Rev 0         7/10/98 Initial release

Rev 1         9/10/98 Major revision, transferred to Word document

Rev 1.1     9/22/98 Page numbers, table of contents, section 4.5 and table 4.5-2

Rev 1.2     10/14/98 Figure 2.1-1 and description updated.

                Note on B-weight added to top of p32.

Rev 1.3     10/30/98 Reformat for HTML, port to website
 

Rev 1.4      02/03/99        Fix unreadable figures

Table of Contents *

1.0 PURPOSE *

2.0 OVERVIEW *

The purpose of this document is to provide the detailed information needed to read the ACE/ULEIS level 1.5 data files (UDF) produced at UMd. For information on commanding and S/C data formats, see "ULEIS: Flight Software System Users Guide", JHU/APL publication SRS-084-97.

2.0 OVERVIEW

2.1 PROCESSING OVERVIEW

Processing of UMd ACE/ULEIS data is sketched in Fig. 2.1-1. Data is broadcast from the ACE satellite, arriving eventually at GSFC where raw data is converted to Level_0 data and is passed on to Caltech.

Figure 2.1-1 UMd ACE/ULEIS Data Processing Overview

At Caltech data is taken from packetized format and put into HDF files on 24-hour boundaries sorted by instrument and put in time order. This is Level_1 data and is distributed to investigators over FTP and via CD-ROMs.

As HDF is not a useful format for retrieving ULEIS data, the Level_1 data is immediately resorted into science data records and stored as UDFs. Time and position data is taken from the ancillary HDF data files at Caltech and put in UDF headers. The bare UDFs are merged with browse data from a separate file and the combined file is archived for later use. Occupying a processing level between the Level_1 data from Caltech and the Level_2 data products, the UDFs can be described as Level _1.5 data.

2.2 SCIENCE DATA RECORD

A science data record (SDR) consists of the collected ULEIS data for a period of ten consecutive spins. (A spin period is approximately 12 seconds). The SDR begins and ends on a S/C sun pulse but its readout occurs over 8 S/C telemetry major frames, beginning and ending on a 8xMajFrPulse. The distribution of the data in S/C telemetry frames is shown in Fig. 2.2-1. In this figure each line represents one major frame and the frames are synchronized to the 8xMajFr pulse. Science records appear one every 8 major S/C frames with no gaps, but they do not represent 100% of the data collected by the sensor: there is typically a dropped spin between SDRs to accommodate the different periods of readout and S/C spin. This timing is shown in more detail in Fig. 2.2-2.

Figure 2.2-1 ULEIS SDR Format in ACE S/C Telemetry Frame Structure
 
  Figure 2.2-2 Uleis Data Timing
 

2.3 LEVEL-1 DATA FORMAT

Time on the S/C is counted in Minor Frames with 1 minor frame = 1 second, with some small, variable error. S/C time is the number of minor frames since the S/C data system was turned on just after launch. S/C time is an integer and by itself allows calculation of time at best to the nearest second.

The ACE Epoch is defined as Midnight of Jan 1 1996. Approximate ACEepoch time can be obtained by adding 52,069,705.0 to the S/C time. Due to accumulated errors and drifts in the S/C clock oscillator, this approximate time will be off GMT by several seconds. Correct ACEepoch time is obtained by multiplying the S/C clock by a daily correction factor and adding a daily offset before adding in the large offset as above. The daily factor and offset are incorporated in the ancillary data file ACE_ANCIL.HDF available via FTP or CD-ROM from Caltech. The work of performing the conversion of S/C time to ACEepoch time is performed by a C routine "SCclock_to_ACEepoch()" which is supplied by Caltech as part of the file "ancil_subs.c"

All ULEIS data is tagged with the "Output Time" which is the S/C time of the 8xMajFrPulse at the beginning of the readout of the science record containing the ULEIS data. Actual collection of the data of course occurred at a somewhat earlier time. The ULEIS DPU begins a Science Data Record on a Sun Pulse and records the minor frame count associated with this sun pulse in the trailer data as "Spin1MinFrCnt". We can calculate the S/C time of the beginning of the SDR in the DPU by taking the Output time, subtracting 128 to get to the start of the previous 8xMajFr cycle and adding in Spin1MinFrCnt to get to the sun pulse. Caltech does this for us and calls it the "Collect Time". Unfortunately, Caltech mis-calculated this time in early versions of the data (before v2-5) so we now perform the computation ourselves and use the result in the calculation of ACEepoch time. It is this latter time, supplied at the beginning of each SDR, which should be used in time calculations for the data. When the time_fix_flag in the SDR Header is 0, the ACEepoch time is good to the nearest second. Times for individual rates or events can be determined by adding their spin and sector times to the ACEepoch: 12 seconds/spin and 1.5 seconds/sector.

