[Fox_Tango] "Fix" circuit on FT-101E speech processor board

[Davis_R]

Hi Wiley, How did things work out?

I am not as familiar with the PB1494 as I am with the PB1534, but it is
actually simpler.

The PB1494 uses a compression amplifier, Q1 followed by 2 diodes D1 and D2
to clip the envelope and the output level is set by VR1. The signal is then
filtered by XF1 to remove the harmonics generated by the diodes' clipping
action.

If the processor is OFF, the IF signal is amplified by Q2 to compensate for
the loss of having the extra filter in line.

The method of alignment is to appy a 1KHz audio signal to the Mic input (I
use the patch input on the back).
Set the mic gain to 12 o'clock, adjust audio input to give 50mV input to the
processor.
With the processor OFF, adjust VR2 for 50mV out at the processor output.
These will be at IF, 3180KHz.

Then switch processor ON, and VR1 fully CCW, peak T1 for maximum processor
output, then adjust VR1 so the processor output equals processor input
(unity gain.)

If it is properly adjusted, increasing the audio to give 100mV input to the
processor should result in about 60mV output. So it sounds like you did
okay.

I don't know of any problem with having a little gain, but I don't know how
to tell when you have too much.
The idea of the processor is to be transparent when it is OFF, and to limit
the peaks while boosting the lows (amplitude wise) of your speech, thus
raising the AVERAGE power.
And  by the way, it really does! When I have the processor OFF, my average
power output, as indicated by my wattmeter reads around 25 - 30 watts.  When
the processor is ON, the wattmeter average reads about 40 - 60 watts.  All
the while, the peaks, using a PEP meter, actually drop a little, from
140-150 watts, down to maybe 110 watts.  So the peaks are lower, reducing
the bandwidth, generating less harmonics, and a 3 dB increase in average
power.  On-the-air reports confirm approximately 3 dB increase in S meter
readings with the processor ON vs. OFF

Pretty cheap 3 db linear aye?

73 and God Bless,
Rick WD8JJA.

----- Original Message -----
From: "Wiley Sanders" <wsanders@wsanders.net>
To: "Davis_R" <WD8JJA@arrl.net>
Cc: <fox_tango@mailman.qth.net>
Sent: Friday, March 14, 2003 12:25 PM
Subject: Re: [Fox_Tango] "Fix" circuit on FT-101E speech processor board


> Thanks Rick,
>
> I have the older board, PB 1494, it does not have the ALC circuitry on it
and
> has two VR's for setting the gain with and without the compressor in.
>
> I followed the alignment procedure in the original operator's manual (I
have a
> separate one-page document on the processor that has a slightly different
but
> equivalent procedure.) The objective of the alignment procedure is
basically
> to feed a 1000 Hz tone into the board and set the output level with the
> processor on and off equal to the input level.
>
> I found the gain with the processor on to be wildly over-set - the VR was
> cranked up all the way and the output level was about 160mv using my
homemade,
> uncalibrated RF probe. I set VR1 and VR2 to provide an output level equal
to
> the input level with compression on and off, and the Mic Gain at 12
o'clock.
> The ALC reading is now between 0.3 and 0.4 for normal speech.
>
> I had a couple of QSOs on 20 after I made the change, and everyone said my
> audio was fine, the test will be on 10 to see if it made any diference on
that
> band. My output power is down to about 50W on 20, 15, and 10, measured
putting
> a 1000Hz tone into a dummy load. Thats OK if that's a tradeoff good audio.
(By
> the way DigiPan provides a clean, useful, uncalibrated, but free audio
> oscillator by way of its "Tune" function.)
>
> My only question is: Is it permissible to maybe allow a little gain in
this
> stage? I had no complaints and a few complements about my audio until a
few
> days ago, when I altered my antenna and started getting some RF in the
shack,
> and I had about 80W output on 20 through 10.
>
> -Wiley KF6IIU
>
> Now, more than ever, QSL.NET is in urgent need of your financial support.
>
> Please help QSL.NET by sending a donation now.
>
> ________
> _______________________________________
> Fox_Tango mailing list
> Fox_Tango@mailman.qth.net
> http://mailman.qth.net/mailman/listinfo/fox_tango
>