Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.
Both have 60 dB SNR at full bandwidth, but HP can improve sensitivity with Resolution bandwidth by adding points to go from 1MHz to 0.3Hz , whereas homebrew is fixed by SAW filter and has noise floor -75dB or -45dBm
Both have 60 dB SNR at full bandwidth, but HP can improve sensitivity with Resolution bandwidth by adding points to go from 1MHz to 0.3Hz , whereas homebrew is fixed by SAW filter and has noise floor -75dB or -45dBm
Thank you. I am curious how you came up with the 60db snr for the homebrew gear?
Also, you are talking about sensitivity, but how about dynamic range? I mean, I think that the FFT has limited dynamic range and that is it's main disadvantage compared to a real spectrum analyzer.
Will this homebrew gear have more dynamic range?
Noise floor to overload point can be an especially big problem, and its not just the amplifiers.
Your filters and mixers may also start to go non linear and create intermodulation products where very strong signals are involved.
While you can very easily compress the whole thing with a log amplifier to display 90 db on the screen, your analog front end may fall well short of that.
Modern state of the art HF communications receivers struggle to reach that kind of dynamic range, its not that easy to do.
Noise floor to overload point can be an especially big problem, and its not just the amplifiers.
Your filters and mixers may also start to go non linear and create intermodulation products where very strong signals are involved.
While you can very easily compress the whole thing with a log amplifier to display 90 db on the screen, your analog front end may fall well short of that.
Modern state of the art HF communications receivers struggle to reach that kind of dynamic range, its not that easy to do.
Even my Tektronix 491 has strict specs on the input signal power in order to be linear. The solution is to use input attenuators to keep the signal to be measured down to levels where overload does not occur. The log detector can help in displaying the low level signals and the high level ones, even if both are attenuated quite a lot.
The general rule is to keep the signals at the lowest level possible to avoid overload.
Don't you think the absence of an input amplifier helps a lot on this? Although it lowers the sensitivity.
Its not just the input amplifier, it may be the very last stage before the detector that is the very first thing that overloads.
If you are trying to measure something with respect to the level of a reference carrier, such as harmonics or phase noise, the carrier has to be able run full blast, while you measure whatever you are measuring below that.
If your noise floor is only 40db below the max carrier amplitude, that is as low as you can measure anything. So a high dynamic range is pretty important.
If a preamp lowers the relative noise floor, it may actually improve dynamic range.
The whole package needs to be optimised for highest overall dynamic range.
Gain distribution is critical, and a strategic attenuator placed somewhere may improve things, but only rigorous hands on testing will tell you.
I don't think you can just look at a schematic and make a value judgement.
It really needs to be fired up on the bench with high amplitude dual equal level signal sources, and a big switched attenuator, and watch for IMD products to appear that you know should not be there, and see where it falls on its face.
It really needs to be fired up on the bench with high amplitude dual equal level signal sources, and a big switched attenuator, and watch for IMD products to appear that you know should not be there, and see where it falls on its face.
You mean an oscillator such as this? https://www.ab4oj.com/test/imdtest/main.html
And then watch for mixer products that should not be there and set the levels of the oscillators so that the analyzer does not intermodulate. The you will know what is the maximum signal level to be handled, ad of course the minimum as well. Am I right?
If I am right, then how will I calculate the dynamic range based on this information?
Dynamic range is between the noise floor and the highest amplitude signal that creates no visible intermodulation products above the noise floor.
All you should see will be your two input carrier frequencies and nothing else.
Any distortion or non linearity anywhere in the system will generate new frequencies.
As long as these cannot poke their heads significantly above the noise floor your spectrum analyser will display only what it should display.
Dynamic range is then the difference between the noise floor and the maximum undistorted amplitude it can display.
On the graph shown at your link, the noise floor is given as -126dbm.
Third order products start to appear above the noise floor as the input signal rises above -39dbm.
So the dynamic range in that case would be:
126dbm - 39dbm or 89db.
Dynamic range is between the noise floor and the highest amplitude signal that creates no visible intermodulation products above the noise floor.
All you should see will be your two input carrier frequencies and nothing else.
Any distortion or non linearity anywhere in the system will generate new frequencies.
As long as these cannot poke their heads significantly above the noise floor your spectrum analyser will display only what it should display.
Dynamic range is then the difference between the noise floor and the maximum undistorted amplitude it can display.
On the graph shown at your link, the noise floor is given as -126dbm.
Third order products start to appear above the noise floor as the input signal rises above -39dbm.
So the dynamic range in that case would be:
126dbm - 39dbm or 89db.
This analyzer uses a trigger pulse to trigger the scope, and then it outputs the different amplitude signals on the scope channel. Obviously, the horizontal timing on the scope, must be set to match the scanning time of the analyzer, so that the signals are spanned in the whole area of the scope display.
I wonder how this analyzer (post#1) can be connected to a PC to be used as a display. A way is to use the LPT port with an A/D converter and an extra pin on the LPT to be used as an external trigger. Can a simple 20KHz speed A/D converter, connected to the LPT, be used? I believe yes, because the horizontal scan rate of the analyzer is not more than 50Hz or so, for a "flicker-free" signal on the display.
But since most modern laptops lack an LPT, I wonder if this can be done with the sound blaster. Wide range of sound blaster audio scopes software is out there and I have found one that supports triggering from one of the two channels of the stereo sound blaster, and the other channel to be used for the input signal.
This would work, but my problem is that the sound blaster does not accept DC input and since the output of the analyzer is a trigger pulse and a dc varying level, this cannot be read by the sound blaster. I was thinking of a circuit that can convert input DC levels accurately to variable amplitude audio tones, so that these tones can be read by the sound blaster. Then an audio scope software, would display this variable amplitude tone from the sound blaster input.
You could probably buy one of those USB virtual oscilloscopes.
The screen resolution might not be brilliant, often only 256 vertical 1024 horizontal. It would need to have external trigger capability.
Should have more than sufficient bandwidth.
That should just about do it all.
Use the spectrum analyser sweep to synchronise the oscilloscope horizontal sweep, and you are then in business with a flicker free display.
A two channel virtual oscilloscope may even have an XY mode so the spectrum analyser sweep can be used directly without the need to synchronise the oscilloscope.
It would also allow you to save screen captures to hard drive and print them out.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
