(no subject)

Noel,
 
 
Good thought actually.... I'll look
 
The 13 100ms is as you state, a delay of 1.3 seconds. That delay is used
for the IR source to come up to a stable condition before doing an A/D
conversion on the output of the thermopile.
 
I've looked at the IR energy out of the thermopile for both modes,
compiled and interpreted, and they are identical with the exception that
the IR-ON, or the IR source, is on slightly longer when interpreting,
which is expected...maybe 50 mS.
 
My only other thinking is that during the interpreting of the typed line
The XTL is communicating with the target serially and potentially
causing a rise in the noise input to the A/D converter. Highly unlikely
as this converter, as do all delta-sigma converters, use a SINC5 filter
which has 120 dB rejection at 60 HZ and multiples there of. You can
pretty noise before you will see any degradation due to noise. Another
reason I don't believe this is the route cause is that on the same board
I'm switching nearly 2 amperes without any noise artifacts showing in
the IR data. We will test this for sanities sake anyway.
 
This is an interesting one. When the super aspect of FORTH,
keyboard-interpret...text-file compile breaks down things get
challenging.
 
Thanks for your reply.
 
Its generally better talking to somebody else after you've talked to
yourself for days...just to make sure your still on planet earth.
 
Gene Silvernail
 
 
-----Original Message-----
From: Noel Henson [mailto:noel_at_noels-lab.com]
Sent: Friday, September 20, 2002 6:47 AM
To: Gene Silvernail
Subject: Re: [swiftx] What the hey !.... ATMEGA103
 
Gene,
Hope you don't mind my butting-in. I deal IR as well but in the realm of

remote controls and such.
I'm assuming that "13 100ms" forms 1.3 second delay. If so, is that time

critical? If so, could there be a difference in BASE from the compiled
version to the interactive version?
Just a thought.
Noel
 
Gene Silvernail wrote:
 
>
>
>What is going on....
>
>The following data comes from an AD, (24 bit) that is used to read IR.
>I get different results depending upon whether the compiled word WWW is

>Executed verses the words that are making up WWW. All this is done from

>the XTL console. Why should running the compiled word WWW make any
>difference versus the running the words that are used to compile WWW.
>The following are copies from XTL and my editor.
>
>Something gotta be nailing the body of the compiled code.TIB ??
>
>Any suggestions on how to trouble shoot this would be appreciated.
>
>
>
>
>
>
>: PICK-CHANNEL ( n-- ) \ Enter with
>CH1...CH8, or NO-CHAN
> SPI-ON
> LTC2408-SPI-MODE
> 2408-DESELECT \ Must
leave
>CS on LTC2408 must be HI
> SPDR C!
> ;
>
>
>{
>=======================================================================
>=
>=
>
> GET-DATA stores the 4 bytes from the LTC2408, storing at
>
> 2408-DATA bits 31-24 offset 1
> 2408-DATA+1 bits 23-16 offset 0
> 2408-DATA+2 bits 15-8 offset 3
> 2408-DATA+3 bits 7-0 offset 2
>
> This is correct for 2@, 2!
>=======================================================================
>=
>= }
>
>
>
>: GET-DATA ( n -- )
> PICK-CHANNEL
> BEGIN AD-DONE? UNTIL
> 0 SPDR C! ?SPI-DONE SPDR C@ 1 2408-DATA + C!
> 0 SPDR C! ?SPI-DONE SPDR C@ 2408-DATA C!
> 0 SPDR C! ?SPI-DONE SPDR C@ 3 2408-DATA + C!
> 0 SPDR C! ?SPI-DONE SPDR C@ 2 2408-DATA + C!
> 2408-DATA 2@ D2/ D2/ D2/ D2/
> 2408-DATA 2!
> 0 2408-DATA 1+ C!
> SPI-OFF
> ;
>
>
>: IR-ON ( -- )
> IR-CONTROL PORTD SET-BIT
> ;
>
>
>: IR-OFF ( -- )
> IR-CONTROL PORTD CLEAR-BIT
> ;
>
>
>: WWW ( -- )
>
> IR-ON 13 100MS CH1 GET-DATA 2408-DATA 2@ IR-OFF D. ;
>
>Compiled WWW code
>
>
>3106 IR-ON RCALL 3AD8
>3108 (LITERAL) CALL 0E943000
>310C LITERAL D
>310E 100MS RCALL BBD8
>3110 (LITERAL) CALL 0E943000
>3114 LITERAL 8
>3116 GET-DATA RCALL F7DE
>3118 (LITERAL) CALL 0E943000
>311C LITERAL 8FF
>311E 2@ CALL 0E944F02
>3122 IR-OFF RCALL 33D8
>3124 D. JMP 0C947506 ok
>
>
>
>
>
>HERE IS A TYPICAL SESSION....
>
>
>
>
>READCOLD 5372 ok
>READCOLD 5788 ok
>READCOLD 5954 ok
>READCOLD 5718 ok
>
>
>Now we turn the IR source on..climbing values are to expected
>
>
>WWW 350322 ok
>WWW 359556 ok
>WWW 367106 ok
>WWW 372434 ok
>
>
>Here are the words direct..big difference.. eh
>
>IR-ON 13 100MS CH1 GET-DATA 2408-DATA 2@ IR-OFF D. 485175 ok IR-ON 13
>100MS CH1 GET-DATA 2408-DATA 2@ IR-OFF D. 493667 ok IR-ON 13 100MS CH1

