Page 1 of 1

1 to 1 scanning from Pentacon Scan5000

PostPosted: Thu Apr 10, 2003 1:05 pm
by Rob Rayner
I am using Silverfast AI 5 within Photoshop 6 (Windows 2000)
Imaging device is a Pentacon Scan5000 camera
This has recently been installed so we are on a learning curve at present

Our intended use in production is to capture large originals that are either too thick or too fragile to put thru a large format roll fed scanner. The TIFF files are then printed using large format roll feed plotting devices.

It is essential in most cases that these images are captured and printed at exactly the same size as the original. Usually these are plans with measurement scales.
The originals also do not necessarily come on standard paper sizes (A1, A2, A0 etc..)

Scan resolution is usually 200dpi (no more than 300dpi)


Am having some difficulty setting up the frame window.
With Q factor at 1, percent at 100% and lpi at 200 the frames created in the prescan do not show the correct width and depth to the originals. They are actually somewhat bigger and this is the size that is transferred into Photoshop

So far, I have found that with Q factor at 1 and lpi at 200, setting the percentage to 70.7 gives the correct 1:! sizing. 250lpi needs a different percentage however. Possible also different Q sizes.

Is this going to be a trial and error process or am I missing something? Is there an easier way (or formula to be applied) that will give us 1 to 1 capturing of the originals in question?

Appreciate any help on this :-?

PostPosted: Fri Apr 11, 2003 7:08 am
by LSI_Flyvbjerg
Dear Mr. Rayner.

In the frame panel, you can see the original size, the scale and the output dimensions.

Please set the output dimensions to the dimensions that you need. The scale should then be adjusted to the correct values.

Best regards.
Eric Flyvbjerg

PostPosted: Fri Apr 11, 2003 5:06 pm
by Rob Rayner
Thanks Eric.

I have been doing this already in order to build a matrix of Scalings to QFactor&LPI settings that produce 1 to 1 sizing.
282lpi (final with Qfactor) seems to be the anchor value which produces 1 to 1 reproduction at 100% scaling.

Original sizes are shown in pixels which is not too helpful

Manually setting the required output size would need our operators to take a ruler to each item being scanned. This is not too practical and I'd hoped to avoid this. It also relies on accurate frame creation!!

Wondered if there was an easier solution here for our operators rather than using rulers on delicate orignals or refering to a complex lookup table.

Rob