ok first a few things about your database
1. Don't use special characters (# , ; * !, spaces, etc.) in your field or object names, the only exception being an underscore (_). They will cause no end of problems as you progress with your database
2. If there is a 1 to 1 relationship between tblTestInfo and tblTestResults you don't need to split these tables at all if there will always be a result for a given test.
3. Your supplier, filter media, media type and filter size should all be options from supporting tables. If these are free-form type in fields a media type of 'pollen' would be different from 'polleen' or 'polen' if the field was mistyped.
for example:
Code:
tblSupplier
S_ID S_Name ----> other supplier specific fields
1 Supplier1
2 Supplier2
tblFilterMedia
M_ID M_Name ---> other media specific fields
1 Filter Media 1
2 Filter Media 2
3 Filter Media 3
tblMediaType ---> other media type specific fields
MT_ID MT_Name
1 Pollen
2 Carbon
tblFilterSize
FS_ID FS_Name ----> other filter size specific fields
1 Filter Size 1
2 Filter Size 2
3 Filter Size 3
This also would allow you to set up criteria for your graph very, very easily rather than relying on free-form fields.
Lastly, you are storing one of the two coordinates as a field name which is really not a good practice and is probably causing your issues with graphing. If your test type sets the air pressures and they are standard for a particular test those should be stored in a separate table and the results stored in your actual test table
i.e.
Code:
tblTestType
TT_ID TT_Name ---> other standard test specific fields
1 Standard Test 1
2 Standard Test 2
tblTestTypePressures
TTP_ID TT_ID TTP_Pressure
1 1 100
2 1 168.78
3 1 253.17
4 1 337.56
5 1 421.95
6 1 506.33
Then your testing results would be split into two tables
Code:
tblTest
T_ID TestName TT_ID S_ID M_ID MT_ID FS_ID
1 FT0001 1 1 1 1 1
2 FT0002 1 1 1 2 2
3 FT0003 1 2 2 1 1
4 FT0004 1 2 3 2 1
tblTestResults
TR_ID T_ID TTP_ID TR_Measurement
1 1 1 8
2 1 2 18
3 1 3 32
4 1 4 47
5 1 5 64
6 1 6 83
7 2 1 8
8 2 2 18
9 2 3 31
10 2 4 46
11 2 5 62
12 2 6 80
etc...
As it stands with your dataset you would have to artificially create the x axis value to go along with your y axis recorded value. It is possible with your current setup but if this is a 'new' project you might want to consider creating a more valid data structure from the start and make it as flexible as possible.