usrp-users@lists.ettus.com
Discussion and technical support related to USRP, UHD, RFNoC
View all threadsRF Switch Time on B200
Hi,
I am using a B200 in half-duplex mode and for my application I need to have fast switching times (couple of nanoseconds) and looking at the B200 it is using RF switch SKY13350-381LF which has delay time of 50nsec and rise/fall times of 50nsec. I'm looking for a way to get faster switching times and would appreciate feedback on following options:
-
Can I physically replace the SKY13350-381LF with a faster RF switch? -
I could add an external RF switch and try and use that. However would I need to modify the FPGA code in order to properly handle the timing? Also if I control from GPIO pin would the GPIO rise/fall times still be too slow? -
I looked at the X300 with different daughter cards (like UBX) but even that still has relatively slow RF switch. Are there any upgrade path options that would have a faster RF switch?
Thanks,
Tim
Hi Tim,
UHD does bring the ability to switch the GPIO the same moments it
switches the on-board antenna switches; [1] applies here.
However, 1/(50ns) = 20 MHz. With the maximum sampling rate of the B200
being 61.44 MHz, that would amount to at most a tiny bit more than three
samples "switching time".
I'd assume there'd be other technical hurdles to be solved here – could
you elaborate a bit on your use case? Maybe we can come up with
something elegant.
Best regards,
Marcus
[1] https://files.ettus.com/manual/page_gpio_api.html
On 21.12.2016 19:35, Timothy Coyle via USRP-users wrote:
Hi,
I am using a B200 in half-duplex mode and for my application I need to
have fast switching times (couple of nanoseconds) and looking at the
B200 it is using RF switch SKY13350-381LF which has delay time of
50nsec and rise/fall times of 50nsec. I’m looking for a way to get
faster switching times and would appreciate feedback on following options:
· Can I physically replace the SKY13350-381LF with a faster RF
switch?
· I could add an external RF switch and try and use that.
However would I need to modify the FPGA code in order to properly
handle the timing? Also if I control from GPIO pin would the GPIO
rise/fall times still be too slow?
· I looked at the X300 with different daughter cards (like
UBX) but even that still has relatively slow RF switch. Are there any
upgrade path options that would have a faster RF switch?
Thanks,
Tim
USRP-users mailing list
USRP-users@lists.ettus.com
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
Hi Marcus,
Thank you for the reference to the GPIO switching information.
I have a radar like application and send a TX signal and then need to turn around and listen to the echo signal coming back but if I have to wait 50ns for switch time then I miss a lot of the signal coming back. It would be great to get as many samples within the B200 bandwidth that I can but I might miss signal entirely with 50ns switch time.
Thanks,
Tim
From: Marcus Müller [mailto:marcus.mueller@ettus.com]
Sent: Wednesday, December 21, 2016 2:05 PM
To: Timothy Coyle; usrp-users
Subject: Re: [USRP-users] RF Switch Time on B200
Hi Tim,
UHD does bring the ability to switch the GPIO the same moments it switches the on-board antenna switches; [1] applies here.
However, 1/(50ns) = 20 MHz. With the maximum sampling rate of the B200 being 61.44 MHz, that would amount to at most a tiny bit more than three samples "switching time".
I'd assume there'd be other technical hurdles to be solved here - could you elaborate a bit on your use case? Maybe we can come up with something elegant.
Best regards,
Marcus
[1] https://files.ettus.com/manual/page_gpio_api.html
On 21.12.2016 19:35, Timothy Coyle via USRP-users wrote:
Hi,
I am using a B200 in half-duplex mode and for my application I need to have fast switching times (couple of nanoseconds) and looking at the B200 it is using RF switch SKY13350-381LF which has delay time of 50nsec and rise/fall times of 50nsec. I'm looking for a way to get faster switching times and would appreciate feedback on following options:
· Can I physically replace the SKY13350-381LF with a faster RF switch?
· I could add an external RF switch and try and use that. However would I need to modify the FPGA code in order to properly handle the timing? Also if I control from GPIO pin would the GPIO rise/fall times still be too slow?
· I looked at the X300 with different daughter cards (like UBX) but even that still has relatively slow RF switch. Are there any upgrade path options that would have a faster RF switch?
Thanks,
Tim
USRP-users mailing list
USRP-users@lists.ettus.commailto:USRP-users@lists.ettus.com
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
Tim,
I think you need to look to architectural solutions, not implementation.
Chasing a precise 2nS turnaround Tx->Rx is way beyond any of these
platforms. If I was you I would be looking at separate well isolated Rx and
Tx radio's running concurrently with engineering to minimize the direct
path.
-Ian
On Wed, Dec 21, 2016 at 11:31 AM, Timothy Coyle via USRP-users <
usrp-users@lists.ettus.com> wrote:
Hi Marcus,
Thank you for the reference to the GPIO switching information.
