[TenTec] TT Synthesizer Accuracy (Orion 2)

[Ron Castro]

Gary:

I believe your assessment is correct, especially in the later firmware versions that set up the 'cancellation' routine.  This also fixed the slight difference in frequency between the main and sub RX's, or at least radically improved it.  The bottom line is that the only thing that needs to be adjusted is the Master Oscillator.

I also have an 8640B, a lucky e-bay purchase that works great, but I still wouldn't trust it for the kind of accuracy I want in my Orion II.  Only WWV is good for that and once calibrated, I use the Orion II to calibrate the 8640B.

          Ron Castro
           N6IE
    www.N6IE.com

Member:
                  ARRL
     Redwood Empire DX Assn.
 Northern California Contest Club
Northern California DX Foundation
  Society of Broadcast Engineers



-----Original Message-----
From: TenTec [mailto:tentec-bounces at contesting.com] On Behalf Of Gary J FollettDukes HiFi
Sent: November 12, 2015 8:51 PM
To: ronc at sonic.net; Discussion of Ten-Tec Equipment
Subject: Re: [TenTec] TT Synthesizer Accuracy (Orion 2)

I am first suspicious regarding the use of either an Icom 775 or an HP 8640B as a calibrated reference down to 100 Hz. 

I have an 8640B myself and recognize that, despite its excellent design, it DOES need calibration in order to be “dead on”. I am also  not convinced that the 775 is any better than an Orion as a reference.

That said, he’s look at what TenTec says on the manual:

“The first local oscillator (A PLL) operates 9 MHz above the received signal. It tracks the 1 ppm stability of the TCXO so its maximum operating error is 10.8 Hz at 1.8 MHz and 39 Hz at 30 MHz.

The second local oscillator is tuned  to 8.545 MHz, locked to the same reference TCXO. Thus, drift in the TCXO cancels when the two drifts are taken together a the 455 KHz IF (almost). 

In other words, there are TWO oscillators with which one must be concerned. If the TCXO is correct at one frequency, then the first local oscillator must be correct to the same ppm accuracy since that frequency only depends upon the reference frequency and the divider number for the VCO (set by the CPU). It CANNOT be wrong by any other error than the reference oscillator frequency and that would be related to frequency. You should be able to plot the error and get a straight line vs. frequency, though the plot line could pass through zero if the reference is off.

Error in the reference would affect the two local oscillators in opposite directions, but not necessarily by the same amount. Thus, if the reference is off, one might be able to calibrate on one band to be correct by the offsetting errors, the system could be off on other bands due to the fact that the error in the 8.545 MHz oscillator is always the same but the error in the first local oscillator is frequency dependent.

It would seem sensible to me to set the reference oscillator directly to its proper frequency with a calibrated counter reading the actual oscillator output (as I suspect Ten Tec did originally). Then the system would be forced by the digital divider to be correct at all frequencies, to 1 ppm (30 Hz at 30 MHz).

Gary




On Nov 12, 2015, at 10:21 PM, Ron Castro <ronc at sonic.net> wrote:
> 
> Jim:
> 
> I can tell you for certain that it's not in the Orion product line 
> since mine generally stays within +- 5 Hz from the broadcast band to 10 meters.
> In the old days when there were two separate oscillators that 
> determined the frequency, tracking errors you described can occur, but 
> unless my memory is getting foggy, the entire frequency routine of the 
> Orion is determined from a single oscillator module.  In that case, if 
> you're off 5 Hz on 5 MHz WWV, you would be off 10 Hz at 10 MHz, 15 Hz 
> at 15.  On ten meters, you would be
> 30 Hz off, all in the same direction, either plus or minus.  IOW a 
> linear deviation.
> 
> If you calibrate the radio exactly on at 15 MHz WWV, how far off is 10 
> MHz and 5 MHz WWV?  Also check 2.5 MHz WWV if you can hear it, and in 
> the day if propagation is good, check 20 MHz WWV.  If you continue to 
> get the same result, there may be something else wrong that the factory needs to look at.
> 
> 
>          Ron Castro
>           N6IE
>    www.N6IE.com
> 
> Member:
>                  ARRL
>     Redwood Empire DX Assn.
> Northern California Contest Club
> Northern California DX Foundation
>  Society of Broadcast Engineers
>