Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
- Softrock40
- Messages
Search
|
Re: New file uploaded to softrock40
Art,
toggle quoted message
Show quoted text
That is a good question. The linearity of any preamplifier is going to depend on biasing, how it is configured, the type of device used, etc. That is why you won't see any device specific spec on it. Obviously the rule of thumb for a preamplifier's gain degrading the intermod performance will not be a problem here, because you want a unity gain device. One big advantage of feedback in an amplifier is that it does a great job of linearizing the amplifier. I suggest using a device with enough feedback to turn it into a unity gain device. One common way of doing this is to use a bipolar emitter follower. You would get a high input impedance and a very low output impedance, depending on the gain of the device by itself. Although there is a resistor in the emitter, the output RF impedance will be much lower than the resistor, so it shouldn't degrade anything in terms of noise figure due to the resistor's RMS noise power, KTB, Boltzman's Constant * Temperature in degrees Kelvin * the bandwidth in Hertz. (This is the same for any value of resistance, so this can be a problem in OP AMP circuits, for example, when high values of resistance are used. Then there will be a higher noise voltage across the resistor for the same noise power. If the OP AMP is a voltage gain circuit with a high resistance at the input as part of the feedback circuit, then this can be a problem in low noise amplifiers.) Bipolars are lower noise at low impedances, and also have less 1/f noise, than do FETs. So I would suggest using a bipolar in this application. Since you are working at such a low frequency and I assume a very short length of coax run in terms of a wavelength, to the antenna, I wouldn't worry about impedance matching the output. Bill, WB5TCO KY1K wrote: Hi Bill and the group, |
|
Re: Software Defined Superhet
--- In softrock40@..., "Bob Hillard" <rhillard@a...>
wrote: ............. That is very good news. A while back I posted a message (No. 484) suggesting the same but did not get any feedback. I suppose 6m is not a high priority to the readers of this group. It would be very interesting if you could post some results regarding your experiments. Why didn't you use 9MHz as IF frequency? You could also have the 30m band? Jean-Claude Abauzit, PJ2BVU |
|
Re: Rocky 1.1 released - new features
Hi Alex,
Great! The support for multi-band will be a huge hit. Also, the wav playback-record is a very welcome addition since (I think) you will be able to play wave files recorded with PowerSDR and vise verse. 73 de Phil N8VB --- In softrock40@..., "Alex, VE3NEA" <alshovk@d...> wrote: frequency display allows you to select one of the pre-defined bands. For eachband, the LO frequency and I/Q balance info are stored. Run the programonce, then open Rocky.ini in a text editor and add/delete/edit the banddefinitions if necessary.signals or the output audio signal. The I/Q recording can be played back inRocky. The recordings of the output signal can be played in MS MultimediaPlayer. Please enter your name, callsign and QTH in the Settings dialog.Rocky will put these data in the comments section of your wav recordings sothat other users will see at a glance when and where the recording was made,who made it, and what the LO frequency was. When playing an I/Q waverecording, you can view this info by moving your mouse cursor over the playbackprogress bar. |
|
Rocky 1.1 released - new features
I have just released Rocky 1.1. Two new functions hav been added.
1. Multi-band support. The drop-down button to the right of the frequency display allows you to select one of the pre-defined bands. For each band, the LO frequency and I/Q balance info are stored. Run the program once, then open Rocky.ini in a text editor and add/delete/edit the band definitions if necessary. 2. WAV file recording/playback. You can record either the input I/Q signals or the output audio signal. The I/Q recording can be played back in Rocky. The recordings of the output signal can be played in MS Multimedia Player. Please enter your name, callsign and QTH in the Settings dialog. Rocky will put these data in the comments section of your wav recordings so that other users will see at a glance when and where the recording was made, who made it, and what the LO frequency was. When playing an I/Q wave recording, you can view this info by moving your mouse cursor over the playback progress bar. 73 Alex VE3NEA |
|
Re: QSD Models
Hi Bill,