3.0 UDF STRUCTURE

3.1 PROCESSING

As detailed above, Level_1 data is organized by time: one L1 file per day starting at or near 00:00:00 and continuing through 23:59:59. UDFs are similarly organized, one UDF being produced from each L1 data file.

UDF files are named for the date of the included data and the version number of the software used to produce it:

ULyyyy_ddd.Pxx

ULyyyy_ddd.Rxx

Where yyyy =4-digit year

ddd = 3-digit day of year

xx = 2-digit version number

.Pxx files include all data

.Rxx files omit the PHA event data

The "version number" is the major version number of the program PROCESS_L1 that is used to generate the UDFs from the L1 data files.

A diagram of the UDF processing is shown in Fig. 3.1-1.

F
Figure 3.1-1

3.2 STRUCTURE

UDFs are unformatted FORTRAN binary files containing a day’s ULEIS data organized as a sequence of Science Data Records. The organization of a UDF is shown in Table 3.2-1.

The file begins with a File Header documenting the elements used in producing the file. Next come a series of Science Data Records and finally the end-of-file. In a typical complete day of data there will be 675 science records (86400 sec/day and 128 sec/SDR).

Table 3.2-1 UDF File Structure
 

Item #	Item
1	File Header
2	Science Data Record #1
3	Science Data Record #2
4	Science Data Record #3
…	…
N+1	Science Data Record #N
N+2	End of File

 

A more detailed picture of the UDF structure is given in Tables 3.2-2 and 3.2-3.

Table 3.2-2 shows the header with its 2 constituent records: a 1-byte Record ID (value = 99 for this data type) and a 16-byte File Header. Details of the file header structure are given below in section 4.

Table 3.2-2 File Header
 

Rec #	Contents	Description	Length (bytes)	Comment
1	"99"	File Header Record ID	1
2	Data	File Header	16

 

Table 3.2-3 shows a complete science record broken into 14 data types each preceded by its own Record ID. Record Ids are summarized below in Table 3.2-4.

Table 3.2-3 Science Data Record Structure
 

Rec. #	Contents	Description	Length (bytes)	Comment
1	"1"	Header Rec ID	1
2	data	Header	54
3	"8"	Mag Browse Rec ID	1	Present only of B_Weight>1
4	data	5-min Mag Browse	18
5	"9"	SEPICA Browse ID	1	Present only if SEP LT>0
6	data	5-min SEP Brwse	40
7	"10"	EPAM Browse Rec ID	1	Present only if EPAM LT >0
8	data	5-min EPAM Brwse	36
9	"11"	ULEIS Brwse Rec ID	1	Present only if ULEIS LT>0
10	data	5-min UL Brwse	44
11	"12"	SWEPAM Brwse Rec ID	1	Present only if SWE wgt >1
12	data	5-min SWE Brwse	24
13	"13"	CRIS Browse Rec ID	1
14	data	1-hr CRIS Brwse	56
15	"14"	SIS Browse Rec ID	1
16	data	1-hr SIS Brwse	20
17	"2"	PHA Events Rec ID	1	Used only if pha events present
18	npha	#of PHA records	2	0<npha
19	data	Block of packed pha events	22n	Npha records at 1 pha event per record
19+n	"3"	1-spin MRate Rec ID	1
20+n	data	Block of single-spin matrix rate data	36*80	80 records of 34 compressed matrix rates
100+n	"4"	2-spin MRate Rec ID	1
101+n	data	Block of double-spin matrix rate data	44*40	40 records of 42 compressed matrix rates
141+n	"5"	Disc Rate Rec ID	1
142+n	data	Block of discriminator rates data	34*40	40 records of 16 compressed discriminator rates
182+n	"6"	Status Rec ID	1
183+n	data	Status Block	112
184+n	data	Status Trailer	128
185+n	"7"	HSKP Rec ID	1
186+n	data	HSKP data	682
187+n	"-1"	End of Sci Record	1

 

