pcmhacking.net

All good. Consider it a late Xmas Present.

Simon

The1 wrote:that's correct but i can't see any calculations, youve done it manually?

I copied the data in the 250 and 500 cells down and used the fill function to fill the empty cells between them.

Then highlighted all data in the cells 250 to 500 and clicked the little box at the bottom right corner of the highlighted section and dragged it across to populate the remaining cells. I'm sure there is a name for that just know how to do it not whats its called.

Not too sure if its the best way to go about it but the numbers look close to where it needs to be

yeh i needed it to be automatic, so people would be able to paste in there data and copy the new data out.

I can tweak the sheet to add in extrapolation at the edges of the table if you like. L me know as it would be a ten minute job.

Simon

attached is a spreadsheet that converts and expolates figures between the old scale and new scale.
Any number you add to the red values will update what it will be in the new table.
Is this how you wanted it?

what slewinson did above is what i was after.

The1 wrote:what slewinson did above is what i was after.

Been thinking more about it and I feel that you could get much better results by doing a polynomial curve fit for each row and then using the resultant formula to populate the lower table. That would potentially give a smoother end result than the stepwise linear interpolation.

Might give it a go when I have a spare half hour or so.

Simon.

sweet

Here they are... Took a bit more "massaging" of the lookup formulas then I anticipated

There are also graphs to show the data on a second tab in the sheet. I used third and fifth order polynomials to get lines of best fit for each row and then extrapolated this into a new table (top right of the four). This is then re-ranged (bottom right) and then used to backfill the blank areas on the original linearly interpolated table (bottom left).

There are some discontinuities but it goves you an idea of what might happen. in this example there is a step that needs to be adjusted at higher flows above 550 ish. Ther are also some -ve numbers that need to be manually culled.

Have a play.
Simon

uploaded new zip files, just tweaks and fix's here and there on xdf/adx's