Not, not at all. It is not a Tayloe detector anymore. The active opamp LPF provides the filtering that the capacitor/antenna impedance of the Tayloe detector did before. 73 de Phil N8VB --- In softrock40@..., Bill Dumke <billd@n...> wrote: opamp LPF source.following it. It is a single switch being driven by a 50 ohm inputThe output of the switch feeds directly into the non-inverting then theof the op amp. If the op amp's feedback resistor is 200 ohms whatconversion loss of the single switch circuit is exactly equal to capacitor".you get with the single switch circuit using the "sampling cycle toThis is for a 25% duty cycle clock. If I increase the duty clock is50% the loss now decreases. Obviously, the duty cycle of the I havechanging the output impedance of the switch as seen by the opamp. I use thea feelingthat we may be able to eliminate the C after the switch and switches?amplifieras a LPF to remove the sum component. and off another turnsinstantaneously, if you have switches turn on while switches tofrom on to off, the timing difference will cause the if bothhavethea short circuit for a brief period. By having less than 25% onwaveform you are insuring that the switches are all off beforeturning one on, and thereby avoid that brief short. capacitor.and deal ratio on aseems to of the equation ------------------------------------------------------------------------Service. <mailto:softrock40-unsubscribe@...?subject=Unsubscribe>YAHOO! GROUPS LINKS ------------------------------------------------------------------------ |
|
Re: Software Defined Superhet
Hi,
toggle quoted message
Show quoted text
This is the path that I started down over a year ago. I was going to basically duplicate the front end and 1st mixer of the K2 along with most of the K2's PLL. The K2 uses a DAC tuned PLL reference to fill in the 5 Khz steps of the PLL. Using the software NCO I could "fill in" the PLL steps in software instead of using the DAC tuned reference osc. The K2 uses an approx. 5 MHz IF so I was going to have a 20 Mhz xtal osc into a divide by four counter to drive the QSD. This arrangement would give coverage of 160 - 10m. Unfortuanetly, the more you get into a design like this you begin to see that a lot of the performance advantages of the QSD are now compromised. 73 de Phil N8VB --- In softrock40@..., "Bob Hillard" <rhillard@a...> wrote:
|
|
Re: Is the UBS-External Sound Card Works?
Bruce Beford
Hello, Hide.
I use a Soundblaster Audigy 2 NX USB external sound card on one of my laptops and it works well. I have no direct expereince with the exact card you ordered. I hope it will work ok for you. By the way, I shipped your 30M crystal yesterday. 73, Bruce N1RX --- In softrock40@..., "qrper723" <qrper72@y...> wrote: ordered the Ext.USB Sound Box as "Sound Blaster Digital Music LX" from |
|
Is the UBS-External Sound Card Works?
qrper723
Hi,
I made a mistake that my PC has no "LINE-IN" input terminal,only MIC (Analog) input exists. So I can not "Null the Images" from adjusting SDR-Software.I've ordered the Ext.USB Sound Box as "Sound Blaster Digital Music LX" from Amazon.com. Is there anyone to try the similar situation with me? It will work I think but I have no confidence now...hi Any suggestions? ja9mat hidehiko |
|
Re: New file uploaded to softrock40
Bill,
The preamp is needed to overcome the 6 dB loss of the double balanced mixer. Bob WA6UFQ --- In softrock40@..., Bill Dumke <billd@n...> wrote: SoftRock 30 Tayloe IF receiver. Even if the preamp was perfectly linear, which ------------------------------------------------------------------------ <mailto:softrock40-unsubscribe@...?subject=Unsubscribe>YAHOO! GROUPS LINKS ------------------------------------------------------------------------ |
|
Re: New file uploaded to softrock40
KY1K
Hi Bill and the group,
toggle quoted message
Show quoted text
I'm trying to sort out a similar preamp issue. I'd like to put my softrock on 185 KHz and 137 KHz. On those frequencies, loops rule as far as receive antennas go. I personally think there will be enough atmospheric noise on those frequencies so that a preamp isn't necessary. Although, loops tend to be a nice quiet receive antenna. But, my loop has around 1000 ohms impedance, so I need a toroid transformer to match the radio front end to the antenna. But, even if I get the impedance matched perfectly though, my nice high Q loop antenna loses 50 percent of it's Q, a problem I'd like to avoid if possible. Maintaining high antenna Q on those frequencies is very desirable! The obvious solution is a buffer amp with low/no gain. The buffer amp should have a high input impedance input so as not to load the loop and a 50 ohm output impedance.so it matches the input impedance of the softrock. And it needs to be very linear with a goof noise figure. But, all of the examples of preamps I find on the web do not have linearity ratings and they all have pretty high gain. I need the linearity but I do not need the gain. Can you suggest a buffer amp that might be appropriate or should I go ahead and take the 50 percent hit on the loaded Q by matching for optimum power transfer? I think I'm beginning to realize why flex-radio didn't include the LF and VLF bands in their SDR-1000. Any ideas? Thanks, Art At 11:38 PM 10/28/2005, you wrote:
It looks like a good idea. It would solve a lot of direct conversion |
|
Re: New file uploaded to softrock40
It looks like a good idea. It would solve a lot of direct conversion and DDS problems we have been talking about.
