Oscilloscope for Ham Radio Use

[Smoke And Flames]

Ok,

After much research and reading, I'm just not sure if I should buy a digital or analog scope.  I want to be able to use it for analyzing audio, as well as RF frequencies (say 25Mhz max).  I found a few opinions stating that digital will not display what you need to see when working with audio/RF circuits.  Specifically, 8 bits of ADC resolution will not be sufficient.  Thanks to Dave, I've blown off the idea of a PC based scope.  I'd love to buy a Rigol 1052E, but I want to be absolutely sure that it's the right choice.   So, is there anyone who has real world experience, that can guide me on what I should buy?   Any help here would be much appreciated.

John

[alm]

Not an RF guy, but my impression is that using a scope to look at modulated signals is fairly useless (except to see than something is happening), you need a spectrum analyzer for that. A 100MHz scope is barely fast enough for a 25MHz signal, a 40MHz scope is clearly too slow.

DSOs can have problems with complex modulated signals, and are not great for low level signals. This is not as true today if you spend thousands on a mid-end model, but the cheap stuff doesn't have all fancy features designed to work around the short comings of DSOs. How about starting with a cheaper used analog scope until you know what you want and need? Even if you get a DSO later, you can always use it for RF and audio if it turns out that it's better for these jobs .

[DaveW]

If you're just using it for RF and audio, then you should be fine with an analogue 'scope, as you've got nice repetitive signals. I've been doing similar stuff for a while now, and I've never had to measure the characteristics of a signal particularly accurately using a scope. If you're measuring amplitude accurately you should be using a DMM or for frequency, using a frequency meter. For most ham work you won't need to look at the output, generally looking at the input signals or the IF, most ham needs are covered by a 40MHz scope, but more is always better!
An analogue scope will normally have a better noise floor than a digital which is useful for RF and audio. As for looking at modulated signals, an analog scope will give you a good idea of what's happening with an AM (modulation index, etc.) or FM signal, you won't be able to make many measurements on an FM signal, but as alm was saying, you'll need a spectrum analyser to do that properly!

[Mechatrommer]

spectrum analyser is darn expensive!

[Lance]

spectrum analyser is darn expensive!

You need the right tool for the right job.

Oscilloscopes give you amplitude vs. time. A spectrum analyzer gives you amplitude vs wavelength or frequency. From what I know you could only really use an oscilloscope to see what was going on with one specific frequency at a time.

[Scrts]

Since RF stuff is almost everywhere sine stuff (no square signals), so 100MHz is sufficient for sure. I would consider a faster (maybe 200-300MHz) analog scope and definitely a spectrum analyzer. We had one at work, which was capable doing FM demodulation... So much fun to work!

[Mechatrommer]

You need the right tool for the right job.

heard that alot, the problem is not everybody knows whats right is right.

Oscilloscopes give you amplitude vs. time. A spectrum analyzer gives you amplitude vs wavelength or frequency.

how spectrum analyzer determine the amplitude of a particular freq? there are a lot of sine'ing oscillation at one particular freq, and different amplitude (modulated?)

From what I know you could only really use an oscilloscope to see what was going on with one specific frequency at a time.

From what I know you could only really use an oscilloscope to see mix of alot of frequency at a time. FFT is a mean to separate them into individual frequency and amplitude.

[tyblu]

With RF circuits you also want to be able to see how your filters are performing, which may operate on signal images higher than your carrier frequency. I would suggest a spectrum analyzer for analog RF work. A sampling oscilloscope with proper inputs (interchangeable bandpass filters) may be able to do double duty, but I'm not really sure, and I believe they are generally much more expensive due to their impressive sampler bridge. Either way, it's more than the DS1052. Used 500MHz analog spectrum analyzers start at over $600 over here, though you may be able to get lucky! I want one, too.

[tyblu]

Oscilloscopes give you amplitude vs. time. A spectrum analyzer gives you amplitude vs wavelength or frequency.

how spectrum analyzer determine the amplitude of a particular freq? there are a lot of sine'ing oscillation at one particular freq, and different amplitude (modulated?)

First they bandpass to your frequency of interest to stop aliasing in following steps, then mix down using a sweeping local oscillator frequency -- across the band of interest -- and then filter again to get rid of higher order image. At this point the digital one would sample it just like a DSO (sampler bridge) and perform DSP on the measurements (high FFT cost overcome using decimation & resampling/cardinal spans). Analog ones use fantastic bandpass filters and measure the power or RMS amplitude coming through a small span.

Analog spectrum analyzers are my heroes.

[Zad]

Wow, people do like to over-complicate things nowadays. To characterise an RF filter, you need a sweep RF generator (sometimes called a wobbulator) and a detector. An oscilloscope is often perfectly fine as a detector, you drive the X-input from the sweep generator VCO voltage. Alternatively, an envelope detector is easy to build.

You do NOT need a vector network analyser for the vast majority of stuff, nor do you need a spectrum analyser (passive filters are linear and don't generate overtones). I get the impression that people think radio amateurs all have shacks full of tens of thousands of pounds worth of HP/Agilent gear just to make sure their 200W amp design works.

[Mechatrommer]

maybe for serious amateurs, they build they own sa/vna? i saw some diy sites online.

[jahonen]

Oscilloscopes give you amplitude vs. time. A spectrum analyzer gives you amplitude vs wavelength or frequency.

how spectrum analyzer determine the amplitude of a particular freq? there are a lot of sine'ing oscillation at one particular freq, and different amplitude (modulated?)

The spectrum analyzer works quite similar way than any ubiquitous superheterodyne (FM/AM) radio. That means that frequency spectrum around frequency of interest is first upconverted to first IF (something like 8 GHz), and then with subsequent operations of filtering and downmixing to lower IF frequencies, we are left with relatively narrow band of signal around conveniently low center frequency (tens of MHz). Now when we put narrow filter at final IF frequency, and sweep the oscillator that feeds the first mixer, we get the spectrum when we measure the amplitude of filter output as a function of the sweep oscillator frequency. Modern spectrum analyzers use FFT in addition of traditional sweep technique. That improves the sweep speed considerably, but does not work well with burst signals.

For more details (and better explanation of what I tried to explain above), see Agilent Application note 150, Agilent Spectrum Analysis Basics.

Spectrum and network analyzer has lot of other uses also than with purely RF applications, although not very common. I have used VNA to characterize bypassing effectiveness on power supply rails (ok, this can be done with sweep generator). Likewise, spectrum analyzer is quite handy to have around when diagnosing spurious resonances of switch-mode power supplies, or hunting emi problems. In oscilloscope the low-level broadband RF noise is just incomprehensible mess.

Regards,
Janne

[Zad]

maybe for serious amateurs, they build they own sa/vna? i saw some diy sites online.

You certainly can, but they still aren't trivial and the oscillator alone can be in the region of $100. The N2PK Network Analyser only runs up to 60MHz or so. Elektor published a sweep generator design for 0-450MHz that is pretty simple.

[Mechatrommer]

You certainly can

i could, but i dont think i could. maybe i'm not serious enough, or maybe my spec is not so simple (based on replies from another thread). i'll find a way, the "non engineering way", but sure i'll keep and thanx to all the links the helpers provide, in case... i have to do the "proper engineering way"... i wish i could, and i wish i have plenty of time for it.

[Lance]

From what I know you could only really use an oscilloscope to see what was going on with one specific frequency at a time.

From what I know you could only really use an oscilloscope to see mix of alot of frequency at a time. FFT is a mean to separate them into individual frequency and amplitude.

I was assuming you had an external method of isolating one frequency.