Matching a monolithic xtal filter

And a 2b's IIP3 (with no pad at the front end) is how much? Under the same conditions, the NF is how much?

Cheers, Tom

I don't remember. Let's put it like the horsepower of a Rolls Royce: Adequate. Even 30 years later that receiver outperformed new gear in terms of operation in the vicinity of strong transmitters. Ok, it's got tubes which gives it a head start.

Why do you think a modest Q-multiplication would worsen that, provided there is plenty of supply voltage? If you just want to reign in the LC before a filter you don't need to reach Qs of a gazillion.

Other question: Where do you have +55dBm right now? And why do you need that much?

Why: because at that level, a couple 0dBm signals from 7MHz broadcast transmitters will yield only -110dBm third-order product that will make unintelligible a microvolt-level signal.

Where: look at what people are doing with H-mode mixers, for example; but that's not the only place.

Cheers, Tom

Besides the nonlinearity problem, Q-multipliers aren't the quietest things on the planet either. That isn't usually a big problem at HF, but (repeat after me) Negative Resistors Don't Have Imaginary Noise. ;)

Cheers,

Phil Hobbs

Sure. However, Q-multipliers are usually placed after there has already been some gain. Also, often noise is less important than selectivity. It's like pulling a trailer up a hill where your car won't run very quiet and not the cleanest but you've got to get up there ;-)

Ok, but at some point it's "good enough". At 7MHz you'll rarely deal with uV level signals. At 20-30MHz that can happen though. Point is, the old Drake 2B performed marvelously in such situations. In ham radio the field days and the contests were the major challenges. Not multiband because you'd be running the 2nd transmitter at another band and the input filters take care of most of it but it was the guys on the next hill half a mile away, competing against your group with a huge antenna and lots of power. Often they were just 20-30kHz away. All the older generation transistor rigs fell off the rocker in that situation. BTDT, many times.

In CW mode (morse code) a steep enough mechanical or crystal filter was not affordable in those days unless you had one of the Rockefellers on the team. So the Q-multiplier saved the day.

On Sep 12, 8:57 am, Joerg wrote: ...

The best currently available is still not quite good enough, I'm afraid. There are those who have valid needs for better. Though you sing the praises of the Drake 2B, I've seen other reports on it that are not nearly so glowing, and we still have no actual measure of its performance. I seriously doubt that it's anywhere near as good at the good modern designs that have IIP3-NF > 40dB, and phase noise well below -100dBc/Hz at 100Hz offset. There's certainly no _need_ for a Q multiplier in front of a first crystal filter to achieve this performance, and I doubt that you could add one there while maintaining the same level of noise and dynamic range performance. I presume the 2B's Q multiplier was at or near the end of the IF strip, not at the RF amplifier or even at the mixer output.

Cheers, Tom

After filtering AFAIR. Again, I agree that xtal filters generally do not need a pre-filter, that was brought up by another poster. In the case of a large ham radio contest that is very different. While you may have a nice 2.3kHz filter for SSB and if you splurged maybe another 600Hz filter that often wasn't enough when the narrow CW portions of a band where crammed with others who also wanted to win that contest. I now own a 300Hz filter here but as a young lad back then I certainly did not have $150 for such luxury. Same for the others. Now, the $100 or so for the booze during the contest, that was not a problem ;-)

The other thing you could do with a Q-muliplier was often even more valuable. You could use it as a very sharp notch filter. All in all I find these things rather useful.