Re: [Flexradio] Image Null Calibration.
What is the ultimate image null obtainable? I was experimenting with this recently and the best I could do with the image 24khz above the fundamental was 83 db down using the automatic cal mode. 73 W9OY __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com
Re: [Flexradio] Image Null Calibration.
At 06:37 AM 9/15/2006, Lee A Crocker wrote: What is the ultimate image null obtainable? I was experimenting with this recently and the best I could do with the image 24khz above the fundamental was 83 db down using the automatic cal mode. 83 dB is quite impressive. The challenge is not so much in nulling at one specific frequency configuration(one DDS frequency + one audio frequency), which is probably mostly determined by the sampling jitter and phase noise of the various oscillators in the process, including the transmitted source. That is, say you had captured a set of data, containing the desired signal and an image. You could find a mathematical sine wave that would match the image, and then subtract it. The problem is that the image signal itself isn't actually perfect sine, so there's some residual power that you just can't cancel. However, the real challenge is in rejecting the image over a wideband (either audio or RF). I think I've convinced myself that on the DDS LO side of the QSD, the mismatch between the two low pass filters is small enough that you probably only need one RF frequency cal point in each ham band (if not even fewer). http://home.earthlink.net/~w6rmk/sdr1000/ddsfilter.pdf has some analysis and thoughts on this. The bigger challenge is the match between I/Q channels in the audio chain. The bandwidth over which phase match must be consistent is much larger (many decades). The current PowerSDR software does a single frequency point cal in the audio band. The two instrumentation amps are probably fairly well matched, and the low pass filter that's part of the QSD is only a single pole, with cutoff well above the frequencies of interest (500kHz or so, in the original SDR1000), so whatever relative phase shift there is between the channels can be expected to be fairly consistent. It's the inside of the audio interface to the PC that I'd have questions about. 73 W9OY __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com James Lux, P.E. Spacecraft Radio Frequency Subsystems Group Flight Communications Systems Section Jet Propulsion Laboratory, Mail Stop 161-213 4800 Oak Grove Drive Pasadena CA 91109 tel: (818)354-2075 fax: (818)393-6875 ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com
Re: [Flexradio] Image Null Calibration.
Would the narrow range encompass a listening bandwidth of say 500 Hz on CW? -- Guy - Original Message - From: Eric Wachsmann [EMAIL PROTECTED] To: 'Lee A Crocker' [EMAIL PROTECTED]; 'Flexradio' FlexRadio@flex-radio.biz Sent: Friday, September 15, 2006 9:47 AM Subject: Re: [Flexradio] Image Null Calibration. 83dB is probably about as good as the automatic routine will get. You can do a bit better (~100dB) by adjusting the up/down controls manually with the display in AVG mode. Note that the 100dB of rejection is only for a very narrow range of audio frequencies though. So getting 100dB (vs 80dB) probably isn't going to help you much practically speaking. Eric Wachsmann FlexRadio Systems -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] radio.biz] On Behalf Of Lee A Crocker Sent: Friday, September 15, 2006 8:38 AM To: Flexradio Subject: Re: [Flexradio] Image Null Calibration. What is the ultimate image null obtainable? I was experimenting with this recently and the best I could do with the image 24khz above the fundamental was 83 db down using the automatic cal mode. 73 W9OY __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com Scanned for viruses by Blue Coat http://www.WinProxy.com/ ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com
Re: [Flexradio] Image Null Calibration.
At 05:36 PM 9/15/2006, Guy Olinger, K2AV wrote: Would the narrow range encompass a listening bandwidth of say 500 Hz on CW? The problem isn't that you can get a good null in 500 Hz for CW, it's getting a null everywhere else (to suppress the image, that is). And, don't forget that in spur reduction mode, the DDS goes in fairly large steps, with the fine tuning occuring in the software downconversion. So, the actual audio frequency out of the hardware varies. If you were interested in nulling a specific troublesome interferer on the image at a fixed frequency, then it would work fine. -- Guy - Original Message - From: Eric Wachsmann [EMAIL PROTECTED] To: 'Lee A Crocker' [EMAIL PROTECTED]; 'Flexradio' FlexRadio@flex-radio.biz Sent: Friday, September 15, 2006 9:47 AM Subject: Re: [Flexradio] Image Null Calibration. 83dB is probably about as good as the automatic routine will get. You can do a bit better (~100dB) by adjusting the up/down controls manually with the display in AVG mode. Note that the 100dB of rejection is only for a very narrow range of audio frequencies though. So getting 100dB (vs 80dB) probably isn't going to help you much practically speaking. Eric Wachsmann FlexRadio Systems Jim, W6RMK ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com
[Flexradio] Image Null Calibration.