toggle quoted message
Show quoted text
But I would recommend avoiding putting a preamp in ahead of the SoftRock 30 Tayloe IF receiver. Even if the preamp was perfectly linear, which it won't be, it will still degrade the 3rd order intermodulation performance of the Tayloe detector by 2/3 of the preamp's gain in dB. For most HF operation, at least on the lower bands, I doubt you will need it anyway. Bill, WB5TCO softrock40@... wrote:
|
|
New file uploaded to softrock40
Hello,
This email message is a notification to let you know that a file has been uploaded to the Files area of the softrock40 group. File : /SDSTransceiver Block Diagram.bmp Uploaded by : cat100bob <rhillard@...> Description : Software Defined Superhet Block Diagram You can access this file at the URL: To learn more about file sharing for your group, please visit: Regards, cat100bob <rhillard@...> |
|
Software Defined Superhet
Consider using the SR40 as an 'IF strip', preceeding it with a MMIC
preamp, some bandpass filtering, and a Diode Balanced Mixer using an AMQRP DDS module as the LO source. I have done that very thing with excellent results. I've converted my SR40 to 30 meters, and use 10.138 mHz as the IF center frequency. I've located the LO above the incoming signal in order to reduce image frequencies. The output of the DDS module ranges from 11.9 mHz on 160 meters to 31.6 mHz on 15 meters. By replacing the DDS module with the newer 60 mHz model when it becomes available, coverage will be increased to include the 12, 10 and 6 meter bands. With the LO above the incoming signal, the radio operates on the lower sideband. However this can be corrected by software for bands that require USB operation. Band switching is accomplished by programming my DDS VFO controller () so that each of the eight programmable configurations represent a band. This gives me a bandswitching VFO with wrap-around, up/down frequency control or direct entry, variable VFO steps, sweep and scan functions, and 20 memory cells per band. Presently I am switching bandpass filters manually, but that will change with the addition of a band pass filter selection circuit similar to that found in the 'Software Defined Radio for the Masses' article. The bandpass filters are also similar to those found in the QEX article. One of the filters, the one I use most oftenly, covers the 15, 17, and 20 meter bands. By using this scheme, the SR40 circuitry is not taxed by trying to stretch it's operation past the limits of its design since it is always operating on 30 meters. However the radio itself can operate from 160M to 6M, with the exception of 30 meters (the IF frequency). Likewise, since the DDS frequency does not have to be divided by four, the useable range of the DDS module is extended to include all bands through 15 meters. What's next? Add a QSE module, a xmit DBM, some driver amps and WALLA, an all band transceiver. Then add an HC908 controller to take care of housekeeping, and a DSP module for operation independent of a computer; a nice winter project. I'll upload a block diagram of the radio. Bob WA6UFQ |
|
Re: Divide by 4 or 2
Milt,
toggle quoted message
Show quoted text
If it needs a symmetrical clock, I would recommend the sine wave to square wave converter used in the QRP2001 RF section. It just takes one gate, in this case it is an exclusive OR gate, but it can be two NAND gates, or anything similar as well. Just feed back some of the output to the input through a resistive T network with a shunt capacitor to ground in the middle to filter the RF. This bias feedback arrangement will square up any sine wave. It is commonly used by high speed logic designers to square up their clocks to obtain maximum frequency capability of their logic circuits. As an RF engineer I used one of these circuits as a high frequency phase stable frequency multiplier about 26 years ago. It used, I think, two ECL 1663 NOR gates which at that time were pretty fast. The sine wave input was at 50 MHz. All I had to do to "tune" it was to look at the output on a spectrum analuyzer and set one of the resistor values for minimum output on the even harmonics. I could get about 30 dB even harmonic rejection over quite a wide frequency range compared to the odd harmonics which are what make up a 50% duty cycle square wave. (Then it was also easier to filter the output signal I wanted at 350 MHz., which is an odd harmonic, since the even harmonics on either side were greatly attenuated. And it had spectacular phase stability, because all devices were differential inputs on the same die, so they were temperature matched as well.) Anyhow, it was an easy way to get a symmetric clock. I think it would work for the JK Flip Flop method as well. Bill Milt Cram wrote: KD5NWA wrote:That is the problem with /2 quadrature generation, you need awhich |
|
Re: QSD Models
Wouldn't that defeat the filter properties of the Tayloe detector?
