New User

Post your feedback here
dgmaley
Posts: 4
Joined: Sun Dec 31, 2017 8:21 pm

New User

Post by dgmaley » Sun Dec 31, 2017 9:04 pm

Afternoon from Iowa, USA. Got a RSP1A for Christmas and finally got to install SDRuno. Put software on two machines. One is an old HP laptop, running XP. It's used more as a toolbox to set up radios. The other is a big desktop with an I7 and lots of ram. The laptop starts 'chugging' audio between one and two megahertz. The desktop will sample at 10MHz but I have to run it wide open and keep an eye on the core temperature.
This little box reminds me of a IC-7300 front end with expanded coverage and needing a lot more CPU for what it can do. It needs a fast machine to do all the math.

Your software is made to give the look and feel of a radio receiver and does a good job at that. I was also wanting to use this also as a tool. Wondering if anyone has software that looks and feels more like a spectrum analyzer? There's no way to control input pads. Would have to do that by hand and be very careful, with a input max of 0dbm. The maximum span would still be 10MHz, but most spectrum analyzers don't do FFT. Accuracy would be OK, but it doesn't cost $10,000 either.

I've retired from communications work. Wish something like this was available ten years ago. Was lugging around 50 pounds of Tektronix spectrum analyzer then.

Dave...
Last edited by dgmaley on Thu Jan 01, 1970 12:00 am, edited 0 times in total.
Reason: No reason

Roger
Posts: 217
Joined: Fri Dec 29, 2017 7:59 pm

Re: New User

Post by Roger » Tue Jan 02, 2018 9:53 pm

There is nothing better than SDRUno for doing Spectrum Analyzer work with an RSP. That is because they have gone to considerable effort to make sure that the dBm levels are accurate as the IF and RF levels are changed by the user or AGC algorithm. All the other packages (SDR-Console, HDSDR, CubicSDR, SDR#) that work with RSPs are not calibrated so the dBm scales are only relative not absolute and if AGC is enabled mean next to nothing. HDSDR can be made to work as a spectrum analyzer if you turn off AGC and calibrate against a known signal level input.

There are SDR products from other manufacturers that work as low cost spectrum analyzers but they cost more money.

BTW have you seen the new low-cost spectrum analyzer box from Rigol - under $700 USD
https://www.rigolna.com/products/spectr ... 00/dsa705/

Reason: No reason

dgmaley
Posts: 4
Joined: Sun Dec 31, 2017 8:21 pm

Re: New User

Post by dgmaley » Wed Jan 03, 2018 4:45 am

Hi Roger.
I’m beginning to find how good their software is. Had an old version of HDSDR on my laptop drive so tried it. Won’t work. Discovered it requires a modified version to work with the Merics .sys driver. Able to calibrate at 10mhz to WWV and get less than 1hz error but only short term stability. Have an old Marconi counter that I can zero beat where it goes through a null about
once every five minutes.
Thanks for the heads up on Rigot. That size wouldn’t be hard to lug around,but was working ona 13ghz microwave system.
500mhz limit OK for what I’m doing now.
Thanks
Dave...
Last edited by dgmaley on Thu Jan 01, 1970 12:00 am, edited 0 times in total.
Reason: No reason

Roger
Posts: 217
Joined: Fri Dec 29, 2017 7:59 pm

Re: New User

Post by Roger » Thu Jan 04, 2018 1:30 am

dgmaley wrote:Hi Roger.
Had an old version of HDSDR on my laptop drive so tried it. Won’t work. Discovered it requires a modified version to work with the Merics .sys driver. .......OK for what I’m doing now.
Thanks
Dave...
Try downloading the latest version of HDSDR from the SDRPlay download site. Nice EXTIO driver that pops up to change settings. At first I didn't like it but I tried it again and find it works very well. The keyboard shortcuts are nice and the built-in Frequency Manager is great for switching bands and looking up Shortwave stations. It starts quickly and uses little CPU so I find that I use it 1/2 the time now. Calibration works well.