The maximum output of the Elacraft XG-2 is -73 dBm. The manual of the SDR-1000 requests a signal of -50 dBm for the Automatic calibration of the Image Null. Is the -73 dBm of the XG-2 insufficient for a reliable I mage Null Calibration ? 73 Eddy ON5UQ. -- next part -- An HTML attachment was scrubbed... URL: /pipermail/flexradio_flex-radio.biz/attachments/20060914/eddcc9ff/attachment.html ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com
Re: [Flexradio] Image Null Calibration.
I'm guessing this is a typo and possibly was meant to say 50uV of which the XG-1/XG-2 is capable (i.e. -73dBm, as you state). I've used my XG-1 to do this calibration and had assumed the signal level (at S9) was adequate. What say, Eric? Pete, N3EVL -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Van de Velde Eddy Sent: Thursday, September 14, 2006 1:49 PM To: flexradio@flex-radio.biz Subject: [Flexradio] Image Null Calibration. The maximum output of the Elacraft XG-2 is -73 dBm. The manual of the SDR-1000 requests a signal of -50 dBm for the Automatic calibration of the Image Null. Is the -73 dBm of the XG-2 insufficient for a reliable I mage Null Calibration ? 73 Eddy ON5UQ. -- next part -- An HTML attachment was scrubbed... URL: /pipermail/flexradio_flex-radio.biz/attachments/20060914/eddcc 9ff/attachment.html ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com
Re: [Flexradio] Image Null Calibration.
No typo here. It is useful to have a stronger signal than -73dBm when doing the image calibration. Considering that the noise floor is going to be around -130dBm, this only gives you a range of about 60dB for a best case null (because it becomes difficult to null below the noise floor). Using a stronger input signal gives a larger image and thus the calibration can more effectively null the signal further (typically all the way into the noise floor). Eric Wachsmann FlexRadio Systems -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] radio.biz] On Behalf Of [EMAIL PROTECTED] Sent: Thursday, September 14, 2006 1:38 PM To: [EMAIL PROTECTED]; flexradio@flex-radio.biz Subject: Re: [Flexradio] Image Null Calibration. I'm guessing this is a typo and possibly was meant to say 50uV of which the XG-1/XG-2 is capable (i.e. -73dBm, as you state). I've used my XG-1 to do this calibration and had assumed the signal level (at S9) was adequate. What say, Eric? Pete, N3EVL -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Van de Velde Eddy Sent: Thursday, September 14, 2006 1:49 PM To: flexradio@flex-radio.biz Subject: [Flexradio] Image Null Calibration. The maximum output of the Elacraft XG-2 is -73 dBm. The manual of the SDR-1000 requests a signal of -50 dBm for the Automatic calibration of the Image Null. Is the -73 dBm of the XG-2 insufficient for a reliable I mage Null Calibration ? 73 Eddy ON5UQ. -- next part -- An HTML attachment was scrubbed... URL: /pipermail/flexradio_flex-radio.biz/attachments/20060914/eddcc 9ff/attachment.html ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com
Re: [Flexradio] Image Null Calibration.
