 gain control and RF input power Adeel, Unfortunately, I planned to operate at 3.4GHz and designed my antenna to match that. A lot of the Minicircuits devices stop at around 3GHz. The limiter really doesn't achieve much relative to the required input power. Also, a circulator may be of little use after considering the reflected energy from the antenna. Even with a >90% efficiency, the reflected power will still be nearly 20dBm (if I understand the calculations right). I have material to construct a second antenna. My new plan is to build that antenna and measure the isolation provided by independent antennas. That probably won't be enough, so I'm thinking about an RF switch controlled via usb/serial and microcontroller (arduino maybe). -Luke On Wed, Jan 22, 2014 at 12:13 PM, Adeel Anwar <adeela...@gmail.com> wrote: Luke, I am also doing a similar project in which i plan to use TR switch instead of circulator because of high TX/RX isolation (60/70 dB) as compared to isolators (20/30 dB). Currently u are using power amplifier ZVE-8G which have output power ~=30 dBm so circulator will work as well but if u plan to go to higher power levels 60/70 dBm (kilo watts) range then T/R switch will be a better option. Otherwise if u just want a Radar test setup, then then simplest approach will be to use commonly available WiFi "Bi-Directional Amplifiers". Then have built-in limiters, power-amps, T/R-switches and LNA's etc. Limiter VLM-63-2W <http://www.minicircuits.com/pdfs/VLM-63-2W+.pdf> has Saturated output-power = +11.5 which is very w.r.t USRP max input power ( < -10 dBm ). Minicircuits ZFLM-252-1WL+ has Saturated-output power =0dBm, so in my opinion this will be a better option compared to VLM-63-2W (provided ur freq of operation is < 2.5G). Hope this helps. -Adeel On Sat, Jan 18, 2014 at 10:57 AM, Marcus D. Leech <mle...@ripnet.com> wrote: On 01/18/2014 12:31 AM, Luke Hough wrote: Get your tomatoes ready, I have attached a proposed block diagram and possible component specs. I have not actually purchased the limiter or the circulator, but I do have the power amp and antenna. The power amp is a ZVE-8G <http://www.minicircuits.com/pdfs/ZVE-8G+.pdf> . I was looking at the VLM-63-2W <http://www.minicircuits.com/pdfs/VLM-63-2W+.pdf> limiter and possibly a JCC3300T3800S10 circulator ( hoping for a sample ). The numbers on the block diagram don't exactly match the specs shown. The numbers are closer to the table values. I have also not taken insertion losses into account. Looking at the the B200 schematic <http://files.ettus.com/schematics/b200/b200.pdf> , I was wondering if during transmit I might set switch U807 to OUT2 while U805 is OUT1. Then on receive switch U807 back to OUT1. Basically, during transmit both RX1 and TX1 are set to use the TXRX1 antenna, but during receive, RX1 is switched back to antenna RX1. Can the switch be made in less than 1µs ? I don't think the switch can be made in under 1us from the host. With suitable mucking-about on the FPGA you might be able to come up with a suitable scheme that amounts to half-duplex switching. In the ordinary scheme of things the ATR state machine will switch the RX chain to the RX2 port during transmit. If this could be done fast enough, that would work fine, and you'd just put a terminator on the RX2 port in half-duplex mode. You could consider a scheme where some external machinery is helping with switching and "scheduling" things. Such machinery would perhaps arrange for a high-isolation path for RX during your TX cycle. This kind of problem is pretty standard in radar designs, so there are probably good solutions out there that could be hybridized to interface to an SDR approach. But radar isn't my particular expertise. On Fri, Jan 17, 2014 at 8:45 AM, Marcus D. Leech <mle...@ripnet.com> wrote: On 01/17/2014 09:37 AM, Luke Hough wrote: As a hobby project, I am developing an active radar. I am primarily familiar with simulation and signal processing, but not so much with RF hardware. So this is a learning opportunity. I do need to Tx/Rx on the same frequency either through a shared antenna or independent. I have constructed an antenna and measured the S11 parameter to be -11dB over a 300MHz band around the resonnant frequency. I was hoping to avoid a GPIO controlled switch. I don't think the B200 has any GPIO capability, so another controller device would be required. Would it be possible to control one of the skyworks switches on the frontend of the B200 in combination with a circulator and a limiter? Basically open the RX1 channel and keep the TXRX1 channel switched to the TX chain. -Luke Well, if this is a half-duplex application, the USRP already does switching. Whenever the unit is transmitting, the RX is connected to the the RX port on the box. Why don't you draw a diagram of what your intended setup is, and we can through metaphoric tomatoes at it, as it were. On Fri, Jan 17, 2014 at 1:34 AM, Ralph A. Schmid, dk5ras <ra...@schmid.xxx> wrote: Hi, > +7dBm is *very* risky. Hmmm...3µs are not very long...but it is a risk, agreed. > If you're feeding a common antenna, the usual approach is to use a > diplexer/duplexer arrangement to isolate the TX frequency from the RX > frequency (assuming different-frequency full-duplex). I guess he uses the same frequency for TX and RX - usage of an isolator/circulator makes me think so :) But this only works for a certain degree and requires no reflected power at all (that means, perfect impedance match) at the antenna port. Depending on the needed timing it may be an option constructing a PIN diode RX/TX switch, operated from some GPIO. > In fixed-purpose applications, like WiFi, where a common antenna is used, > there's a duplexor, usually implemented in some kind of ceramic > resonator technology that has bandpass and band-stop components to it, > to keep the RX isolated very deeply. This will not work for WiFi, as this transmits and receives on the same frequency, they usually apply the above mentioned diode method to rapidly switch between RX and TX path. Those ceramic diplexers are common for cellphones and some digital LMR systems, as they have the need for full duplex on different frequencies. > -- > Marcus Leech > Principal Investigator > Shirleys Bay Radio Astronomy Consortium > http://www.sbrac.org Ralph. -- Ralph A. Schmid Mondstr. 10 90762 Fürth +49-171-3631223 <tel:%2B49-171-3631223> ra...@schmid.xxx http://www.bclog.de/ _______________________________________________ USRP-users mailing list usrp-us...@lists.ettus.com http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com _______________________________________________ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org https://lists.gnu.org/mailman/listinfo/discuss-gnuradio _______________________________________________ USRP-users mailing list usrp-us...@lists.ettus.com http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
_______________________________________________ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