BTW - Speaking of calibration you need to give the RSP some time to temperature stabilize before you calibrate. I have mine calibrated to less than 1 Hz. now

Reason: No reason

dgmaley
Posts: 4
Joined: Sun Dec 31, 2017 8:21 pm

Re: New User

Post by dgmaley » Thu Jan 04, 2018 8:47 pm

Roger;
Downloaded the patched version of HDSDR that works with SDRPlay hardware. It's OK. Now looking like old technology. I do like the front end block diagram with variables.(EXIO). Used to seeing this on real world transmitters. Suspect this is part of the technology that will be going away.
Got the calibration worked out. Like to see calibrated error float above and below zero instead on one side. My last dongle, I had 10ppm and was stuck. Nice to see something under 0.1ppm and a way to maintain it.
The additional LNA is wonderful at 450MHz.
Found the schematic for RSP1 but not for 1A. Not sure it it's the same LNA.
The broadcast band filters are good for about 30db. This would be good for normal use. I live 1/4 mile from a 5KW AM tower on 600Khz and a 14.5KW FM station. Another 5KW at 1600KHZ within a mile. The two AM's measure 0dbm on my gutters and downspouts to a 50 ohm load. I use an old T style antenna tuner that cleans them out.
Last edited by dgmaley on Thu Jan 01, 1970 12:00 am, edited 0 times in total.
Reason: No reason

someYguy
Posts: 128
Joined: Sun Apr 10, 2016 7:12 pm

Re: New User

Post by someYguy » Tue Jan 09, 2018 4:13 am

For serious spectrum analyzer type measurements, far better to use SDRSharp (older version - v1.0.0.1361 works best) for this.

See my posts on this forum at the link below for some experiences regarding this:

viewtopic.php?t=1744

Controls in SDRSharp much more closely relate to and mimic those of a genuine spectrum analyzer, and the spectrum display grid is clearer than that of SDRUno and other SDR software.

Controls on SDRSharp which closely relate to spectrum analyzer controls are as follows:

"Zoom" slider = Frequency Span.
"Offset" slider = Reference Level
"Resolution" control(under the "FFT Display" tab) = Video Resolution.

There are others also.

As far as I know there are no dedicated spectrum analysis programs for the RSPx which have real spec analyzer features such as markers, dedicated frequency readouts, delta functions, etc. but SDRSharp comes closest.

Reason: No reason

someYguy
Posts: 128
Joined: Sun Apr 10, 2016 7:12 pm

Re: New User

Post by someYguy » Tue Jan 09, 2018 4:38 am

Roger wrote:There is nothing better than SDRUno for doing Spectrum Analyzer work with an RSP. That is because they have gone to considerable effort to make sure that the dBm levels are accurate as the IF and RF levels are changed by the user or AGC algorithm. All the other packages (SDR-Console, HDSDR, CubicSDR, SDR#) that work with RSPs are not calibrated so the dBm scales are only relative not absolute and if AGC is enabled mean next to nothing.
Not necessarily. SDR# can also be calibrated to a signal of known strength and from then on is very accurate within many mhz of the signal frequency. I made a calibration chart for just such use using calibration signals from a calibrated HP model 8656B signal generator.

Please don't see this as a bashing of SDRUno, nor a shout-out for SDR# but I've tried both and also HDSDR for SA uses and personally have found SDR# to be closer to a real spectrum analyzer than the others.

Reason: No reason

Roger
Posts: 217
Joined: Fri Dec 29, 2017 7:59 pm

Re: New User

Post by Roger » Wed Jan 10, 2018 12:03 am

someYguy wrote:
Roger wrote:There is nothing better than SDRUno for doing Spectrum Analyzer work with an RSP. That is because they have gone to considerable effort to make sure that the dBm levels are accurate as the IF and RF levels are changed by the user or AGC algorithm. All the other packages (SDR-Console, HDSDR, CubicSDR, SDR#) that work with RSPs are not calibrated so the dBm scales are only relative not absolute and if AGC is enabled mean next to nothing.
Not necessarily. SDR# can also be calibrated to a signal of known strength and from then on is very accurate within many mhz of the signal frequency. I made a calibration chart for just such use using calibration signals from a calibrated HP model 8656B signal generator.

Please don't see this as a bashing of SDRUno, nor a shout-out for SDR# but I've tried both and also HDSDR for SA uses and personally have found SDR# to be closer to a real spectrum analyzer than the others.
SDR# may look more like a Spectrum Analyzer but there problems when using it with an RSP. Only an old version of SDR# (1361 was last) will work with an RSP and that requires using EXTIO.dll. I won't go into the reasons why - it is easily found if you use google.

