e: [Icom] re pro output power]

[George, W5YR]

Bill, your note prompted me to revisit the SSB performance of my PRO using
various mics and adjustments. The results confirm that indeed the SM-8 can
provide full output from the PRO, and I would expect the same of the SM-20,
although I did not have one to test.

First though, we must be careful to note that there are at least four ways
to measure the SSB power output of the PRO or any other transmitter:

1. oscilloscope trace of the output r-f envelope

2. transmitter panel meter (and with the PRO, the digital meters as well)

3. an external average-power meter

4. an external peak-power meter

Of these, only the first is capable of responding accurately and completely
to the question "How much SSB output power can the transmitter generate?"
followed closely by an equally important proviso " and maintain a clean
output waveshape with no clipping or flat-topping."

The remaining methods each have inherent errors associated with calibration
accuracy, meter dynamics such as response time and damping factor, and the
like. Thus, we must be careful in interpreting the readings of any of the
metered approaches in comparison with the scope approach.

For these tests, I adjusted the PRO to produce 100 watts RTTY
single-frequency output as measured on the MFJ 989C wattmeter. This meter
tracks the output power of the Kachina 505 DSP at any reading from 10 watts
to 100 watts to within the width of the indicator needle. The 505 output
power is controlled digitally to within 0.1 db. Hence, I am confident in
relying upon the 989C wattmeter for these largely comparative readings of
both average and peak envelope power.

The scope gain was set to provide a display of the unmodulated r-f output
envelope that was 4 cm peak to peak. Under these conditions, any SSB
modulation waveform that created peaks reaching the 4 cm limits was by
definition 100% modulating the transmitter, i.e., producing 100 watts peak
envelope power. By the same token, any signal reaching those limits and
evidencing clipping distortion disclosed a condition of overdrive
distortion in the transmitter. 

This all too frequent situation with SSB transmitters makes the meters read
higher but unfortunately the added power readings are largely the result of
distortion products. So, my boundary condition for "100% modulation" was to
observe that the voice input waveform caused peaks that reached the 4 cm
boundaries and yet remained clean and unclipped at those levels.

The Transmit EQ for the PRO was set at 0 dB for the low end and +5 dB for
the high end. Four microphones were used: the SM-8; the HM-36 hand mic
provided with the PRO; the Heil HM-i; and the Heil GM-5 Goldline. The first
three were plugged directly into the front panel mic jack and the internal
mic amplifier circuitry was used. The Goldline was operated through a
Behringer MX602A Mixer/EQ which was set for flat EQ; only the output level
was varied to control the input signal level to the ACC1 connector on the
rear panel of the PRO.

The audio signal used for these tests was yours truly speaking the
time-honored and consistent word "FIIIIVVVVVEEEE" as a test waveform. This
word has the unique property of providing salient peaks which are easily
observed and allow for ready detection of clipping distortion. All audio
was generated at about the same voice level with the mics about 3-4 inches
from my lips.

The followng table compares the results of operation with these four
different mics:

 MIC	 MIC GAIN  PEP (Scope)	PEP (PRO Meter*)  PEP (MFJ Meter) AVG (MFJ
Meter)	
-----    --------  -----------  ---------------  ---------------
--------------- 
HM-36  	11 o'clock  100 watts        75 watts       50-60 watts     30-35
watts

SM-8     9    "      "    "           "   "           60     "      30-40  
"

HM-i     3    "      "    "           "   "            "     "      20-30  
"

GM-5     N/A         "    "           "   "          50-60   "      25-30  
"

-----------
* In every instance, the PRO digital meter and the analog panel meter
rapidly rose to a reading of about 3/4 full scale or 75 watts with each
input word. The analog meter, in particular appears to be very lightly
damped compared with the MFJ meter. Because of the rapidity of the PRO
meter movements both up and back, their maximum reading could only be
estimated, as inertia and damping prevented a 100% indication. 

These results are remarkably consistent, considering the inevitable
variability of the speech sounds used for excitation. In particular,
comparing the scope waveforms with the meter readings, one can conclude
that

1. the PRO is capable of being driven to 100 watts PEP clean SSB output
from voice input with any of the four mics that were tested using
reasonable mic gain settings and speech levels.

2. when driven to 100 watts PEP output, the PRO output signal remains clean
and unclipped; for all tests, the digital ALC reading was driven to the
maximum and the lack of peak clipping is a fair indication of the
smoothness and effectiveness of the ALC action.

3. analog meters fail to read the full 100% PEP with normal speech input
since not all speech segments contain peaks that consistently reach 100%
PEP, and few analog meters are so lightly damped that they can follow
speech elements closely enough to give accurate readings. Professional "VU"
meters are the exception, but the PRO analog meter is surprisingly agile in
this respect.

4. due to the peak to average power ratio of some 10-14 db in the typical
unprocessed male voice, it is not surprising to find that SSB analog power
meter readings can be quite low with normal speech compared to actual PEP
output. It is not uncommon for a typical average-power wattmeter to read in
the vicinity of only 10-15 watts when a 100-watt transmitter is being
driven to its full PEP output.

After doing these tests, I have concluded that Julius Jones' remark about
the PRO being "set up for only 60 watts SSB output" was probably based upon
observations of a peak-reading analog wattmeter. The PRO *is* fully capable
of being driven to 100 watts PEP output power with any conventional mic, it
would appear, but one must bear in mind that analog metered output
indications may be substantially lower than the 100 watts actually being
developed on modulation waveform peaks.

One final point: although the PRO is designed and rated for 100 watts PEP
output within acceptable distortion specifications, any effort to extend
the power output - such as by changing internal pot settings, etc. - to
gain a few extra watts will (a) cause the output device temperatures to
rise rapidly and (b) the distortion products to rise even more rapidly.
This effect is clearly pronounced when operating PSK31 where a 40-50 watt
PEP output is about the limit if IMD readings of -30 db or better are
desired.

I hope that this more detailed quantitative information will make up for
any confusion my earlier posting may have caused.

73/72/oo, George W5YR - the Yellow Rose of Texas         
Fairview, TX 30 mi NE of Dallas in Collin county EM13qe   
Amateur Radio W5YR, in the 56th year and it just keeps getting better!
QRP-L 1373 NETXQRP 6 SOC 262 COG 8 FPQRP 404 TEN-X 11771 I-LINK 11735
Icom IC-756PRO #02121  Kachina 505 DSP  #91900556  Icom IC-765 #02437

All outgoing email virus-checked by Norton Anti-Virus 2002
  
Bill Tarkington wrote:
> 
> George, realize your comments are relative to the factory hand mike's
> insufficiency.  For those of us who are using the front panel mike
> connector, can we safely assume the SM20 and/or SM8 desk mikes will drive
> the transmitter to full output (assumed to be 100 watts)?