Eric said: No typo here. It is useful to have a stronger signal than -73dBm when doing the image calibration. So anyone have a good recommendation for a stable signal generator that won't break the bank? -Tim --- Integrated Technical Services Too much of everything is just enough. -Rob Barlow -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Eric Wachsmann Sent: Thursday, September 14, 2006 2:44 PM To: [EMAIL PROTECTED]; [EMAIL PROTECTED]; flexradio@flex-radio.biz Subject: Re: [Flexradio] Image Null Calibration. No typo here. It is useful to have a stronger signal than -73dBm when doing the image calibration. Considering that the noise floor is going to be around -130dBm, this only gives you a range of about 60dB for a best case null (because it becomes difficult to null below the noise floor). Using a stronger input signal gives a larger image and thus the calibration can more effectively null the signal further (typically all the way into the noise floor). Eric Wachsmann FlexRadio Systems -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] radio.biz] On Behalf Of [EMAIL PROTECTED] Sent: Thursday, September 14, 2006 1:38 PM To: [EMAIL PROTECTED]; flexradio@flex-radio.biz Subject: Re: [Flexradio] Image Null Calibration. I'm guessing this is a typo and possibly was meant to say 50uV of which the XG-1/XG-2 is capable (i.e. -73dBm, as you state). I've used my XG-1 to do this calibration and had assumed the signal level (at S9) was adequate. What say, Eric? Pete, N3EVL -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Van de Velde Eddy Sent: Thursday, September 14, 2006 1:49 PM To: flexradio@flex-radio.biz Subject: [Flexradio] Image Null Calibration. The maximum output of the Elacraft XG-2 is -73 dBm. The manual of the SDR-1000 requests a signal of -50 dBm for the Automatic calibration of the Image Null. Is the -73 dBm of the XG-2 insufficient for a reliable I mage Null Calibration ? 73 Eddy ON5UQ. -- next part -- An HTML attachment was scrubbed... URL: /pipermail/flexradio_flex-radio.biz/attachments/20060914/eddcc 9ff/attachment.html ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com
Re: [Flexradio] Image Null Calibration.
On 14/09/06, Tim Ellison [EMAIL PROTECTED] wrote: So anyone have a good recommendation for a stable signal generator that won't break the bank? -Tim Tim, Why not use Analog Devices evaluation board for AD9952 http://www.analog.com/en/prod/0,,770_843_AD9952%2C00.html? Does USD250.00 break the bank? 73, Ahti OH2RZ ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com
Re: [Flexradio] Image Null Calibration.
At 12:02 PM 9/14/2006, Tim Ellison wrote: Eric said: No typo here. It is useful to have a stronger signal than -73dBm when doing the image calibration. So anyone have a good recommendation for a stable signal generator that won't break the bank? I assume the Elecraft unit has a pad to get the level down to sub nanowatt levels? Why not remove or change the pads. There are a fair number of inexpensive quiet oscillators out there. If you aren't concerned about frequency accuracy, then any of the inexpensive low jitter clock oscillator modules would work. You can follow it with a pad, a LC BPF, and a pad. Minicircuits sells inexpensive IF filters for, e.g. 10.7 MHz, and they do quite nicely for removing harmonics from a 10MHz source. If you want good frequency accuracy, low phase noise, and a bunch of power, what about a 10 MHz OCXO from someone like Wenzel or surplus HP/Agilent unit. ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com
Re: [Flexradio] Image Null Calibration.
