View Single Post
Old 21st Apr 2021, 2:58 am   #24
ortek_service's Avatar
Join Date: May 2018
Location: Northampton, Northamptonshire, UK.
Posts: 629
Default Re: Buffer Amplifier for Digital Frequency Counter

Yes, you'd hope they would have had a basic HF RF signal generator and good enough 'scope / level meter, to measure the bandwidth at least, before making any claims on this, even if they were unable to measuring the noise performance, or check it wasn't oscillating at some very high frequency - One of RF design's Murphy's laws, used to be that whenever you tried to make an amplifier, you often end up with an oscillator. And just simulating it the circuit alone, without the layout strays / coupling, won't necessarily tell you if it will always be stable. I've known even designs by professional Engineers to have oscillated at frequencies beyond the range of their spectrum analysers.

A lot may depend on the origins of the circuit, as it could have just been an example circuit from somewhere, that was never actually built (or it was designed to go into a high impedance rather than 50R etc. load - Although it does look like that simulation has also been done like that initially, so it may well be even worse into a Low impedance.)

Back in the 80's, it was also not uncommon for there to be many circuit and veroboard etc. layout errors in Electronics magazines published projects (even ones that didn't have corrections in following issues, so I once sent them a long list of ones I'd found!).
As I found out when building several, that never worked (although I did eventually learn quite a bit about fault-finding and bad designs from these!). Lack of access to much test equipment back then, also meant you didn't really know what was happening. So later-on getting given a 'scope, was a major step forward.

But there were some basic unforgivable errors, like the experienced Maplin designer of their 25W amplifier project that claimed the inputs were 100k impedance, as they used a 100k level preset. But they had followed it with a classic inverting OpAmp that only had 1k into its -ve input 'virtual-earth'. And I ended up building a proper non-inverting stage veroboard replacement!

So although free pSpice (DOS text file entry) etc. simulation software didn't become too easily available until the late 80's on PC-compatibles etc. / this didn't have much of a model library - It was a few years before release of SwitcherCAD (later LT-Spice) - most people just had to build and debug prototypes, rather than able to at least get rough expected frequency response etc.

And more specialist higher-frequency RF-simulation software is still beyond the reach of many small companies (The same with antennas, and there's probably been quite a few dodgy-claims commercially-sold ones!)

Last edited by ortek_service; 21st Apr 2021 at 3:05 am.
ortek_service is offline   Reply With Quote