Re: [Hamlib-developer] Yaesu S-Meter Level Calculation

[Berndt J. W. <wu...@pi...>]

On Wed, 22 Oct 2003 01:18 pm, Nate Bargmann wrote:
> Hi Berndt.

Hi Nate,

[...]
> > 2.) The template for the calcuation of the value as currently applied in
> > hamlib is hard coded with sample data obtained from one particular rig,
> > which may not be universally applicable to other rigs. The true value
> > greately depends on rig calibration, component tolerance etc.
>
> Certainly.  When I wrote the code in question earlier this year I hoped
> to get feedback from '890 users and I could fine tune the algorithm with
> a larger sample.  So far it's been very quiet.
>

Just a quick reply to your comment above to provide you with the data I've 
obtained from an FT-990 rig -

----------------------------------
S-value	| Data	| Delta
----------------------------------
1	| 7	| 
2	| 23	| 16
3	| 40	| 17
4	| 56	| 16
5	| 72	| 16
6	| 86	| 14
7	| 99	| 13
8	| 112	| 13
9	| 124	| 12
+10	| 146	| 22
+20	| 167	| 22
+30	| 189	| 21
+40	| 210	| 21
+50	| 232	| 22
+60	| 255	| 23
----------------------------------

Applying above data to a polynominal with twiddle factors obtained from a 
least square fit calculation should give reasonable accuracy for this 
particular rig and model. However, it will be interesting to see how closely 
matched these values are when applied over a larger set of samples for this 
model.

73, Berndt
VK5ABN