Note: n = number of PHA events in the science record = npha

Table 3.2-4 Summary of Record IDs
 

Rec ID Value	Data Type
1	SDR Header
2	PHA Events
3	Matrix Rates Accumulated over 1 Spin
4	Matrix Rates Accumulated over 2 Spins
5	Discriminator Rates
6	Status
7	HSKP
8	Magnetometer 5-Min Ave Browse
9	SEPICA 5-Min Average Browse
10	EPAM 5-Min Average Browse
11	ULEIS 5-Min Average Browse
12	SWEPAM 5-Min Average Browse
13	CRIS 1-hour Average Browse
14	SIS 1-hour Average Browse
99	File Header
-1	End of Science Record Marker

 

4.0 RECORD STRUCTURE – LEVEL-1 DATA

This section gives details of the structures of the data types listed above with record Ids from 1 through 7. This covers all the items included in the LEVEL_1 data files from Caltech. Browse data is considered below in section 5.

4.1 FILE HEADER

The detailed structure of the 16-byte File Header is shown in Table 4.1-1. The file header is read out as a single record.

Table 4.1-1 File Header Record Format ( Rec. ID = 99)
 

Item	Bits	Name	Comment	Type
1	8	PROCESS_L1, Major Rev #		byte
2	8	PROCESS_L1, Minor Rev #		byte
3	8	C modules, Major Rev #	Caltech-supplied code	byte
4	8	C modules, Minor Rev #		byte
5	8	Data: Major Rev #	Encoded in Lev 1 Data file names	byte
6	8	Data: Minor Rev #		byte
7	8	Spare		byte
8	8	Spare		byte
9	8	Spare		byte
10	8	Spare		byte
11	8	Spare		byte
12	8	Spare		byte
13	8	Spare		byte
14	8	Spare		byte
15	8	Spare		byte
16	8	Spare		byte

 

4.2 SCIENCE DATA RECORD HEADER

The detailed structure of the 54-byte Science Data Record Header is shown in Table 4.2-1. The header is read out as a single record.

Referring to Table 4.2-1, the ACEepoch, Collect time and Output time are discussed above in section II.

Attitude, Position and Velocity vectors are passed on from the ancillary data. No checking or processing is performed on them at this time. A quality flag is available for the attitude data but is ignored by the current software and is not in the header. For future reference, the flag has the following properties:

 
 

Flag = -1	clock out of bounds	Attitude data wrong
Flag = 0	OK	Attitude data correct
Flag= +1	maneuver in progress	Attitude data suspect

 

The QAC is a bit appearing in each minor frame of data which is set if there are any known problems with the data in that minor frame (e.g. data missing, data out of sync, etc.). QAC_Count is the number of such bits set in a given Science Record. The number may be as high as 128. The preferred value is 0.

The ULEIS DPU calculates a checksum for the entire data record and includes it in the status trailer at the end of the SDR. Caltech generates a similar checksum on the received data and compares it with the DPU’s value. The result is a check sum flag (Chk_sum_flag), equal to zero if the transmitted and received checksums match and to one if they differ.

As the ACEepoch time calculation depends on several data items being received correctly, there are a number of ways it can go wrong. As long as there are no indications of bad data, various data are consistent with each other and the resulting ACEepoch is consistent with the ones that came before, the Time_fix_flag is set to zero. Here the time data is assumed to be good and the ACEepoch calculation is performed straightforwardly. The resulting ACEepoch is good to the nearest second. If however, the QAC or CHK_SUM flag is non zero, or the output time varies too much from its predecessor, or the Spin1_min_fr_cnt is too large, the time data is assumed to be bad and an attempt is made to correct it. This is done primarily by bringing the bad or missing data into alignment with its preceding values. For these cases the Time_fix_flag is set to a value greater than zero indicating that the ACEepoch time may be off from a couple of seconds to as much as a spin.

Table 4.2-1 SDR Header Record Format ( Rec. ID = 1)
 

Item	Bits	Name	Comment	Type
1	32	ACE_epoch	collect time in seconds since Jan1 96	int*4
2	32	Attitude: R component	s/c attitude in RTN coordinates	real*4
3	32	Attitude: T component		real*4
4	32	Attitude: N component		real*4
5	32	Position: X component	s/c position, km, GSE coordinates	real*4
6	32	Position: Y component		real*4
7	32	Position: Z component		real*4
8	32	Velocity: X component	s/c velocity, km/s, in GSE coordinates	real*4
9	32	Velocity: Y component		real*4
10	32	Velocity: Z component		real*4
11	32	collect time	SC time (minor frames since launch)	int*4
12	32	output time	SC time (minor frames since launch)	int*4
13	32	QAC_Count	# of minor frames with QAC set in SDR	int*4
14	8	chk_sum_flag	0=chk sums matched, 1 = chk sum error	byte
15	8	time_fix_flag	0=time ok, 1=time problem had to be fixed	byte

 

4.3 PULSE HEIGHT EVENT DATA

As is shown in Table 3.2-3 above, the PHA data for a given SDR are arranged as an event-count record, "NPHA", followed by the events themselves, one event per record.