toggle quoted message
Show quoted text
Bill WB5TCO Phil Covington wrote: P.S. I put a model of the capacitorless circuit to my blog at: |
|
Re: QSD Models
P.S. I put a model of the capacitorless circuit to my blog at:
73 de Phil N8VB --- In softrock40@..., "Phil Covington" <p.covington@g...> wrote: the and dealwaveform you are insuring that the switches are all off beforeturning one on, and thereby avoid that brief short. seems to Service. |
|
QSD Models
Hi All,
For those of you interested in the performance of the QSD circuit in Softrock40: A few hours of modelling of the SR40 circuit leads me to these conclusions. 1) The circuit behaves like a sample and hold detector since the time constant of the 0.1uF holding capacitor in parallel with the 10 ohm input resistor to the opamp is several times the period of the incoming RF signal. Thus, the circuit does not discharge much during each cycle. 2) Unlike many sample and hold circuits, which are driven from a very low impedance, this circuit has a relatively high impedance consisting of the transformed impedance of the source (50 ohms) plus 10 ohms plus the switch on-resistance. This prevents the holding capacitor from being charged to the peak value of the input signal. (I did not include the transformer in my model) 3) Where many sample and hold circuits use a very short pulse duration for the sampling, the SR40 uses a 25% duty cycle pulse. This tends to compensate for some of the lost charging due to the impedance of the circuit. Further, as the sampling duration is reduced below a certain point, the effective resistance of the source increases, decreasing the charging of the holding capacitor. 4) In my model, posted earlier, reducing the 10 ohm resistor on the transformer side of the switch, can significantly increase the output of the detector--as much as 10 dB, if the switch resistance is only 1 to 2 ohms. The effect will be reduced if the switch resistance is higher. 5) My comments yesterday suggesting that 50% sampling interval would be better, applies only to the case of an integrating detector. In such a case 50% sampling of a sine wave gives about 1.6dB more output than a 25% sampling interval. Reducing the sampling interval further reduces the output. 6) The "Ideal" sample and hold circuit should charge to the peak voltage of the input for very short sample intervals (i.e. much less than 25% of a cycle). Reducing the circuit resistance can increase the signal, but may introduce an unacceptable unbalance between the I and Q channels due to variations in the on-resistance of the switches. 7) The integrating detector should have some advantage over the sample and hold with regard to noise. However, if there is appreciable band limiting ahead of the detector, the advantage will likely be minimal. 73, Milt W8NUE -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.1.362 / Virus Database: 267.12.5/150 - Release Date: 10/27/2005 |
|
Re: QSD Models
Hi Phil,
I think that you are correct that we could eliminate the signal integrating capcitor. One of the circuits that I have been playing with does exacly as you have described - no C and an active opamp LPF following it. It is a single switch being driven by a 50 ohm source. The output of the switch feeds directly into the non-inverting input of the op amp. If the op amp's feedback resistor is 200 ohms then the conversion loss of the single switch circuit is exactly equal to what you get with the single switch circuit using the "sampling capacitor". This is for a 25% duty cycle clock. If I increase the duty cycle to 50% the loss now decreases. Obviously, the duty cycle of the clock is changing the output impedance of the switch as seen by the opamp. I am going to play with it more and report the results. Hope you had a nice trip. 73 de Phil N8VB --- In softrock40@..., pvharman@a... wrote: result. The conversion loss is lowest when the on time is 25%.determine the reason for this. Going to work on this on the plane home but I havea feeling that we may be able to eliminate the C after the switch and use theamplifier as a LPF to remove the sum component.have Service.a short circuit for a brief period. By having less than 25% on thewaveform you are insuring that the switches are all off beforeturning one on, and thereby avoid that brief short. |
|
Re: 50% QSD with four caps
KD5NWA
Interesting, if my capacitor is too big, I get a big hump in the center of the band-pass, where have we seen that before? When I say lower it, I mean really lower it, in this test to .001uF, signals went up, hump in the middle went down,
Interesting thing to try out, lower the integrating cap, increase the filtering in the op-amp to keep trash past 24KHz out of the sound card. In a quickie emulation with software you are not familiar with you might get results that don't pan out in real life, but it's a strange coincidence, might be worth checking out.. At 02:59 PM 10/28/2005, you wrote: I did a rough approximation, and using a SPDT switch into aCecil Bayona KD5NWA www.qrpradio.com I fail to see why doing the same thing over and over and getting the same results every time is insanity: I've almost proved it isn't; only a few more tests now and I'm sure results will differ this time ... |
Alex, VE3NEA
to navigate to use esc to dismiss