I built one from an article in RADCOM (UK RSGB monthly, March 2001
article by G3GKG) used an LF398 sample and hold chip to grab very short
pulses. Has a compensation circuit for meter under/over shoot.
Adjustable time constant so you can have it follow or remain steady with
speech. Peak hold by disabling the discharge time constant. May not be
perfect but holds up against anything else I have, quite impressive the
way it just jumps up to the power level then drops back when you stop
speaking.
MArtin, HS0ZED
Larry wrote:
> Alex,
>
> Yes! Exactly why I'm looking at as many schematics as I can find. The
> time constants are everything. Most of the kludge jobs I've seen over
> the years did a very poor job of catching the true peaks simply because
> not enough attention was paid to the times.
>
> First thing is to make sure the peak detector is fast enough to catch a
> short signal pulse. But how short, that's the question. For SSB, you
> might have pulses with a rise time of 83 microseconds and a duration of
> not much more than that. Can the circuit capture that? Most can't. On CW
> you surely need to be able to capture a single "dot" at any normal
> sending speed. My "digital" power meters used on HF do that well enough
> but I haven't have any analog meters here in the shack that do.
>
> Next item is "hang time". How long does the circuit need to hold the
> reading? That depends on what type of indicator you use. If it's an
> analog meter like I *might* use in this project, then the internal
> damping will effect the circuit design. Hang time is determined by the
> time constant of the resistor and capacitor in the peak detector in an
> analog implementation for the most part. However OP amps typically don't
> like to drive a capacitive load. If you make the cap small, you need to
> make the resistor large and long time constants will be limited by the
> leakage of the components and circuit board. Large caps will also limit
> the fast pulse capture due to finite output impedance of the op amp and
> diode as well as the slew rate of the OP amp. Details Details Details
>
> None of the circuits I've seen so far have any offset drift
> compensation. Is this not needed? Or, is it just ignored like so many
> other details.
>
> Truthfully I have not personally evaluated the Bird adapter board, or
> any of it's clones, but I have information from several others that I
> trust that it falls short for amateur service.
>
> 73, Larry
>
>
> Larry - W7IUV
> DN07dg
> http://w7iuv.com
>
> Saandy Eban wrote:
>
>> NOOO, IT'S NOT!!!!
>> At least Heathkit had on the market a dual function wattmeter (published
>> also in QST)and MFJ has also at least one model- not a tuner with
>> SWR/Wattmeter included- but a stand alone meter that can also measure
>> average or peak power. The trick is in selecting the right time constants
>> for the averaging circuits: long for peak readings, short for average
>> readings. Bad thing is that selection has to take into account the
>> mechanical characteristics of the meter movement!
>>
> _______________________________________________
> Amps mailing list
> Amps@contesting.com
> http://lists.contesting.com/mailman/listinfo/amps
>
>
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|