The detailed structures of the 2-byte NPHA record and the 22-byte PHA Data records are given in Table 4.3-1. Each packed event contains 11 integer*2 words whose contents are arranged as in Table 4.3-2. In this table, S1 is the START1 wedge located at the front of the telescope. The Wedge, Strip and Zigzag are the 12-bit position signals from this wedge. Similarly, S2 is the START2 or middle wedge. STOP is the Stop wedge at the rear of the telescope. The SSD E is the energy signal associated with this event, selected from among the signals from the seven solid-state detectors. TOF1 is the time of flight between START 1 and STOP, and TOF2 is the time of flight from START2 and STOP. Status words 1 and 2 contain the flags associated with the event. They are defined in Tables 4.3-3 and 4.3-4. The spin and the sector indicate when the event was processed inside the DPU (a few milliseconds after detection in the telescope). The spin is the spin number 0-9 within the science data record. The sector is 0-15 as opposed to sectors 0-7 for rates.

RATE Sector = INT(PHA Sector / 2)

In Table 4.3-2, the L and H are used to denote the low-order and high order portions of a 12-bit word when that word has been broken up. In each entity, the less-significant bits are at the right end.

Record ID 2 and the associated PHA data appears only in .Pxx UDFs and then only in those SDRs for which NPHA > 0.

Table 4.3-1 PHA Record Structures (Rec ID = 2)

NPHA Record
 

Item #	Bits	Name	Comment	Type
1	16	NPHA	# of PHA events in SDR	Int*2

 

Event Record
 

Item #	Bits	Name	Comment	Type
1	16	Packed PHA word 1	See table 4.3-2	int*2
2	16	Packed PHA word 2		int*2
3	16	Packed PHA word 3		int*2
4	16	Packed PHA word 4		int*2
5	16	Packed PHA word 5		int*2
6	16	Packed PHA word 6		int*2
7	16	Packed PHA word 7		int*2
8	16	Packed PHA word 8		int*2
9	16	Packed PHA word 9		int*2
10	16	Packed PHA word 10		int*2
11	16	Packed PHA word 11		int*2

 

Table 4.3-2 PHA Packing Scheme
 

Name

15

14

13

12

11

10

9

8

7

6

5

4

3

2

1

0

Word 1

S1 Strip (L)

S1 Wedge

Word 2

S1 Zigzag (L)

S1 Strip (H)

Word 3

S2 Wedge

S1 Zig (H)

Word 4

S2 Zig (L)

S2 Strip

Word 5

STOP Wedge (L)

S2 Zig (H)

Word 6

STOP Strip

STOP Wedge (H)

Word 7

SSD E (L)

STOP Zigzag

Word 8

TOF1 (L)

SSD E (H)

Word 9

TOF 2

TOF1 (H)

Word 10

Status2 (L)

Status 1

Word 11

Spin

Sector

Status 2 (H)

 

Table 4.3-3 PHA Status 1 Definition
 

Bit #	Normal Mode	Cal Mode
11 (msb)	HAZ	CE11
10	LA1	CE10
9	LA0	CE9
8	SA1	CE8
7	SA0	CE7
6	D7	CE6
5	D6	CE5
4	D5	CE4
3	D4	CE3
2	D3	CE2
1	D2	CE1
0 (lsb)	D1	CE0

 

Table 4.3-4 PHA Status 2 Definition
 

Bit #	Normal Mode	Cal Mode
11 (msb)	0	SSD ID 2
10	0	SSD ID 1
9	Box Number 5	SSD ID 0
8	Box Number 4	ES
7	Box Number 3	CAL STEP 2
6	Box Number 2	CAL STEP 1
5	Box Number 1	CAL STEP 0
4	Box Number 0	CM
3	CO	CO
2	ES	0
1	TOF2	TOF2
0 (lsb)	TOF1	TOF1

 

In the above status tables:

HAZ = Hazard flag generated in the AE box indicating that two PHA events have occurred within the resolving time of the energy electronics and that therefore the energy signal may be corrupted by the previous event. Events with the HAZ bit set should be used with care.