Tim Ellison wrote: Eric said: No typo here. It is useful to have a stronger signal than -73dBm when doing the image calibration. So anyone have a good recommendation for a stable signal generator that won't break the bank? -Tim --- Integrated Technical Services Too much of everything is just enough. -Rob Barlow -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Eric Wachsmann Sent: Thursday, September 14, 2006 2:44 PM To: [EMAIL PROTECTED]; [EMAIL PROTECTED]; flexradio@flex-radio.biz Subject: Re: [Flexradio] Image Null Calibration. No typo here. It is useful to have a stronger signal than -73dBm when doing the image calibration. Considering that the noise floor is going to be around -130dBm, this only gives you a range of about 60dB for a best case null (because it becomes difficult to null below the noise floor). Using a stronger input signal gives a larger image and thus the calibration can more effectively null the signal further (typically all the way into the noise floor). Eric Wachsmann FlexRadio Systems -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] radio.biz] On Behalf Of [EMAIL PROTECTED] Sent: Thursday, September 14, 2006 1:38 PM To: [EMAIL PROTECTED]; flexradio@flex-radio.biz Subject: Re: [Flexradio] Image Null Calibration. I'm guessing this is a typo and possibly was meant to say 50uV of which the XG-1/XG-2 is capable (i.e. -73dBm, as you state). I've used my XG-1 to do this calibration and had assumed the signal level (at S9) was adequate. What say, Eric? Pete, N3EVL -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Van de Velde Eddy Sent: Thursday, September 14, 2006 1:49 PM To: flexradio@flex-radio.biz Subject: [Flexradio] Image Null Calibration. The maximum output of the Elacraft XG-2 is -73 dBm. The manual of the SDR-1000 requests a signal of -50 dBm for the Automatic calibration of the Image Null. Is the -73 dBm of the XG-2 insufficient for a reliable I mage Null Calibration ? 73 Eddy ON5UQ. -- next part -- An HTML attachment was Use your XG-1 for frequency calibration, but use a transmitter into a dummy load for image rejection calibration. -- Cecil KD5NWA www.qrpradio.com www.hpsdr.com Sacred Cows make the best Hamburger! Don Seglio Batuna ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com
Re: [Flexradio] Image Null Calibration.
Build a simple crystal or LC oscillator fom the ARRL handbook. Tom Thompson W0IVJ Jim Lux wrote: At 12:02 PM 9/14/2006, Tim Ellison wrote: Eric said: No typo here. It is useful to have a stronger signal than -73dBm when doing the image calibration. So anyone have a good recommendation for a stable signal generator that won't break the bank? I assume the Elecraft unit has a pad to get the level down to sub nanowatt levels? Why not remove or change the pads. There are a fair number of inexpensive quiet oscillators out there. If you aren't concerned about frequency accuracy, then any of the inexpensive low jitter clock oscillator modules would work. You can follow it with a pad, a LC BPF, and a pad. Minicircuits sells inexpensive IF filters for, e.g. 10.7 MHz, and they do quite nicely for removing harmonics from a 10MHz source. If you want good frequency accuracy, low phase noise, and a bunch of power, what about a 10 MHz OCXO from someone like Wenzel or surplus HP/Agilent unit. ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com -- next part -- An HTML attachment was scrubbed... URL: /pipermail/flexradio_flex-radio.biz/attachments/20060914/f41e184e/attachment.html ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com
Re: [Flexradio] Image Null Calibration.
I simply used my back-up station transceiver (FT-1000) into the dummy load with the drive cut all the way back. I also disconnected the antenna from the SDR 1000. It provided a nice strong stable signal for calibration. Bob K8MLM In a message dated 9/14/2006 6:00:50 P.M. Eastern Daylight Time, [EMAIL PROTECTED] writes: Build a simple crystal or LC oscillator fom the ARRL handbook. Tom Thompson W0IVJ Jim Lux wrote: At 12:02 PM 9/14/2006, Tim Ellison wrote: Eric said: No typo here. It is useful to have a stronger signal than -73dBm when doing the image calibration. So anyone have a good recommendation for a stable signal generator that won't break the bank? I assume the Elecraft unit has a pad to get the level down to sub nanowatt levels? Why not remove or change the pads. There are a fair number of inexpensive quiet oscillators out there. If you aren't concerned about frequency accuracy, then any of the inexpensive low jitter clock oscillator modules would work. You can follow it with a pad, a LC BPF, and a pad. Minicircuits sells inexpensive IF filters for, e.g. 10.7 MHz, and they do quite nicely for removing harmonics from a 10MHz source. If you want good frequency accuracy, low phase noise, and a bunch of power, what about a 10 MHz OCXO from someone like Wenzel or surplus HP/Agilent unit. ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com -- next part -- An HTML attachment was scrubbed... URL: /pipermail/flexradio_flex-radio.biz/attachments/20060914/f41e184e/attachment.html ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com -- next part -- An HTML attachment was scrubbed... URL: /pipermail/flexradio_flex-radio.biz/attachments/20060914/a001ed6b/attachment-0001.html ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com
