Where to Look for Good *Modern* Designs for Homebrew

[Bazz4CQJ]

I wanted to build an RF millvoltmeter with a high-impedance (FET?) input and capable of working from 1-150MHz. I don’t have anywhere near the expertise to design such a thing and spent some time on Google looking at designs on various website/blogs. I finally chose a design which seemed to compare well against other designs and seemed to draw on some earlier designs which appeared to have "heritage".

The design did not work at all well and got filed in the “pending tray”. There has been a really informative thread running on the Forum in recent weeks on “FET Selection”. When I threw the design I had built in to the discussion, the reaction was that it was a deeply flawed design, best avoided, that view being based on a lot of work being done by forum members.

In years gone by, I would have looked for designs in magazines like PW, SWM or Radio Constructor, but that’s no longer an option. Can anyone suggest where to look for reliable modern designs (as may be desirable for such items as test equipment etc) for homebrew? Just as getting your news from the web needs doing with care, getting your circuits from the web is also fraught with danger

B

[Skywave]

I would just like to add that my thread of 'F.E.T selection' features a circuit that not only appears in the much-revered book "The Art of Electronics" but also in a relatively old copy of the "National Linear Applications" book. However, in each source, only the cct. is presented, devoid of any performance parameters.

Al.

[Terry_VK5TM]

For a modern design, look for circuits that use an AD8307.

There are several in the ARRL QST/QEX magazines from a few years ago (with plenty of sites on the web with variations of them).

Terry VK5TM

[Bazz4CQJ]

Many thanks for that suggestion, which I'll be looking at closely. In more general terms than broadband amps, perhaps designs originating from QST might be one good place to look for 'modern' stuff?

B

[Radio Wrangler]

Most of the better technical articles that would have gone in QST are creamed off and printed in QEX instead.

There's also a lot of project designs and circuit designs in the ARRL handbook, though I have to declare a connection.

There was some good stuff in Wireless World, but it may seem dated, but good designs age well.

David

[mhennessy]

The thing is, none of us were born with the ability to design electronic circuits. But just building circuits won't really help. Where we learn is from our failures. If you go into a project with the expectation of it being just an assembly exercise, then you're setting yourself up for disappointment and won't be in the right place mentally to debug the design (learning as you go).

I've a feeling I've said something similar recently, but I believe that the best designs are to be found in service manuals. I know that's stating the obvious, perhaps, but there's a lot to be learnt from the designers at the big companies like HP and Tek

There's also value in consumer gear as well, where the designers have had to achieve their goal at the lowest possible cost. That goes against the grain in some respects, and saving every penny is clearly not essential for a one-off project at home, but there are still valuable lessons to be found there. After all, anyone can make something work by stringing together all the building blocks they've just learnt at university, but it requires a certain flair to distil a circuit down to the absolute minimum while still meeting the specification and working reliably and repeatedly in production.

Circuits found in app-notes or text books or magazines or internet designs must be taken for what they are - unless clearly stated, that are usually little more than someone's one-off that probably wasn't rigorously tested for performance and tolerance spread, etc. That's why I'm so clear on my website that I don't expect anyone to copy my designs and have them work first time - I present them for inspiration only. That said, I've had plenty of "success stories" from folk who have made them work in some form - but then, it's all simple stuff on there - you need to pay me for the good stuff

I've never copied a complete design from anywhere - there's no learning benefit from doing that; it's barely more advanced than building a Velleman kit.

But I might copy a tiny extract - having first prototyped it and explored it and understood it, and learnt something new in the process. In other words, I expect to do my own engineering.

Ultimately, the topic in that (fascinating) FET thread is rather specialist, and I certainly wouldn't expect perfect results straight away from a schematic (which is only part of the circuit, after all). I'd go into it expecting it to not work, knowing that simulation would be an essential part of the journey as observations of many nodes will be impossible unless you already have a working low-C probe! At those frequencies, there are so many "traps" to catch you out...

By coincidence, I discovered that we had a paper copy of the Marconi 2374 manual at work yesterday (the probe itself vanished decades ago, sadly), and I have all the parts except the FET, so that would be my starting point for any experimentation I might do...