LA = The two-bit encoding of which large SSD was selected for this event:
 

LA = 0	D5 or none
LA = 1	D6
LA = 2	D7
LA = 3	Not defined

 

SA = The two-bit encoding of which small SSD was selected for this event:
 

SA = 0	D1 or none
SA = 1	D2
SA = 2	D3
SA = 3	D4

 

D1-D7 = Individual discriminator bits showing which SSDs had signals above threshold for this event.

Box Number = Shows which matrix rate this event was accumulated into. See Ref 1 for more details.

CO = Cal mode / Normal mode bit, 1 = Cal Mode

ES = Identifies which energy system was used for this event. 0 = large SSDs, 1 = small SSDs

TOFn = Identifies which TOF system fired for this event. 1 = valid stop present

When the instrument is in CALIBRATE mode, status bit definitions shift somewhat. New status bits include:

CE = 12-bit number indicating the calibrator energy step associated with the event

SSD ID = 3-bit identification of which SSD is being calibrated

CAL STEP = 4-bit calibrator tof step which produced this event

CM = Calibrator mode. 1 = short.

(Note that the CO bit always stays in the same place so we can identify which events are calibrate events and which are not.)

4.4 SINGLE SPIN MATRIX RATES

Single-spin matrix rates are a group of 34 sectored rates read out each spin for best time resolution (12 secs). In the science record they appear as a block of 80 records of 34 rates each, each record representing a single sector of a single spin. The structure of the data block is shown in Table 4.4-1 and the detailed structure of a single matrix rate record is shown in Table 4.4-2.

Table 4.4-1 Single Spin Matrix Rate Data Block (Rec ID = 3)
 

Record	Bytes	Name	Comment
1	36	Spin 1, Sect 0 Rates	spin, sect, 34 rates, see below
2	36	Spin 1, Sect 1 Rates	spin, sect, 34 rates, see below
…
8	36	Spin 1, Sect 7 Rates	spin, sect, 34 rates, see below
9	36	Spin 2, Sect 0 Rates	spin, sect, 34 rates, see below
…
80	36	Spin 10, Sect 7 Rates	spin, sect, 34 rates, see below

 

Each line of the above table represents a single record, and each record contains 34 compressed rates and the spin and sector number structured as follows:

Table 4.4-2 Single Spin Matrix Spin Record Structure
 

Item	Bits	Name	Comment	Type
1	8	Spin number	1-10	byte
2	8	Sector Number	0-7	byte
3	8	Small SSD Background	Box # 64	byte
4	8	H S1	65	byte
5	8	H S2	66	byte
6	8	H S3	67	byte
7	8	H S4	68	byte
8	8	H S5	69	byte
9	8	3He S1	70	byte
10	8	3He S2	71	byte
11	8	3He S3	72	byte
12	8	3He S4	73	byte
13	8	3He S5	74	byte
14	8	4He S1	75	byte
15	8	4He S2	76	byte
16	8	4He S3	77	byte
17	8	4He S4	78	byte
18	8	Large SSD Background	0	byte
19	8	3He L1	1	byte
20	8	3He L2	2	byte
21	8	3He L3	3	byte
22	8	3He L4	4	byte
23	8	3He L5	5	byte
24	8	3He L6	6	byte
25	8	4He L1	7	byte
26	8	4He L2	8	byte
27	8	4He L3	9	byte
28	8	4He L4	10	byte
29	8	4He L5	11	byte
30	8	4He L6	12	byte
31	8	4He L7	13	byte
32	8	4He L8	14	byte
33	8	4He L9	15	byte
34	8	4He L10	16	byte
35	8	4He L11	17	byte
36	8	4He L12	18	byte

 

Each of the 34 rates making up a single-spin record was stored in the DPU as a 16-bit number and was compressed from 16 to 8 bits before transmission to the S/C. It can be decompressed as follows:

Compressed rate is of the form: eeeemmmm

If eeee = 0, value = mmmm

Else value = (16+mmmm)* 2^(eeee-1)

The Box # in the above table refers to the DPU code for the matrix rate in question.