The problem is that the dBm values shown in HDSDR and SDR# are only accurate for the RF gain and LNA settings that were used during calibration. So if you go into the EXTIO and change the LNA or IF gain (which is the normally the case as you do different measurements) the dBm values shown are no longer valid. If you just change the IF Gain you can do an offset calculation without recalibration because they are in 1 dB steps. That is not the case with LNA gain which has does not have fixed dB changes between steps. SDRUno goes to great effort to use calibration tables that have adjustment values for all LNA and IF Gain settings. So it is the only program that gives an accurate dBm reading referred to the antenna input.

Reason: No reason

someYguy
Posts: 128
Joined: Sun Apr 10, 2016 7:12 pm

Re: New User

Post by someYguy » Thu Jan 11, 2018 1:51 am

Roger wrote:
someYguy wrote:
Roger wrote:There is nothing better than SDRUno for doing Spectrum Analyzer work with an RSP. That is because they have gone to considerable effort to make sure that the dBm levels are accurate as the IF and RF levels are changed by the user or AGC algorithm. All the other packages (SDR-Console, HDSDR, CubicSDR, SDR#) that work with RSPs are not calibrated so the dBm scales are only relative not absolute and if AGC is enabled mean next to nothing.
Not necessarily. SDR# can also be calibrated to a signal of known strength and from then on is very accurate within many mhz of the signal frequency. I made a calibration chart for just such use using calibration signals from a calibrated HP model 8656B signal generator.

Please don't see this as a bashing of SDRUno, nor a shout-out for SDR# but I've tried both and also HDSDR for SA uses and personally have found SDR# to be closer to a real spectrum analyzer than the others.
SDR# may look more like a Spectrum Analyzer but there problems when using it with an RSP. Only an old version of SDR# (1361 was last) will work with an RSP and that requires using EXTIO.dll. I won't go into the reasons why - it is easily found if you use google.

The problem is that the dBm values shown in HDSDR and SDR# are only accurate for the RF gain and LNA settings that were used during calibration. So if you go into the EXTIO and change the LNA or RF gain (which is the normally the case as you do different measurements) the dBm values shown are no longer valid. If you just change the LNA you can do an offset calculation without recalibration because they are in 1 dB steps. That is not the case with RF gain which has does not have fixed dB changes between steps. SDRUno goes to great effort to use calibration tables that have adjustment values for all LNA and Gain settings. So it is the only program that gives an accurate dBm reading referred to the antenna input.
Yup I've known for a long time about the SDR#/Airspy/SDRPlay feud but I use SDR# 1361 and it does indeed provide accurate amplitude measurements here in my system. For reference, here is how I do it:

1) Set my HP 8656B signal to the same frequency as the signal to be measured (or close, doesn't need to be exact), and set the output level to, say, -40dbm. Reference level (top line) in SDR# is at -30dbm
2) Connect the 8656B to the RSP1 with a very short - about 12" - 50-ohm cable.
3) Open the RSP1 Device controller and set the level to match that of the generator output. The 8656B is recently calibrated and so is accurate. The LNA of the RSP1 is always off. I never use it, ever.
4) Good to go within the frequency range that I need. I've also verified amplitude with a calibrated HP 400E RF AC voltmeter, and also with an Instek GSP-810 spectrum analyzer.

I can vary the amplitude from the 8565B from reference level to near noise level and the RSP1 tracks it nicely.

But you are correct in that the amplitude accuracy is only good within a limited range from the calibrated frequency, although I've found it to hold accuracy over several mhz so it hasn't been an issue for me.

SDRUno is quite accurate also, especially if you use the RSP Tuner AGC, which I do not. But the user interface is horrible for spectrum analyzer use, colors are not adjustable enough, and frequency span(zoom) and FFT resolution are not continuously adjustable as they are in SDR#.

Reason: No reason

Roger
Posts: 217
Joined: Fri Dec 29, 2017 7:59 pm

Re: New User

Post by Roger » Thu Jan 11, 2018 2:31 am

someYguy wrote:
SDRUno is quite accurate also, especially if you use the RSP Tuner AGC, which I do not. But the user interface is horrible for spectrum analyzer use, colors are not adjustable enough, and frequency span(zoom) and FFT resolution are not continuously adjustable as they are in SDR#.
Have you looked at the most recent version of SDRUno? There have been some improvements for frequency span, zoom and resolution bandwidth.

Note - I made some changes to the my previous post. I made a mistake with LNA and IF gain comments.

Reason: No reason

Post Reply