[Orion] V2.059d Vs. v1.373b5 Noise Reduction Code + My experiences with 1.373b5 + End of Current Chapter

[Joe Giacobello]

I just wanted to follow up on my experiences with Bill's recommendations 
on AGC settings, etc.

When I wrote the preliminary assessment below, I had not done any A/B 
comparisons with my FT-KMP.  I now have done so on 20, 30, 40 and 160M 
CW.  I tried to keep the operating parameters for both radios as close 
as possible.  I was using an Inrad 600 Hz roofing filter and 400 Hz DSP 
filter in the Orion and the Inrad roofing filter in the MP.  The IF 
filters in the MP were an Inrad 400 Hz cascaded with the stock 500 Hz.  
The MP's DSP features were off for the comparisons.

When using FW 1.373b5 in the Orion, the MP consistently outperformed the 
Orion.  In most cases, the differences were dramatic, and I was tempted 
to record them because I know there are those here whose experiences are 
just the opposite.  (I also know Bill has an MP and has favorably 
evaluated the Orion vs the MP.)  I am not talking about selectivity 
here, but comparative S/N.  Invariably, identical signals on the MP 
stood out well above the noise while they seemed to be competing with 
the noise on the Orion.  In making these comparisons, I repeatedly 
switched between the two units to make sure I wasn't hearing changing 
band conditions.  (I had set up the Orion as Bill had recommended, but I 
did try manipulating RF gain and decay in order to optimize the 
signals.  Usually, I was using 18 dB attenuation on the Orion.)

I then switched back to FW 2.059d.  Here, as best as I could tell, the 
performance of the two units was a lot closer.  The problem was that the 
audio level and punch on the Orion were so low, it was hard to compare 
them at equal levels even with headphones.  (I have both units feeding a 
dual input KWD speaker, and I take the headphone output off the 
speaker.)  Nevertheless, it seemed that the MP was better.  (I should 
also add that I had been corresponding with Paul Clinton, TT's service 
manager, on another subject, and he said that FW 1.373b5 had been 
developed to solve a specific problem reported by someone and that the 
resulting improvements had been incorporated in subsequent FW versions.)

It is also important to say that I was experiencing hardware problems 
with the Orion, but these seemed unrelated to its receive performance.  
(The transverter power output control did not work properly and the 
transverter transmit audio was quite distorted.  In addition, even 
though I had changed the optical encoder, which Paul had kindly provided 
free of charge, the main dial was not working and I was using the remote 
pod to tune.)  Maybe there is a clue here to my unit's inferior 
performance, but it seems a stretch to me.

Nevertheless, I shipped my Orion back to TT on Friday for repairs.  I 
also described the current comparative receive issue and asked Paul to 
carefully check the receive performance of my unit.

Let's see what happens, but at this juncture my objective assessment is 
that MY MP outperforms MY Orion in MY shack.  Obviously, YMMV.

73, Joe
K2XX



Joe Giacobello wrote:
> Bill, I just wanted to let you and the list know that I had a chance 
> to check out your procedure for RF gain, etc. adjustment this weekend 
> on FW 1.373b5.  There was a vast improvement in performance, although 
> I didn't spend as much time as I wanted to with it, and I'm still 
> evaluating.
>
> I wound up using 18 dB of attenuation and the RF gain at 60-80 with a 
> threshold of 1.02 uV on 30 and 40M CW.  When I invoked the NR, the 
> noise (gain) did increase, but readjusting the RF gain got it back 
> under control.  I haven't really checked the settings out on very weak 
> sigs, etc, but there's no doubt that a vast improvement was obtained.
>
> I did have to do a master reset (after the initial reset after loading 
> the new FW) when I lost all output power.  Others have reported 
> similar glitches with 1.37x FW.  I had occasionally observed power 
> output, mike gain or audio output losses with 1.37x myself.
>
> Another issue:  If the NR works by narrowing the receive bandwidth in 
> a manner similar to manually reducing selectivity of the filters (I 
> presume that the NR uses different algorithms for CW and SSB.), why 
> have an NR at all except for its automatic feature?   Interestingly, I 
> had reported that FW 2.059d seem to clip the signals when using NR at 
> a "1" setting.  However, I also noted that the clipping could be 
> minimized by very carefully centering the sig in the passband, which 
> seems to support the argument that NR works by reducing BW.  OTOH, 
> with weak sigs no amount of centering helps.
>
> 73 and thanks for your help.
>
> Joe
> K2XX
>
> Joe Giacobello wrote:
>> Bill, thanks.  I remember Sinisa's paper, but I never had much luck 
>> with it.  Nevertheless, I'll reinstall 1.373b5 and try exactly what 
>> you suggest.  There is no doubt that there was too much RF gain.  
>> However, I did manipulate the attenuation, RF gain, and threshold and 
>> I would never use the preamp on 40M anyhow.  However, I didn't 
>> manipulate them in an orderly fashion such as you have described.  
>> I'll let you know how I make out.
>>
>> Thanks for your help.
>>
>> 73, Joe
>>
>> 73, Joe
>>
>> Bill Tippett wrote:
>>  
>>> K2XX:
>>>  >I replaced 2.059d with 1.373b5 last night and the change in 
>>> performance
>>> was both dramatic and negative.    There was an enormous increase in RF
>>> gain, which was further increased when invoking NR.  I tried
>>> manipulating the Threshold, Decay, Attenuator and NR setting without
>>> much improvement.  To make 40M CW even marginally listenable, I had to
>>> use 12 dB of attenuation.  The NR seemed to act properly with the
>>> attennuation.  The Threshold setting seemed to have no effect and there
>>> seemed to be a non-linearity in the RF gain control.
>>>
>>>          Almost certainly an indication RF GAIN is set too high.
>>> Also, AGC Threshold and RF GAIN are now linked in all V1
>>> versions following 1.371.  To set RF GAIN correctly:
>>>
>>> 1.  Set AGC Threshold = 1 uV for all bands, all AGC options.  Set AF 
>>> GAIN to mid-scale.
>>> 2.  Tune to an area of the band with antenna connected but no
>>> signals.
>>> 3.  Adjust the following, in order, until you cannot hear band
>>> noise when switching between no antenna and your antenna (e.g.
>>> ANT 1 vs ANT 2 with nothing connected to ANT 2):
>>>
>>> a.  Preamp Off
>>> b.  Attenuator (6/12/18 dB)
>>> c.  RF GAIN
>>>
>>> 4.  When you can no longer hear band noise, then increase RF
>>> GAIN until it just becomes comfortably audible (but not beyond
>>> that point).
>>>
>>>          Anytime I hear people complain about Orion being
>>> "noisy", it is usually a sure indication there is too much
>>> gain in their setup.  The above is adapted from a paper
>>> Sinisa YT1NT wrote below, but it was prior to 1.372 when
>>> Threshold and RF GAIN were integrated into one control.
>>>
>>> http://www.geocities.com/va3ttn/UsingOrionRX.pdf
>>>
>>>                                          73,  Bill  W4ZV
>>>
>>>
>>>
>>> _______________________________________________
>>> Orion mailing list
>>> Orion at contesting.com
>>> http://lists.contesting.com/mailman/listinfo/orion
>>>
>>>
>>>       
>>
>> _______________________________________________
>> Orion mailing list
>> Orion at contesting.com
>> http://lists.contesting.com/mailman/listinfo/orion
>>
>>
>>   
>