4.5 SPIN-PAIR MATRIX RATES

Spin-pair matrix rates are a group of 42 sectored rates accumulated in the DPU and read out every other spin to save bit rate at the expense of some loss in time resolution (24 seconds). In the science record they appear as a block of 40 records of 42 rates each, each record representing a single sector of a single spin pair. The structure of the data block is shown in Table 4.5-1 and the detailed structure of a single matrix rate record is shown in Table 4.5-2.

Like the single-spin matrix rates, the spin-pair rates are compressed. Each of the 42 rates making up a single-spin record was accumulated in the DPU as a 16-bit number and was compressed from 16 to 8 bits before transmission to the S/C. It can be decompressed as follows:

Compressed rate is of the form: eeeemmmm

If eeee = 0, value = mmmm

Else value = (16+mmmm)* 2^(eeee-1)

Table 4.5-1 Spin-Pair Matrix Rate Data Block (Rec ID = 4)
 

Record	Bytes	Name	Comment
1	44	Spin 1\2, Sect 0 Rates	spin, sect, 42 rates, see below
2	44	Spin 1\2, Sect 1 Rates	spin, sect, 42 rates, see below
…
8	44	Spin 1\2, Sect 7 Rates	spin, sect, 42 rates, see below
9	44	Spin 3\4, Sect 0 Rates	spin, sect, 42 rates, see below
…
40	44	Spin 9\10, Sect 7 Rates	spin, sect, 42 rates, see below

 

As with the single-spin rates above, each line of the above table represents a single record, and each record contains 42 compressed rates and the spin and sector number structured as in Tables 4-5-2 below. A DPU table upload on Feb 17-18 1998 added rate O L7 and bumped the following rates ahead one step. The before and after assignments are given in the ...a and ...b versions of Tables 4.5-2

Table 4.5-2a Spin-Pair Matrix Rate Record Structure: Launch – Feb 17 1998
 

Item #	Bits	Name	Comment	Type
1	8	Spin number	1-10	byte
2	8	Sector Number	0-7	byte
3	8	C S1	Box # 79	byte
4	8	C S2	80	byte
5	8	O S1	81	byte
6	8	O S2	82	byte
7	8	Ne-S S1	83	byte
8	8	Ne-S S2	84	byte
9	8	Fe S1	85	byte
10	8	Fe S2	86	byte
11	8	C L1	19	byte
12	8	C L2	20	byte
13	8	C L3	21	byte
14	8	C L4	22	byte
15	8	C L5	23	byte
16	8	C L6	24	byte
17	8	C L7	25	byte
18	8	C L8	26	byte
19	8	O L1	27	byte
20	8	O L2	28	byte
21	8	O L3	29	byte
22	8	O L4	30	byte
23	8	O L5	31	byte
24	8	O L6	32	byte
25	8	Ne-S L1	33	byte
26	8	Ne-S L2	34	byte
27	8	Ne-S L3	35	byte
28	8	Ne-S L4	36	byte
29	8	Ne-S L5	37	byte
30	8	Ne-S L6	38	byte
31	8	Ne-S L7	39	byte
32	8	Fe L1	40	byte
33	8	Fe L2	41	byte
34	8	Fe L3	42	byte
35	8	Fe L4	43	byte
36	8	Fe L5	44	byte
37	8	Fe L6	45	byte
38	8	Fe L7	46	byte
39	8	Fe L8	47	byte
40	8	Fe L9	48	byte
41	8	Unassigned	reads out as 0	byte
42	8	Unassigned	reads out as 0	byte
43	8	Unassigned	reads out as 0	byte
44	8	Unassigned	reads out as 0	byte

Table 4.5-2b Spin-Pair Matrix Rate Record Structure: Feb 18 1998 - present
 

Item #	Bits	Name	Comment	Type
1	8	Spin number	1-10	byte
2	8	Sector Number	0-7	byte
3	8	C S1	Box # 79	byte
4	8	C S2	80	byte
5	8	O S1	81	byte
6	8	O S2	82	byte
7	8	Ne-S S1	83	byte
8	8	Ne-S S2	84	byte
9	8	Fe S1	85	byte
10	8	Fe S2	86	byte
11	8	C L1	19	byte
12	8	C L2	20	byte
13	8	C L3	21	byte
14	8	C L4	22	byte
15	8	C L5	23	byte
16	8	C L6	24	byte
17	8	C L7	25	byte
18	8	C L8	26	byte
19	8	O L1	27	byte
20	8