[CQ-Contest] Re: SO2R with Orion

[Bob Henderson]

Several of you responded directly  to my enquiry so rather than reply 
individually, I will do so here.  There may be others with an interest too.

For my SO2R contesting, I have used two FT1kMPs for the past 20 months.  
The primary radio is my original MP and the secondary is a MkV Field.  I 
am  impressed by the design philosophy of the Orion, its specs and the 
findings in the ARRL reports.  The two MPs have given good service in 
SO2R but there are some areas in which I feel they could be better.  
These include, strong signal handling, filtering flexibility (worse on 
the Field than the original MP), DSP effectiveness and the complete lack 
of ability to adjust the CW envelope without dismantling the radio and 
then the application of a soldering iron.  I ordered an Orion around the 
end of October and took delivery 10 days ago.  My plan has been to 
evaluate it and then if satisfied order a second.

So far there are many things that impress me about the Orion and I mean 
seriously impress.  There are a few niggles.  Though I believe most of 
these will either be fixed in future firmware updates or I will grow 
accustomed to them.  Then there is one BIG issue which currently has me 
feeling the Orion is incompatible with use in an unassisted SO2R 
application.

Before I set out my observations let me qualify them by reference to my 
useage:  I am not cut out for phone contesting so I don't do it.  My 
contest effort is directed primarily to CW with some RTTY. These factors 
define my terms of reference.  Further, I don't need to tell you that 
there hasn't been a major contest in the last ten days so the baptism by 
fire is yet to occurr.

My experience to date with Orion has me convinced it significantly 
outperforms the FT1kMP in all of the MPs weak areas mentioned above.

Strong signal handling is noticeably better
The filtering scheme, comprising as it does, selectable xtal roofing 
filters in the first I/F (9 MHz) and a superb I/F DSP system which 
provides b/w selectable from 100 Hz to 6000 Hz in 10 Hz steps through a 
rotary control.  As those who are familiar with the TS870 will know; 
despite some other shortcomings, the I/F DSP filtering in the 870 is 
superb.  Well in the Orion it is not as good, it's BETTER.
The default CW envelope is excellent and gradients are adjustable from 
the front panel.
Not a factor for SO2R but I do find the QSK the best I have ever used 
(T.O.R.  I currently have TS50, TS570, TS870, TS930, FT1kMP, Field, 2 x 
K2, and an Omni VI+)

The problem with the Orion for SO2R is born out of something no other of 
my radios does......

The Orion, according to its specs and according to the ARRL review has 
pretty much the cleanest synthesiser I've seen.  Whilst I am prepared to 
accept at face value these claims there seems to be a fundamental 
problem with the way the synthesiser is managed by the Orions control 
microprocessor.

There are some limitations on the set ups available for tuning rate and 
synthesiser step size  The best of those available appears to be 10 Hz 
step size with High encoder rate selected from the menus.  This gives a 
tuning rate of 2.5 kHz per revolution.  There is currently no way to set 
a 10 Hz step with 5 or 10 kHz per revolution, which I would prefer.  
Unfortunate but hardly the end of the world.

The BIG problem is this:  The Motorola Dragonball microprocessor used 
for housekeeping Orion operations appears to be way short of being up to 
the job.  There are numerous consequences of this, including:

When you press a button on the Orion control panel there is a noticeable 
delay between what should happen and it actually happening.  Fractions 
of a second but it's there.
The Orion's computer I/F runs at 57600 baud, a veritable race horse 
compared to many other radios but any benefit is lost as the Dragonball 
processes commands at an altogether pedestrian rate.
When you change band from the front panel or under computer control it 
takes about two seconds for the frequency readout to update.

Niggles but not stoppers?  Probably.  But.........

With encoder rate at maximum and step size at 10 Hz, if you tune real 
slowly the steps are barely discernable, as they should be.  The 
illusion that you are controlling the rig with an incredibly stable 
linear VFO is complete.  You only have to tune a little faster though 
and the bottom falls out of this illusion real quick!
The Dragonball can't cope.  It collapses panting on the floor.  Instead 
of processing frequency change requests, it opts instead to merely count 
the encoder pulses until it finds enough breath to perform a qsy.  How 
does this sound?  Well, rather like you have one of the nasty 1980 
vintage synths.  This really cheapens the feel of the radio but that's 
not the worst of it.

Unassisted contesters have to find multipliers quickly and efficiently.  
We can't spend too much time trawling the band at a snail's pace because 
we'll loose valuable run time.  Efficient mult hunting requires rapid 
tuning with the ability to read some code as you traverse the band 
without always stopping on stations in order to copy them.  It's an art 
you fellas will understand and one which I have been unable to practice 
with the Orion, because the choppiness of the synthesiser creates such 
an overhead of confusion.  I haven't used the rig in earnest in a big 
contest SO2R but I can imagine what will happen when I am running at 3-4 
a minute whilst trying to mult hunt in 5-7 second bursts with my second 
radio while the run radio sends exchanges under computer control.  This 
just isn't going to fly!

The Orion is superb in many ways but it can't replace my SO2R FT1kMPs 
until this problem is solved.  And I have a feeling it will take a 
significant rework of the Orion's CPU/DSP assembly to do so.

Over to Ten Tec!

73

Bob, 5B4AGN, P3F



The Orion is an interesting rig.  I've only had it for 10 days though 
I've spent a lot of time with it, so it seems like longer.  There are a 
lot of things I like about it and a couple I'm less keen on.  Nothing I 
have seen makes me wish I hadn't bought it, though I'm not sure it's 
going to be any good for the purpose I actually bought it for!

