Re: LF: Re: WOLF/WSPR Sound card issues

[Andy Talbot <andy.g4jnt@googlemail.com>]

My Dell Inspiron Laptop, when  11025Hz sampling rate is requested, actually gives 10800Hz (near enough) which was enough to completely kill JT4 decoding.  (JT4 is a WSJT mode, not totally unlike WSPR in its coding/error correction and use of four frequency FSK with 50% embedded sync, but with a much wider frequency shift and 48 second Tx period with 4.375Hz symbol interval).

 

The WSJT software itself measures and reports sampling rate errors and allows the user to enter the values into the setup - quite why Joe never made this an automatic procedure I wonder....

 

the 'JNT theory why it was so far out...

The old sampling rate of 11025 has factors of (3 * 3 * 5 * 5 * 7 * 7)

However 12000Hz has factors   5 * 5 * 5 * 3 * 2 * 2 * 2 * 2 * 2 * 2

So I'm guessing 10800 (factors of  5 * 5 * 3 * 3 * 3 * 2 * 2 * 2 * 2)  is adopted as it has more in common with 12000 and keeps the firmware simpler.

 

Frequency error of 2.08% which I'm sure a musician would detect

 

As WSPR is the first from the WSJT stable to use 12000 sampling, presumably he saw no need to put the correction into the original WSPR software.  

 

Andy
www.g4jnt.com

(working on modifying WSPR beacon software for the ancient AD9851 DDS, as I have a couple spare now)

this email has been scanned for political correctness by the nanny state

2009/9/12 Graham <g8fzk@g8fzk.fsnet.co.uk>

Jim,

Re the  size of error

With the  delta-4 card, the  tx frame could be observed to  be  some  2 or more seconds short of the  expected time  window

1 % ~  1.2 seconds , 2 %  ~ 2.4  seconds, assumedly indicating the clock on the  card runs fast, but as to the  influence of the system by the software, dma, irq etc  Im not  sure, one problem is I expect is the soundcards are not  designed to be the  interface for  a accurate  frequency meter / generator and as such compromises 'must' exist ...

The only  comparison I can think of is the running of a movie at  shot at 24 fps  for projection , on a tele cinie system at  25 fps to  sync with the uk 50 hz frame rate  , giving a lift in the sound  pitch of + ~ 4 % ? , I know that this dose generate  irate phone  calls from musicians pointing out that orchestras appear to  be   'sharp' but not many  hi .. so its possible a tolerance of  2 %  may be accepted

G .

--------------------------------------------------
From: "James Moritz" <james.moritz@btopenworld.com>
Sent: Saturday, September 12, 2009 1:33 PM

To: <rsgb_lf_group@blacksheep.org>

Subject: LF: Re: WOLF/WSPR Sound card issues

Dear Dave,

AC 97
(http://download.intel.com/support/motherboards/desktop/sb/ac97_r23.pdf )
specifies 48kHz as the basic, obligatory sample rate for compliant sound
cards, so yours should certainly be able to do it - to find more info, you
could try looking for the audio codec chip on the motherboard - usually a
very tiny square 48pin thing near the audio connectors, and searching using
its type number. But I guess what sample rate is actually used will depend
on software, drivers and so forth - perhaps someone with experience of PC
sound card programming could advise. Looking at the audio settings in the
Windows control panel may well tell you what sample rates are possible.
Also, some software allows you to select different sample rates, for
instance Spectrum Lab, but not Argo as far as I know.

The errors you are getting in the measured frequency seem to be enormous - 2
or 3%, which would certainly give rise to problems.

Cheers, Jim Moritz
73 de M0BMU

----- Original Message ----- From: "Dave Sergeant" <dave@davesergeant.com>
To: <rsgb_lf_group@blacksheep.org>
Sent: Saturday, September 12, 2009 11:59 AM
Subject: LF: WOLF/WSPR Sound card issues


It seems that both WOLF and WSPR are

designed for current sound cards which have 48,000 and 96,000 clock
rates. Mine is obviously from the old stable so has 44,100 etc rates.

For information it is the sound card interface built into the
motherboard, an ASR Rock K7S41GX dating from 2003 with generic AC'97
sound. Zilch more given about its spec in the manual.

No virus found in this incoming message.
Checked by AVG - www.avg.com
Version: 8.5.409 / Virus Database: 270.13.93/2365 - Release Date: 09/12/09 06:37:00