forums.silverfrost.com

[colt1954]

Any ideas why this program which attempts to sift certain array values hence the first IF/DO loop section (relevant snapshot below) it prints out sensible expected values within the initial do loops so it works thus far (i put the write command in simply as a check), but when I finally change the order it prints out (which I need to do) in the second set of do loops the values come out stupid e.g 4.034E0129....the FX,FY..etc arrays are declared as real*8, setting them up simply as dimension FX(6,12)..simply does not work etc. The GENG arrays are complex.

DO J=1,6
DO M=2,512
IF(ABS(THETA(2,M)-(RPM*J/120)).LT.0.001)THEN
FX(J,NRPM)=ABS(GENG(31,M))
FY(J,NRPM)=ABS(GENG(32,M))
FZ(J,NRPM)=ABS(GENG(33,M))
MX(J,NRPM)=ABS(GENG(34,M))
MY(J,NRPM)=ABS(GENG(35,M))
MZ(J,NRPM)=ABS(GENG(36,M))
WRITE(*,*)RPM,FX(J,NRPM),FY(J,NRPM),FZ(J,NRPM),MX(J,NRPM),MY (J,NRPM),MZ(J,NRPM)
ENDIF
ENDDO
ENDDO
IF(NRPM.EQ.11)THEN
DO J=1,6
DO N=1,NRPM
ESP=RPMIN+(N-1)*RPMI
EO=J*0.5
WRITE(*,*)ESP,EO,FX(J,N),FY(J,N),FZ(J,N),MX(J,N),MY(J,N),MZ(J,N)
ENDDO
ENDDO
ENDIF

Here is a sample snapshot of output sets for last write section for two rpm values 1500/2000:

1500.00 0.500000 0.00000000000 0.00000000000 0.00000000000
0.00000000000 0.00000000000 0.00000000000
2000.00 0.500000 0.00000000000 -1.420806014199E+0241 0.00000000000
2.336470151337E-0307 2.368279619495E-0308 0.00000000000

[davidb]

I can see a few things. But it is difficult to see what you are trying to do.

You seem to be overwriting the elements of the arrays with each pass through the first M loop. E.g. should you have FX(J,M) = ... and not FX(J,NRPM) = ..' ? Or should NRPM be calculated inside the M loop? or inside the IF block (I think the latter.)

I think your check PRINT simply gives you the last calculation. It doesn't tell you that you are over-writing all the earlier ones.

This is my guess at what you need. I hope it helps some. You might want to consider whether the indenting makes it clearer too, and whether it would have helped you seen the answer yourself.

[colt1954]

This problem was sorted by making the new set of arrays exactly the same size as the original set (in the array declaration, not shown above) even though the second set only needed a fraction of the first array data...not elegant I realise and I'm sure there must be a better way, but hey it worked.

[JohnCampbell]

You appear to be searching for the closest value of THETA. An alternative approach could be:
1) Find the closest value of THETA (see code below) or,
2) Find the 2 closest values of THETA and do a linear interpolation between these two.

The simpler closest value search could be coded as:

[colt1954]

John, really appreciate your efforts...no the NRPM integer is called via a subroutine in stepwise fashion...after NRPM calls to subroutine NRPM=10 say, so the arrays are seeded with 10 sets of values etc thats when the second loop is invoked, so it does work. As I stated my problem was I had to make the array sizes equal which seems a waste of space...but hey with modern computers it does not seem to mind.

Your comments re. the If statement <0.001 well really that is more than accurate for what I'm doing its a simple sieve on quite distinct vibration order values so it good enough !!

[JohnCampbell]

For your info, I thought I'd explain a way you could use linear interpolation from your table of values.
By allowing linear interpolation between values, the size of the look-up table can be much smaller.
This approach makes no assumption on the order of values in THETA(2,: )

[JohnCampbell]

Further to my last post, if THETA(2,: ) is ordered, then you can use linear interpolation for the values, using