I should explain......

My game plan was to buy an Orion with the intention of getting to know 
it and then to buy a second to upgrade from my two FT1kMPs in the SO2R 
station here.  The things that attracted me to this idea included:

Better BDR
IF/DSP with good xtal roofing filters in the first I/F
Second receiver that wasn't third class, like the one in the FT1kMP
Tailorable CW envelope (with the unmodded FT1kMP I got a fair number of 
click complaints and with the modded MP I get no click complaints but 
the other guys with clicks sidle right up close and cause me QRM 
problems)  I wanted a control like there is on the TS870 so I can run a 
little dirty when I need to!)

The Orion has completely met or exceeded my expectations in all of these 
categories.  I especially like the I/F DSP filtering which is rather 
like but better than, that in the TS870.  I/F b/w is adjustable in 10 Hz 
steps from 100 Hz to 6 kHz. using a rotary encoder knob.  The second RX 
is light years better than that on the FT1k.  Strong signal handling 
capability on the primary rx is truly excellent.  The transmitted signal 
on both CW and SSB is of excellent quality.  I haven't got around to 
trying RTTY yet.  The CW envelope can be adequately tailored to meet my 
contesting needs.  In addition the menu system is good clear and easy to 
operate.  Synethiser noise factor is very low.  QSK is excellent.

I'm undecided about the AGC system currently.  It's a digital system and 
is quite different to anything I have used before.  The comments I've 
had from those who have used it for longer than me is that it's actually 
extremely good.  I suspect I will come to agree with them but it's early 
days here.

What is not quite so good about it?  Well......

Whilst styling has improved over the traditional retro look of the Omni 
series and earlier, the construction still retains a bit of the Ten Tec 
homebrew feel.
The 64 momentary push buttons on the front panel are spongy and lack 
positive tactile feedback.

Finally.....and this takes us towards the problem area.

The synthesiser has step size selectable at 1/10/100/1k/5k/10k/100k Hz.  
It then has an encoder rate selectable Low or High.  On low that 
provides 64 steps per rev and on high 256.  Only the 10 Hz setting is 
practical for CW/SSB work in my view and with High encoder rate that 
gives 2.5 kHz per rev.  I would prefer 5 kHz per rev but that's not 
available.  To get to 5kHz per rev they will probably have to introduce 
a 20 Hz step.  OK there's an optimum tuning rate issue here but that's 
not the end of the world.

So why do I think the Orion is not up to replacing my two FT1kMPs?

Well it's this:  Primary control funtions and all housekeeping are 
accomplished with a Motorola Dragonball processor, 68EZ328.  This is the 
control processor they used in the Jupiter, apparently with great 
success.  The problem is that there are a lot more things to housekeep 
in the Orion and the device appears way out of its depth.  This 
manifests itself in a number of ways:-

When you press a button there is a pause before what should happen 
actually does happen.
When you change band it takes a full 2 seconds before the frequency 
readout follows.
The computer interface runs like a race horse at 57600 baud but command 
execution time is pedestrian.

All of these are irritations but hardly stoppers.  Then there is the 
problem with the synthesiser.........

With the synth set up for 10 Hz steps you should hardly be able to 
discern audible steps even if you listen really hard.  Tuning should be 
smooth and uninterupted giving the illusion that you are using a rock 
steady linear VFO.  Well, if you tune real slowly that's the way it is 
but if you tune just a little faster (and I mean a little faster; not 
knob spinning fast!) the microprocessor can't keep up with the synth 
update requests so it gives up and just counts pulses until it finds 
time to sort itself out and actually shift the synthesiser frequency.  
This means if you tune anything faster than very slowly with a 
theoretical step size of 10 Hz you can jump in much greater increments.

For day to day use and DXing I don't think this is that much of a 
problem even though the choppy tuning is an irritation and significantly 
cheapens the feel of the radio.  In contesting it's a bigger deal!  In 
single op unassisted efforts you have to spend time tuning the band as 
rapidly as you can looking for mults.  The manic choppy sound you 
encounter doing this is a real put off and substantially detracts from 
the ability to read calls whilst tuning.  In SO2R where the operator 
overhead is a lot higher the problem is bigger still.  Running on one 
radio and mult hunting on a second is a demanding way to work.  
Frequently, you can only tune across a band looking for mults in 7-10 
second bursts while exchanges are being sent by computer on radio 1.  
 From the tests I have done, the synth choppiness completely screws up 
my ability to do this with any success and it increases stress levels 
massively.  I don't need that!

So far as I can determine, the A7 CPU/DSP assembly in the Orion would 
need to be reworked and perhaps completely redesigned to resolve this 
problem.  It's way too early to say whether Ten Tec will rise to that 
challenge.  Without a fix though, the Orion looks to be completely 
incompatible with second radio SO2R use.

There you have it Fred.  That's pretty much what I think so far.

73

Bob




Fred Handscombe wrote:

> Hi Bob
>
> Happy New Year!
>
> I was interested to read in todays newsletter that u have got an 
> Orion.  I
> was going to order one before Xmas to pick up in San Jose next week but
> they are out of stock and building for the end of Jan.  I have time to
> think as my next trip will be in March although of course they say the
> next batch is "selling fast"!
>
> What do u think about it?
>
> Is it better than the MP/Mk5?  A lot better?
>
> I have read various reviews but dont really trust ON4UNs judgement.  I
> believ they gave him one anyway!
>
>
>
> 73 Fred
> G4BWP
>
>
>
>  
>