>GET-DATA 2408-DATA 2@ IR-OFF D. 503249 ok IR-ON 13 100MS CH1 GET-DATA
>2408-DATA 2@ IR-OFF D. 509294 ok
>
>
>Repeating WWW
>
>WWW 347811 ok
>WWW 353093 ok
>WWW 361707 ok
>WWW 368330 ok
>
>
>Repeating words direct..
>
>
>IR-ON 13 100MS CH1 GET-DATA 2408-DATA 2@ IR-OFF D. 487014 ok IR-ON 13
>100MS CH1 GET-DATA 2408-DATA 2@ IR-OFF D. 497636 ok IR-ON 13 100MS CH1

>GET-DATA 2408-DATA 2@ IR-OFF D. 506249 ok IR-ON 13 100MS CH1 GET-DATA
>2408-DATA 2@ IR-OFF D. 513419 ok
>
>Redoing WWW to this was no help..forced an output value
>
>Also forced an output value from within GET-DATA..worked fine
>
>: WWW
>
> IR-ON 13 100MS CH1 GET-DATA
> { 2408-DATA 2@ }
> 1234567.
> IR-OFF D.
> ;
>
>
>IR-ON 13 100MS CH1 GET-DATA 1234567. IR-OFF D. 1234567 ok IR-ON 13
>100MS CH1 GET-DATA 1234567. IR-OFF D. 1234567 ok WWW 1234567 ok
>WWW 1234567 ok
>WWW 1234567 ok
>IR-ON 13 100MS CH1 GET-DATA 1234567. IR-OFF D. 1234567 ok
>IR-ON 13 100MS CH1 GET-DATA 1234567. IR-OFF D. 1234567 ok
>
>
>----------------------------------------------------------------------
>swiftx_at_forth.com The SwiftX programming discussion email list
>To unsubscribe, send subject "unsubscribe swiftx" to listar_at_forth.com
>For help with listar commands, send subject "help" to listar_at_forth.com
>Archives are located at http://www.forth.com/swiftx -- check them out!
>----------------------------------------------------------------------
>THIS LIST IS NOT FOR BUG REPORTS! Send bug reports to
>support_at_forth.com.
>
>
>
 
 
 
 

----------------------------------------------------------------------
swiftx_at_forth.com The SwiftX programming discussion email list
To unsubscribe, send subject "unsubscribe swiftx" to listar_at_forth.com
For help with listar commands, send subject "help" to listar_at_forth.com
Archives are located at http://www.forth.com/swiftx -- check them out!
----------------------------------------------------------------------
THIS LIST IS NOT FOR BUG REPORTS! Send bug reports to support_at_forth.com.
Received on Fri Sep 20 2002 - 12:28:50 PDT