GSFC/NGIMS-FSW-020




EEPROM
Interface Control Document

for

CONTOUR/NGIMS
Flight Software





Prepared By:

Michael Paulkovich, Lead Software Engineer

for
Hasso Niemann, Paul Mahaffy, Co-Investigators

NASA/GSFC Code 915
Laboratory for Planetary Atmospheres




October 31, 2001 

Vers. 2.8




DOCUMENT CHANGE HISTORY

Version
Date
Change
2.7
Oct 29, 01
First CM version.  Remove auto-start data.
2.8
Oct 31, 01
Mods to AMB-38 through -41;
Add to "notes" section: v3.7.00 Alt Boot statistics



I  Purpose
This document describes the layout and data formats for the AMB (Adaptive Memory Block) area of EEPROM for the NGIMS instrument.  
The AMB area consists of: QP (quadrupole) computation parameters (for instance, Config Tables); Subscan Tables; appropriate calibration data; software adaptation data; and software constants.  These represent constants that may be defined in any of the requirement, design or code phases, that are normally adjusted during testing and calibration phases, and thus should be modifiable and maintained in non-volatile memory (in this case EEPROM) for automatic application upon boot-up.  
II Overview
The FSW (Flight Software) maintains default values for all AMB parameters (which can be uploaded to EEPROM) in PROM (which, of course, can not be changed).  The AMB parameters are shadowed in (that is, loaded into) RAM for use by the FSW during program operation.  
Values in EEPROM can be altered after the final PROM has been inserted into the instrument, so EEPROM AMB values aren't necessarily the same of those placed into PROM.  Note that {Patch} telecommands can alter EEPROM, but they don't automatically alter their shadowed RAM counterparts.  Execution of the {TableUpdate} command results in loading of RAM with the AMB segment of EEPROM (with no checksum test).
The EEPROM undergoes a checksum test upon boot-up.  If the EEPROM checksum tests pass, the EEPROM values for AMB parameters are loaded into RAM for use by the FSW.  Of course, if the EEPROM checksum tests fail, the "as-launched" PROM values are used by the FSW, and any uploaded EEPROM values are ignored.  
1.  Conventions
This section describes the conventions and special nomenclature used to describe the interface data.  
1.1  Word Definition
The word size for the 1750 FC (Flight Computer) is 16 bits.  The bit numbering is such that the MSB is bit 0, the LSB is bit 15.  

1.2  Data Types
Below are the basic data types for the AMB data objects:
Type
Notes
# Bits
Int16
MSB is sign.  In this document, also refers to any unsigned 16-bit numbers
16
Boolean
LSB is the only applicable bit.  1=TRUE. 
16
Float
32-bit Mil-Std-1750A floating point type*
32
Fixed 
Radix point between two indicated bits; bits to right of radix point are fractional. 
16

* Two-word Floating Point Format (see diagram below) - 
 8-bit signed Exponent in LSBs of 2nd word
 23-bit Mantissa in bits 1..23
 1-bit Sign in MSB of 1st word

0
1
15
s
Mantissa MSBs



0
7
8
15
Mantissa LSBs
s
Exponent
The decimal point in the mantissa is between bits 0 and 1.  The value is computed as 2**exponent*mantissa.


1.3  Data Subtypes

There are specific subtypes of the above basic types used in the NGIMS AMB data as shown below.  The Size column indicates the number of words of EEPROM used to store the data type: 

Subtype
Base Type
Range
LSB
Size
Notes
Scale_8
Fixed
-126.0 .. 126.0
1/256
1
Used for Correction Limits
Scale_14
Fixed
-1.999 .. 1.999
1/16384
1
Used for Temperature Coefficients
Dac_Index
Int16
1..29
1
1
1=Vac, 2=Vdc, etc - see DAC memo
Rf_Dac
Int16
1..2
1
1
Subset of Dac_Index: 1=Vac, 2=Vdc
Amu_Code
Int16
0..4020
1
1
0..301 = AMUs, 4000..4099 = Bands
Band_Amu
Int16
4000 .. 4020
1

index 0..20 
Mass
Int16
0..301
1
1
Used for "L" lookup table
Freq
Int16
0..2
1
1
0=XX, 1=XX, 2=XX
Config
Int16
0..3
1
1
Selected Config:
0=Open_Source, 
1=Closed_Source, 
2=Ion_Mode, 
3=Neutral_Thermal
Lconfig
Int16
0..4
1
1
"Logical" Config. Computed based on:
  - "selected" Config
  - Mass Code
  - Ion Mode Cutoff Mass
0=Open_Source, 
1=Closed_Source, 
2=Ion_Mode_Lo_E, 
3=Ion_Mode_Hi_E, 
4=Neutral_Thermal
Omega
Int16
0..3
1
1
0 = 0 degrees
1 =  0.5 degrees
2 = 2.5 degrees
3 = 6.0 degrees
IP_Index
Int16
1..15
1
1
Used for indexing Select Words
SS_Table
Int16
0..255
1
1
Each SS_Table contains 15 Select Words
SS_Rec
Record
2 wds + 15 IPs
n/a
17
Subscan Table layout:
   Adaptive Flag (note 1)
   Source (note 2)
   Subscan (note3)
User_Mux
Int16
1..91
1
1
"User" Mux ID number, 1..91.  See 0770491
AD_Index
Int16
0..255
1
1
Index into table of 256 A/D IDs
	note 1: Set = 1 means this table is a special subscan of all Bands for the Adaptive Scanning Algorithm
	note 2: Source: 0=OS, 1=CS, 2=IM, 3=NT
	note 3: The Subscan consists of 15 Select Words.

