From: Henk de Groot (ormvbn@bishoplynch.org)
Date: Tue Aug 20 2002 - 23:47:41 EEST
Hello,
WARNING: Technical content!
Here is more information about the problems I found in the TH-D7. It has
been posted before but seeing the discussion here it may be of interest.
Kind regards.
Henk.
------------------------------------------------------------------
Hello All,
I promised a lot of people to report about the latest Kenwood update for
the TH-D7E handheld tranciever with build-in TNC.
For those who are not aware of it the following. Kenwood has now at
least 3 versions of the TH-D7E in the field.
The first version, which is usually referred to as v1.0, was the
European version of the TH-D7A but was unfortunately lacking some
functionality essential for Europe. This was KISS support and DAMA
support.
The second version, v2.0, fixed this. This second version also corrected
some other things like the ability to use 9600 baud in APRS mode -
needed for watching DX clusters - and the ability to handle higher
precision NMEA position strings, like transmitted by newer eTrex GPSses
for example.
The third version, which also identifies itself as v2.0, fixed an
annoying bug in the second version, known as the "long packet"
problem. Kenwood can update existing TH-D7E's, but unfortunately only
the TH-D7E v2.0. I know this fix is available in the UK and in Belgium.
My unit was send to Belgium via my dealer. A round-trip can be as short
as 2 weeks, but for me it was over 8 weeks due to the fact that bacause
of warantee I had to send it via the dealer. So far I've not heard it
was available in Germany, which kind of surprises me.
I just got my TH-D7E back and tested some of the known problems.
First of all some data about the upgrade itself. From the outside there
is no visible difference between this new version and the old v2.0
version. None of the menus are changed and even the TNC startup message
is the same as before the upgrade (same date, same 2Chip version
number). To me it seems that the only way to verify if it is new or an
old TH-D7E v2.0 is by watching the behaviour with the known bugs of the
old v2.0 version.
---------------------------------------------------------------------------
Problem: Long UI packet problem in APRS
Symptom: Decoding packets stopped, Beacon transmission stopped. Cycling
the TNC is needed to bring APRS to life again.
Report :
In the TH-D7E v2.0 this problem occured on reception of a packet which
length, written down in TAPR TNC format, exceededs 195 characters. This
was very reproducable. After the upgrade I tried to reproduce the
problem. It looks like the Engineers at Kenwood found the weak spot.
Even UI frames with 256 bytes of data, the maximum length accroding
to the AX.25 spec, doesn't do any harm anymore! I'm very pleased with
this I can tell you!
I let it run for half a day on our heavily loaded LAP QRG, after that
torture, which killed the APRS function for sure with the old v2.0, the
TH-D7E still ran happily.
After I pressed the TNC button once more to enter Packet mode I noticed
that it hung up on me, I was not able to send anything. So still
something gets messed up after reception of the long packet, but I don't
think this is a problem. If TH-D7E is just switched on an then set to
Packet there is no problem as far as I could see.
---------------------------------------------------------------------------
Problem: Hangup in KISS mode after some time of use
Symptom: "STA" and "CON" burning on the display, no transmission and
reception of packets anymore, no response from the TNC anymore
Report :
This problem occured with WinAPRS in KISS mode on the TH-D7E after a few
hours. There was however a much quicker way to get this hangup. That was
by using the TH-D7E in KISS with the DIGI_NED digipeater software. Then
the hangup occured in minutes. This problem still exists. It is however
a *lot* better now. It now takes about 30 minutes to have the same
problem again, using the digipeater method. So definately something is
improved. Unfortunatly its not eliminated however. With the old v2.0 it
was possible to regain control by sending the command "TC 1", which
switched the TNC off (important, "TC" in capitals). Then with "TC 0" the
TNC could be activated again. This method still works with the new v2.0.
---------------------------------------------------------------------------
Problem: PERSIST works the wrong way arround
Symptom: With PERSIST=255 the TNC never transmits instead of always
With PERSIST=0 the TNC always transmits instead of never
Report :
This is anoying if you use the latest version of TFPCX which, when
connected to a DAMA node, programs the PERSIST parameter to 255 to get
better DAMA behaviour. The TH-D7E refuses to transmit in that case. This
can be verified in terminal mode. Just set PERSIST to 255 and try to
transmit something. The transmission will be queued in the TH-D7E. If
you then set PERSIST to 0 all the queued data is transmitted. On the new
v2.0 this problem still exists. A bypass for those who want to work in
DAMA with the latest TFPCX version can switch the PERSIST channel access
method off and use the DWAIT method with a time of 0 for direct access.
To start KISS use the following commands to do this:
cmd: PPERSIST OFF
cmd: DWAIT 0
cmd: KISS ON
cmd: RESTART
WARNING: Only use this with DAMA, not for normal connections! For normal
connections you can use this table to find out the correct PERSIST
values which depends on the total number of the stations on the channel:
nr of users TAPR THD7E
on the channel PERSIST PERSIST
1 (only you) 255 0
2 128 128
3 85 171
4 64 192
8 32 224
The assumption here is that the TH-D7E persist is 256 - PERSIST
---------------------------------------------------------------------------
Problem: RESptime timer doesn't work
Symptom: Slow connection because the TH-D7E doesn't take initiative when
the node doesn't send a POLL bit on the last I frame
Report :
A lot of protocol stacks used in Western Europe do not send a poll bit
on the last transmitted I frame before witching to reception. This makes
the gaps between the transmission and the RR ack a bit wider so other
stations have a chance to "break-in". The RR response of the reciever is
triggered by the internal RES timer instead. In de Kenwood this timer
doesn't seem to work. Now the link becomes slow bacause the TH-D7E waits
until the node or BBS polls the TH-D7E due to an elapsed FRACK timer.
This has not changed with the new v2.0 version.
---------------------------------------------------------------------------
Problem: Slow 9600 baud in KISS
Symptom: Only 1 frame at the time is handled in KISS at 9600 baud
Report :
It looks that after reception of 1 frame the TH-D7E is so busy to
transmit it to the serial KISS port that the next frame is always
missed. As far as I could see this has not changed with the new v2.0
version. In 1200 baud there seems to be no problem in KISS (except for
the hangup problem mentioned above), but that was already the case with
the previous v2.0 version.
---------------------------------------------------------------------------
Summary:
As far as I could see the advantage of the upgrade is the ability to
sustain long packets in APRS mode, which was, for me at least, the most
annoying problem and I am really pleased that Kenwood managed to fix
this. Futhermore the KISS mode seems to be more stable.
Although the other problems still remain, the advantage of this upgrade
is worth more than enough to have to part with the TH-D7E for a few
weeks. After sending it off I started to realize however how much I
missed this nice handheld device, I'm glad it is back now. The upgrade
itself was free of change (under warantee), I only covered the postage
from me to my dealer and back.
I hope this information was usefull for you, and many thanks to Kenwood
for listening and fixing the most annoying problem. I am a happy man!
Kind regards,
Henk.
I accidently found a new small bug. Set the TNC in APRS mode. Press LIST
once. Now if you press the PTT the TH-D7E jumps back to the normal
screen. If you however do this while recieving an APRS message the
transmitter goes on and stays on even after release of the PTT button!
Pressing the PTT again for a short time stops the transmission. I don't
know if the old v2.0 version has this too.
-
To unsubscribe from this list: send the line "unsubscribe linux-hams" in
the body of a message to nbm@akvasmart.cl
More majordomo info at http://vger.kernel.org/majordomo-info.html
This archive was generated by hypermail 2b30 : Tue Aug 20 2002 - 23:53:14 EEST