SDR Offsets?

[Bob Egan]

Is there s way to set an offset for an SDR connected to a Rx IF l

I have and RTL-SDR v4 connected to the IF  of my r7000.....which is at 10.7MHZ making the waterfall kinda not cool :-)

[Marcus PY2PLL]

This will be useful to those (like me) that uses 144 and 432-435MHz transverters on a 28MHz IF.

[Alex VE3NEA]

I have received several requests for the transverter offset option, but no details were provided. Is this offset the same on all bands, or should it be defined per-band? How should it work on the non-ham frequencies, such as 400 MHz band where many satellites transmit? Should the offset be the same for RX and TX? Should it be set separately for SDR and CAT control? I do not use a transverter and thus cannot answer these questions myself. Please provide these details to me, and any other details that may be important.

The best way to reqest a feature like this is by creating a PRD. md document (Project Requirements Definition) that includes all details. There is an easy way to create such PRD: go to one of the free AI web sits, such as Claude, ChatGPT or Gemini, give it a brief description of the feature, and ask it to create the file for you. Review the file and tell it to make changes, until you are satisfied. If you post such file in this forum, there will be much better chances to have the feature implemented.

[Marcus PY2PLL]

I'll try to create an initial PRD. In this meantime, I'm answering the questions above:

" Is this offset the same on all bands, or should it be defined per-band? ": if we consider standards V and UHF transverters, they usually convert 28-30MHz to 144-146 or 435-437 MHz. So the local oscillators will be 116MHz when VHF and 407MHz when UHF. One extra offset on UHF when operating AO-7 mode U-V: On this mode the uplink is right above 432MHz (so the range is 432-434 for a 28-30MHz IF, 404MHz local oscillator). The request is: make the SDR tune from 28 to 30MHz to be connected to the transverter RX IF output, keeping the doppler correction as is (same rate), keeping the same displayed frequencies on SkyRoof VFO and Waterfall.

Note: Elecraft K3S have a transverter config that allows me to tell the radio: XVRT1: IF=28MHz, RF=144MHz ; XVRT2: IF=28MHz, RF=435MHz and so on. SkyRoof via Rigtld send the VHF or UHF frequencies to the radio (this radio specifically), not 28-30MHz ones. Because the radio looks like a regular VU transceiver. For other rigs that does not have this XVRT configuration, the use of the SDR at the transverter port will be the same BUT the frequency information sent to the radio must have the same offset used to tune the SDR. One example: a simple IC-735 via CI-V: it will not understand 145.950, need to be 29.950.

"How should it work on the non-ham frequencies, such as 400 MHz band where many satellites transmit?" : hmm, most sats in this range transmit basically telemetry, LoRa, may be some SSTV or other imagery? I'm not sure. In this range, keep the SDR attached to the antenna, no offset at all.

"Should the offset be the same for RX and TX?" : only when the sat is VV or UU (packet, APRS, simplex apps, store and forward?); otherwise one offset for V (at 116MHz range if the IF is 28MHz), another for U (404 or 407MHz) and even something on SHF range for 2.2-2.3GHz.

"Should it be set separately for SDR and CAT control?": well, when using rigs that does not have transverter configuration such as an IC-735 and many others older or newer, seems to me that the frequency information is slightly different in the HF rig, that will be the TX , and the SDR that will be the RX. I'm saying that to agreed with your proposal / concept that consider the SDR as the main receiver for half or full duplex operations. Let me use FO-29 as an example (I turned doppler off to make numbers round):

Downlink => 435.835 - 407.000 = 28.835MHz on SDR
Uplink => 145.965 - 116.000 = 29.965MHz on HF RIG

One can choose to receive at the HF rig as well. This type of offset exist on HDSR for example, may be in other apps such as SDR Console but I don't use it. But for sat work things became more tricky because there's uplink and downlink in different bands, sometimes with sub-bands as the common IF frequency available is not so wide (28 to 30MHz). And there's the unique of AO-7 uplink exception, that require a second local oscillator.

Sorry for the BW.

73

MR  

[Bob Egan]

I'll see what I can cook up - in my case my radio (the r7000 being one) is a 10.7mhz IF which is very common on Rxvrs. Many transverters use 28 mhz

[Alex VE3NEA]

Here is a draft PRD, please review and comment:

https://github.com/VE3NEA/SkyRoof/blob/master/design-docs/PRD-transverter-offset.md

[Marcus PY2PLL]

Hi Alex:

4. Allow separate LO offset values for the SDR path and the CAT path per band, since the SDR transverter and the TX transverter may use different local oscillators.

Comment: Is this a case where someone uses a "dumb" transceiver or receiver and the SDR is connected to an IF port?

Passband coverage: with the satellite frequency ranges used in the default band table, all IF spans fit within the SDR's 3 MHz maximum bandwidth. The only marginal case is UHF 70cm (3 MHz IF span = 3 MHz max SDR bandwidth); if the SDR is configured at less than its maximum bandwidth, signals at the 435 or 438 MHz edges may go off-screen. All other bands are well within the passband at any realistic SDR bandwidth setting.

Comment: AirSpy Discovery have max BW ~ 725KHz when using 912ksps with the most recent firmwares. The older FW versions 610KHz BW with 768ksps. Why not re-center waterfall when selecting a new sat? Today I force this re-center keeping some zoom applied to waterfall while selecting some sat, and it works like a charm. Even if some signal at the edge goes out of screen zooming it (the screen) and re-selecting the sat forces it to the center of the screen anyway, out of this 3MHz. Example: some zoom applied, enter some terrestrial frequency. I know that it's not a satellite freq range but seems that this 3MHz "shifts" freely, which is ok.

O.T. How am I using waterfall: mostly the linear sats zoomed to screen extent, mouse drag and drop VFO to match any station frequency. Vy agile and smooth btw. I need to enable RIT often because users that are not using some doppler tracking scheme usually hit their TX frequency. I keep my TX + doppler correction "locked" and follow his drifting frequency using RIT + mouse wheel.

Default Band Table

Comment: Understood. Seems that there will be some freedom at the LO frequencies for eventual non standard ones. And even add extra bands. Nice to test external LOs OCXO/GPSDO driven on non standard frequencies (e.g 117MHz instead of 116MHz etc).

73
Marcus PY2PLL

[Alex VE3NEA]

I have updated the PRD based on the feedback and other considerations. Please review again.

marcus...@gmail.com  bobe...@gmail.com  ecoal...@gmail.com - I would like to hear your comments. If all is good, this PRD will be implemented.

[Marcus PY2PLL]

Alex et al;

As far as I understood, seems a good candidate for tests. I'll explore the LO setup table on the tests to see how I can handle some peculiarities of the transverters use: most of the IFs on 28-30 have, in fact, a flat pass band from 28 to 29.7 which is the usual 10m band limits. Seems that most rigs keep the RX band pass filters inline and this is ok, even if the xverter conversion puts the sat segment at 29.8-30MHz, a bit at the edge. Bottom line, the transverter use is an alternative for those that does not have a specific full duplex VHF/UHF rig.

I have a personal setup I'm preparing to test: will install those transverters at the back of the antennas, just 1 meter jumper cable there, sealed box. The 28-30 IF TX/RX (separated cables for TX and RX) will travel via a shielded CAT5 or 6 ethernet cable with baluns on both ends. PTTs and VU/UV commands using the remaining pairs.

73 Marcus PY2PLL