1.4  Data Structures

There are two types of AMB data structures - Simple structures, and Arrays.  

Simple structures have a straightforward layout in EEPROM -- each Simple data object consists of one object, and in fact, just one 16-bit word (that is, there are no Simple structures of type Float).  

But arrays may be accessed by one, two, or more indices.  The simplest example is an array of one dimension (one index); the following nomenclature form in this document: 


... represents a one-dimensional array, accessed by an index of type Dac_Index (range 1..29).  Each data item in such an array is one 16-bit word of type Int16 (16-bit integer).  

An array of this type yields the following layout in EEPROM: 
		
Array Index
Address Offset
(16-bit words:)
1
0
X X X X X X X X X X X X X X X X
2
1
X X X X X X X X X X X X X X X X
3
2
X X X X X X X X X X X X X X X X
4
3
X X X X X X X X X X X X X X X X
5
4
X X X X X X X X X X X X X X X X
6
5
X X X X X X X X X X X X X X X X
7
6
X X X X X X X X X X X X X X X X
8
7
X X X X X X X X X X X X X X X X
9
8
X X X X X X X X X X X X X X X X
10
9
X X X X X X X X X X X X X X X X
11
10
X X X X X X X X X X X X X X X X
12
11
X X X X X X X X X X X X X X X X
13
12
X X X X X X X X X X X X X X X X
14
13
X X X X X X X X X X X X X X X X
15
14
X X X X X X X X X X X X X X X X
16
15
X X X X X X X X X X X X X X X X
17
16
X X X X X X X X X X X X X X X X
18
17
X X X X X X X X X X X X X X X X
19
18
X X X X X X X X X X X X X X X X
20
19
X X X X X X X X X X X X X X X X
21
20
X X X X X X X X X X X X X X X X
22
21
X X X X X X X X X X X X X X X X
23
22
X X X X X X X X X X X X X X X X
24
23
X X X X X X X X X X X X X X X X
25
24
X X X X X X X X X X X X X X X X
26
25
X X X X X X X X X X X X X X X X
27
26
X X X X X X X X X X X X X X X X
28
27
X X X X X X X X X X X X X X X X
29
28
X X X X X X X X X X X X X X X X

In this document, an example array of two dimensions is indicated by two indices in parenthesis, e.g. the following form:
	X_Array : array (Freq) (Rf_Dac) of Float

	...would indicate an array with one dimension of Freq and a second dimension of Rf_Dac; and each data item in the array is of type Float.

A "real" example of a multi-dimensional array is the "L" array of GSFC/NGIMS-FSW-015, which would be specified in this document as:	
	Config_L : array (Freq) (Mass) (Rf_Dac) of Int16

 ... which selects data of type Int16, based upon three indices:
	Freq (range 0..2), 
	Mass (range 0..301); and 
	Rf_Dac (range 1..2).

It is depicted pictorially as follows: 	

AMU

0
1
2
3
4
5



...






300
301
LLF
vac, vdc
vac, vdc
vac, vdc
vac, vdc
vac, vdc
vac, vdc



...






vac, vdc
vac, vdc
LMF
vac, vdc
vac, vdc
vac, vdc
vac, vdc
vac, vdc
vac, vdc



...






vac, vdc
vac, vdc
LHF
vac, vdc
vac, vdc
vac, vdc
vac, vdc
vac, vdc
vac, vdc



...






vac, vdc
vac, vdc

The Tartan Ada compiler allocates such an array in memory as shown below.  The order of the indices dictates the grouping of the data blocks, in this example, first by Freq, then Mass, then Rf_Dac.  The size of the data objects in the array, of course, along with the index ranges, dictates the size of each data block. 
	
Rel. Addr
Freq 
Mass
Data
0000
0001
0002
0003
0004
0005
0006
0007
...

LF
0

1

2

...

301

Vac
Vdc
Vac
Vdc
Vac
Vdc
Vac
Vdc
Vac
Vdc
etc

MF
0

1

...

