EMRFD Message Archive 4182
Message Date From Subject 4182 2010-02-03 20:45:38 ajparent1 origin of Pretty good and the minimim gain superhet
Pretty good Superhet and the minimum gain superhet designs.
Both are interesting and resemble some of the mimimum art
SSB transceivers (receive side).
One particular item is the filters. It's been years since I've seen anything using user designed half lattice filters. I'd be curious as to the performance of those filers as depicted as they would seem narrow
(CW bandwidth).
For laughs I built a hybrid of the two using the pretty good RF amp
and the minimum gain post amp (fewer coils and caps then the fet IF)
with a filter from an dead HW32 (4crystal half lattice 3.3934mhz)
with L network matching. Not a bad 40m SSB RX only needs earsaver AGC.
Allison4183 2010-02-03 22:23:00 KK7B Re: origin of Pretty good and the minimim gain superhet
Several years ago QRP-ARCI invited me to speak at Four Days in May. I laughed and said, "well, I'm primarily a receiver designer/builder/operator and it seems to me that QRP only applies to transmitters, so what should I talk about?" My old friend George G3RJV suggested that some receivers embrace the same aesthetic as simple QRP transmitters--elegantly designed and simply constructed using a few components.
So I cracked open EMRFD and started sketching a few simple receivers starting with the primitive 3 transistor receiver that kicks off Chapter 8--adding only enough circuitry to obtain a single-signal superhet, but keeping the theme of three terminal active devices.
Those of you who know me understand that I enjoy trying things a bit off the main stream. The lower sideband ladder filter first popularized by Wes and others decades ago is a great performer--but I wanted to experiment with something a different. I had built half-lattice and cascade half-lattice crystal filters in research receivers decades ago using custom crystal frequencies. I also played with some overtone versions (similar to what you probably designed for your 6m rig.)
One day I was sketching a half-lattice in my notebook and I figured out a different way to explain its operation. To verify my model with an experiment, I needed to try a half-lattice filter with two identical crystals, one offset by a series capacitor. There is a trade-off between stop-band performance and bandwidth. If the series offset crystal is much larger than the parallel capacitance of the crystal, then the parallel crystal capacitances still cancel in the half-lattice configuration. As the series capacitor gets larger, the bandwidth increases to something usable for SSB, but the simple model collapses. But it still works even at SSB bandwidths using a 27pF crystal. The only thing elegant about the design is that the bandwidth is controlled solely by that one series crystal. It's a circuit that "wants to work."
OK Ed, I know you won't let me get away with that bit of balderdash and piffle--it's a circuit that exhibits a broad global maximum over all the variables....even the ones I haven't yet figured out how to define.
In these particular experiments, the half-lattice filter with two identical crystals and an offset capacitor works about as well as a two crystal lower-sideband ladder filter, but it does require a tightly coupled transformer. I understand how it works well enough to use it again--but I suspect that the ladder filter makes more sense for most applications.
I've built at least 20 receivers for every one I've written about--most of them with some unusual experimental circuitry. It's a lifelong habit. I secretly built a crystal radio from a schematic in a boys book when I was 7 years old, to listen to forbidden rock and roll stations under the covers at night--so I've been designing and building receivers for half a century.
Even when I build one for pure research instrumentation applications, I bring it home, connect an antenna, and make a contact or two. Until you actually listen with an antenna, you don't even know what you don't know. Many circuits that look great in the simulator and on the bench have serious flaws in practice.
So the "Simple Superhet" and "Pretty Good Superhet" were experiments with a new (to me) type of crystal filter, based on design sketches for simple receivers using only 3 terminal devices. They work well given their simplicity, when used with a full sized dipole in the Northwest US. Neither has enough gain to need AGC, unless you have a very local 40m station as a neighbor.
Except in very unusual circumstances, basic HF receiver performance has been good enough for a few decades. Wes and others attacked the last major problem and solved it in the late 70s. Since then a few designers have continued to push the dynamic range envelope. But I seldom experience any intermod problems at HF, even with the simplest receivers I build. So I've backed off from designing receivers that attempt to improve on some spec or other, and have instead focused on designs that are interesting or enjoyable in some way.
I've been inspired for decades by the writings of the late Phil Bolger, a naval architect who spent a lifetime designing interesting boats. The titles of two of his books: "Boats with an Open Mind;" and "40 Odd Boats," will provide a clue to his thought process. Many of his boats are design sketches to try out an interesting idea. He called his design sketches "cartoons." A few years ago Phil Bolger came up in a conversation with Bob Larkin. He said, "I'm building one of his designs in my garage this winter." I built a Bolger designed boat 22 years ago, and restored it this year. So 2 of the 3 authors of EMRFD have built boats designed by the same obscure guy from Maine. That can't be a coincidence.
Anyway--I brought that up because I've thought about writing a book called 40 Odd Receivers, or perhaps Receivers with an Open Mind. If I did, the "Simple Superhet" and "Pretty Good Superhet would be featured as examples of the type. Not optimized for anything--just built with parts I had4185 2010-02-04 06:55:52 w0ep Re: origin of Pretty good and the minimim gain superhet
Just for my edification, what was the "last major problem"?
>
>
> Except in very unusual circumstances, basic HF receiver performance
> has been good enough for a few decades. Wes and others attacked the
> last major problem and solved it in the late 70s. Since then a few
> designers have continued to push the dynamic range envelope. But I
> seldom experience any intermod problems at HF, even with the simplest
> receivers I build. So I've backed off from designing receivers that
> attempt to improve on some spec or other, and have instead focused on
> designs that are interesting or enjoyable in some way.
>4186 2010-02-04 07:05:34 Chris Trask Re: origin of Pretty good and the minimim gain superhet
> I very much enjoyed reading this exchange.
>
> Just for my edification, what was the "last major problem"?
>
> >
> > Except in very unusual circumstances, basic HF receiver performance
> > has been good enough for a few decades. Wes and others attacked the
> > last major problem and solved it in the late 70s. Since then a few
> > designers have continued to push the dynamic range envelope. But I
> > seldom experience any intermod problems at HF, even with the simplest
> > receivers I build. So I've backed off from designing receivers that
> > attempt to improve on some spec or other, and have instead focused on
> > designs that are interesting or enjoyable in some way.
>
Chris Trask
N7ZWY / WDX3HLB
Senior Member IEEE
http://www.home.earthlink.net/~christrask/4187 2010-02-04 08:22:38 ajparent1 Re: origin of Pretty good and the minimim gain superhet
FB Rick. Yes I've done same and experimented with the simple superhet
idea for years. No two alike and many worked extremely well and a few not so much. All however were worth the effort as they did show what was required and what was purely optional.
Allison
4188 2010-02-04 08:34:04 ajparent1 Re: origin of Pretty good and the minimim gain superhet