[aprssig] Kenwood D700 TNC Transmit Equalization Screwed Up!
Stephen H. Smith
wa8lmf2 at aol.com
Mon Apr 4 01:09:06 CDT 2005
I am in the process of producing an audio CD recording of various types
of APRS packet signals. I intend for it to be a source of exactly
repeatable noisy and distorted signals for comparing the performance of
different TNCs under various less-than-ideal conditions. The idea is to
be able to use any audio CD player or old CD-ROM drive to play back test
signals into each TNC in turn and compare the results. I have been
using an audio editor (Adobe Audition - formerly known as Cool Edit) to
record various types of packet bursts off-the-air. These recordings are
being made using the extremely low distortion wide-band demodulator of
my IFR-1500 service monitor (instead of the discriminator of an
ordinary radio) in order to get the highest-possible-quality reference
signal unbiased by the IF and audio response characteristics of any
It's generally accepted that 1200 baud packet is transmitted through the
MIC jack of a radio, and therefore should acquire the same audio
preemphasis as a voice signal (i.e. the 2200 Hz tone of standard
1200-baud should have almost TWICE the deviation as the 1200 Hz tone).
For the best TNC receive performance, corresponding DE-emphasis needs to
be used at the receiver, either by taking the RX audio off the speaker,
or by de-emphasis networks inside the TNC. (Higher-speed formats are
normally transmitted "flat" due to direct coupling to the transmitter
In the course of capturing signals with the recording setup, I was
shocked to discover that the Kenwood D700 transmits 1200 baud "flat"
with no pre-emphasis at all!
Here are two GIF images that are screen captures of the Cool Edit
screen showing the envelope of the AFSK 1200 baud packet burst as
captured off the IFR's discriminator:
This is a display of the waveform over an entire 1/2-second Mic-E burst
from the D700. You can see the sync tones at the beginning of the burst
- note the long delay before the data starts. The very tall "grass" at
the extreme left and right of the picture is unsquelched white noise
before and after the burst. If you look closely at the right side, you
can see a millisecond or so of dead carrier between the end of the
packet tones and the radio unkey.
The Cool Edit display lets you zoom into the display until you can see
the individual cycles of the audio wave form. Here is about a 700%
enlargement zoomed in at the end of the sync and the beginning of the
data. Note that the low tone (fewer cycles per unit distance) is
actually HIGHER in level than the high tone!
The waveform display clearly shows that the D700 is transmitting 1200
baud WITHOUT the customary preemphasis. This would seem to suggest that
D700s (At least at 1200 baud.) are at a real disadvantage compared to
other APRS devices when transmitting to digipeaters and other users
employing standard receive de-emphasis. De-emphasis applied to a signal
already transmitted "flat" will result in the high tone being 6 dB LOWER
in level than the low tone. This is especially bad news for TNC2s (i.e.
UI-Digi installations) whose modems have a well-known intolerance for
high tone being at a lower level than the low tone.
Stephen H. Smith wa8lmf (at) aol.com
Home Page: http://wa8lmf.com
New/Updated "Rev G" APRS http://webs.lanset.com/wa8lmf/aprs
Symbols Set for UI-View,
UIpoint and APRSplus:
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the aprssig