I have a radar like application and send a TX signal and then need to turn
around and listen to the echo signal coming back but if I have to wait 50ns
for switch time then I miss a lot of the signal coming back. It would be
great to get as many samples within the B200 bandwidth that I can but I
might miss signal entirely with 50ns switch time.
Thanks,
Tim
From: Marcus Müller [mailto:marcus.mueller@ettus.com]
Sent: Wednesday, December 21, 2016 2:05 PM
To: Timothy Coyle; usrp-users
Subject: Re: [USRP-users] RF Switch Time on B200
Hi Tim,
UHD does bring the ability to switch the GPIO the same moments it switches
the on-board antenna switches; [1] applies here.
However, 1/(50ns) = 20 MHz. With the maximum sampling rate of the B200
being 61.44 MHz, that would amount to at most a tiny bit more than three
samples "switching time".
I'd assume there'd be other technical hurdles to be solved here – could
you elaborate a bit on your use case? Maybe we can come up with something
elegant.
Best regards,
Marcus
[1] https://files.ettus.com/manual/page_gpio_api.html
On 21.12.2016 19:35, Timothy Coyle via USRP-users wrote:
Hi,
I am using a B200 in half-duplex mode and for my application I need to
have fast switching times (couple of nanoseconds) and looking at the B200
it is using RF switch SKY13350-381LF which has delay time of 50nsec and
rise/fall times of 50nsec. I’m looking for a way to get faster switching
times and would appreciate feedback on following options:
· Can I physically replace the SKY13350-381LF with a faster RF
switch?
· I could add an external RF switch and try and use that. However
would I need to modify the FPGA code in order to properly handle the
timing? Also if I control from GPIO pin would the GPIO rise/fall times
still be too slow?
· I looked at the X300 with different daughter cards (like UBX)
but even that still has relatively slow RF switch. Are there any upgrade
path options that would have a faster RF switch?
Thanks,
Tim
USRP-users mailing list
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
USRP-users mailing list
USRP-users@lists.ettus.com
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
Hi Tim,
I can't argue that with 50ns switch time, you'd miss 15m of close-range
reflections.
So this is basically a monostatic pulse radar?
The classical "radar" way of dealing with this is architecturally
different – if possible by means of isolation, you'd use circulators to
RX/TX on the same antenna. If that's impossible, chances are you run
into problems anyway – single-PCB transceivers simply can't have
hundreds of dBs of TX-RX crosstalk attenuation.
Assuming you don't have the dynamic range problem (because you say
"radar like application"), you'd still only lose 3 samples worth of info
– and in that situation, also only 3 samples worth of energy; that not
being a recoverable problem by the estimator that follows is not a good
sign with respect to that estimator's variance...
Could you hence explain a bit on your signal model?
Best regards,
Marcus
On 21.12.2016 20:31, Timothy Coyle wrote:
Hi Marcus,
Thank you for the reference to the GPIO switching information.
I have a radar like application and send a TX signal and then need to
turn around and listen to the echo signal coming back but if I have to
wait 50ns for switch time then I miss a lot of the signal coming back.
It would be great to get as many samples within the B200 bandwidth
that I can but I might miss signal entirely with 50ns switch time.
Thanks,
Tim
*From:*Marcus Müller [mailto:marcus.mueller@ettus.com]
Sent: Wednesday, December 21, 2016 2:05 PM
To: Timothy Coyle; usrp-users
Subject: Re: [USRP-users] RF Switch Time on B200
Hi Tim,
UHD does bring the ability to switch the GPIO the same moments it
switches the on-board antenna switches; [1] applies here.
However, 1/(50ns) = 20 MHz. With the maximum sampling rate of the B200
being 61.44 MHz, that would amount to at most a tiny bit more than
three samples "switching time".
I'd assume there'd be other technical hurdles to be solved here –
could you elaborate a bit on your use case? Maybe we can come up with
something elegant.
Best regards,
Marcus
[1] https://files.ettus.com/manual/page_gpio_api.html
On 21.12.2016 19:35, Timothy Coyle via USRP-users wrote:
Hi,
I am using a B200 in half-duplex mode and for my application I
need to have fast switching times (couple of nanoseconds) and
looking at the B200 it is using RF switch SKY13350-381LF which has
delay time of 50nsec and rise/fall times of 50nsec. I’m looking
for a way to get faster switching times and would appreciate
feedback on following options:
· Can I physically replace the SKY13350-381LF with a
faster RF switch?
· I could add an external RF switch and try and use that.
However would I need to modify the FPGA code in order to properly
handle the timing? Also if I control from GPIO pin would the GPIO
rise/fall times still be too slow?
· I looked at the X300 with different daughter cards (like
UBX) but even that still has relatively slow RF switch. Are there
any upgrade path options that would have a faster RF switch?
Thanks,
Tim
_______________________________________________
USRP-users mailing list
USRP-users@lists.ettus.com <mailto:USRP-users@lists.ettus.com>
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com