301

Vac
Vdc
Vac
Vdc
Vac
Vdc
Vac
Vdc
etc
HF
etc
etc


2.  AMB Data Formats and Addresses


EEPROM Address (hex)
AMB Data
#Words (dec)
AMB1-5
0000
spare
1
AMB-06
0005
Config Table A coefficients --
Config_A : array (LConfig) (Dac_Index) of Float
2 words each, 5x29
290
AMB-07
127
Config Table B coefficients --
Config_B : array (LConfig) (Dac_Index) of Float
2 words each, 5x29
290
AMB-08
249
Config Table G coefficients --
Config_G : array (LConfig) (Dac_Index) (Omega) of Float
2 words each, 5x29x4
1160
AMB-09
6D1
Config Table H coefficients --
Config_H : array (LConfig) (Dac_Index) of Int16
(note: valid values are range +-3000)
1 word each, 5x29
145
AMB-10
762
Config_K1 : array ((Dac_Index) of Int16
1 word each, 5x29
145
AMB-11
7F3
Config_L : array (Freq) (Mass) (Rf_Dac) of Int16
1 word each, 3x302x2
1812
AMB-12
F07
Config_E : array (Freq) (Dac_Index) of Float
2 words each, 3x29x2
174
AMB-13
FB5
Config_C : array (Int16, Int16, Scale_14, Int16)
4 words each (Lower, Upper, C1, C2), 2x12
96
AMB-14
1015
Config_K2 : array ((Dac_Index) of Int16
1 word each x 29
29
AMB-15
1032
RF_Corr_YN_Array : array (Dac_Index) of Boolean
1 word each x 29
29
AMB-16
104F
Temp_Corr_YN_Array : array (Dac_Index) of Boolean
1 word each x 29
29
AMB-17
106C
RFMon_Corr_Limit : array (Freq) of Int16
3
AMB-18
106F
Temperature_Corr_Limit_RF : not used
1
AMB-19
1070
Temperature_Corr_Limit_Non_RF : not used
1
AMB-20
1071
RFMon_Avg_Sample_Number : Int16
1
AMB-21
1072
Temperature_Avg_Sample_Number (spare, PROM value of 3 is used)
1
AMB-22
1073
RFMon_Nominal_LF : Int16
1
AMB-23
1074
RFMon_Nominal_MF : Int16
1
AMB-24
1075
RFMon_Nominal_HF : Int16
1
AMB-25
1076
Ion_Mode_Mass_Switchover : Amu_Code
1
AMB-26
1077
spare
609
AMB-27
12d8
Subscan_Tables : array (SS_Table) (SS_Rec) of Int16
1 Select Word each (256)x(2+15).  See GSFC/NGIMS-FSW-015 §10.1.
3165
AMB-28
23d8
Mux_Array : array (AD_Index) of User_Mux
256
AMB-29
24d8
DAC Overrides : array (29) of [Flag / Int16] records
58
AMB-30
2512
InitMode_DACs : array (29) of Int16
46
AMB-31
2540
Band_DAC : array (Band_Amu) (Dac_Index) of Int16
1 word each 21x29x3
2436
AMB-32
2EC4
spares
11

AMB-33
2ECD
spares
3
AMB-34
2ED0
Config Table ID block
48
AMB-35
2F00
ETCBoot Version# 
1
AMB-36
2F01
ETCBoot EEPROM Checksum computed byf LOG2CTB program
1
AMB-37
2F02
Patch "FAT" (in "Copytable" format)  <256 required for load image; 5*256 added for up to 256 patches.
256 + 1280 =1536
AMB-38
3502
Load Image (in "Copytable" format) to ~E700
<44000
AMB-39
~E700
Patch data area - start addr depends on NGIMS load image
~2300
AMB-40
F000
spares
1088
AMB-41
F440
Allocated for ATCS - spare
3004
AMB-42
FFFC
AMB Load Flag: set to AB12h to cause FSW to do TableUpdate on boot
1
AMB-43
FFFD
ETCBoot Load Counter: decrements to zero on each reboot/power cycle. Value of zero means don't ETCBoot.
1
AMB-44
FFFE
spare
1
AMB-45
FFFF
Checksum
1



NOTES

EEPROM 0 Spares = ~ 609+11+1280+3840+1088=~ 6828d

Sample output of LOG2CTB program: 

Starting Storage Addresses were:
 for Version:    2f00
 for Checksum:   2f01
 for CTB FAT:    2f02
 for CTB Data:   3502

Total number of zeroblock words:  22425  (5799h)
High EEPROM Data Address:        dd67
High EEPROM FAT Address:         2fc6
Compressed data size (decimal):   43109
Compressed data size (hex):      a865

v3.7.00 Alt Boot statistics Oct 30 2001:

Starting Storage Addresses were:
 for Version:    2f00
 for Checksum:   2f01
 for CTB FAT:    2f02
 for CTB Data:   3502

Total number of zeroblock words:  19912  (4dc8h)
High EEPROM Data Address:        e738
High EEPROM FAT Address:         3016
Compressed data size (decimal):  45622
Compressed data size (hex):      b236

END