Delta 100 and very long exposures

[hoffy]

Howdy,

I would like to do long exposure with Delta 100 (OK, truth be known, I need to cycle a roll through the camera.....it has a problem). I have done the research and have found the following formula to calculate the reciprocity failure

Corrected time = 1.15567 x (Metered time)^1.4379

The metered exposure time is approximately 240 secs, which using the above formula would equate to 3057 secs (or 51 mins for those who don't want to use a calculator!)

Firstly, do I have my calcs correct? I know reciprocity failure isn't linear, but this does seem rather high! If the calcs are correct, am I going to get some really bizarre contrast?

Cheers

 

[marco.taje]

Hi, Hoffy.

I don't want to sound obvious or patronizing, but why don't you use the corrections listed here?

http://www.ilfordphoto.com/Webfiles/201062894918374.pdf


Rgds,
Marco

 

[hoffy]

Not patronizing at all Marco, but yes, I have checked out the graph on the data sheet. The drama is that this graph only goes as far as 35 secs. I am looking at a metered time of nearly 8 times that!

Cheers

 

[Tom Kershaw]

hoffy,

The ILFORD data is not very helpful. Try this article from Photo Techniques:

http://www.phototechmag.com/index.php/archive/reciprocity

Tom

 

[hoffy]

Wow! Thanks Tom. My corrected exposure of 50 mins was way way out compared to the above!

 

[steven_e007]

I've got a fair bit of Delta 100 to use up and was also looking for reciprocity failure information. I found it strange that Fuji Acros had such excellent reciprocity characteristics - whilst the Delta 100 information published by Ilford indicates it is actually quite poor... Indeed, the Ilford sites still seem to have the same chart for all films, From HP5+ to Delta 100 - which just doesn't seem right.

After reading a few sites (especially astronomical websites) it appears that, according to the few people who have actually done tests, in fact Delta 100 is one of the better films for very low level exposure - nearly as good as Fuji Acros. It is indeed the reciprocity chart published by Ilford that appears to be dodgy!
One site suggested that the chart even over-corrects for FP4+

I haven't done any tests myself to verify this, it is just what I have found over the last few days - but if it is true it does suggest that Ilford might well be missing a trick, since in many places on the net you will find that the popular view seems to be that for very low light / long exposures... you should use Fuji Across as it has the best reciprocity characteristics. Similarly some astronomers and other low light users often discuss Kodak T-Max films - but rarely Ilford. Put off by that reciprocity chart, perhaps? :confused:

 

[hoffy]

Spot on there Steven. If what Tom's link shows is correct, the chart for Delta 100 would have had me over exposing by quite a lot. 50mins V's 9mins 20 seconds is quite a difference. I have to admit that If this is correct, to actually publish this data is quite bad on Ilfords behalf.

Anyhow, I have done my exposures tonight (I had to use the delta, as I needed to get a film out of a camera for a repair). I'll see what they are like when I process them on the weekend.

Cheers

 

[Tom Kershaw]

Spot on there Steven. If what Tom's link shows is correct, the chart for Delta 100 would have had me over exposing by quite a lot. 50mins V's 9mins 20 seconds is quite a difference. I have to admit that If this is correct, to actually publish this data is quite bad on Ilfords behalf.

Anyhow, I have done my exposures tonight (I had to use the delta, as I needed to get a film out of a camera for a repair). I'll see what they are like when I process them on the weekend.

Cheers

hoffy,

I think Harman / ILFORD Photo should be more proactive in terms of publishing accurate data concerning their films.

Tom

 

[Lee L]

Looks like you got your equation from a post in Large Format Forum. That's where I found your posted equation using google. I think I've seen this pop up here on APUG before. Your calculations are correct, but the equation is definitely not good for Delta 100, even though the equation appears to be derived from the Ilford data sheet that Marco linked to, and fits those recommendations. Kodak and Ilford haven't changed their generic reciprocity correction curves for 40 years or more. Most careful workers find them pointless with modern films.

From the Bond article:

In publications dated September 2002 (sent to me in January 2003), Ilford still had a single old curve for correcting reciprocity departure for HP-5+ and 100 Delta. I found the behavior of these films to be quite different from the curve, and also different from one another.

Here's an equation for Delta 100 that I regressed against the Bond data from the article referenced in Tom Kershaw's post #4 to this thread.

Corrected time = 0.0569814253867811 x (Metered time)^1.57549715077228 + Metered time

You'll find that this formula matches the Bond data within 3%-4% with metered times over 4 seconds, and the adjusted time for 240 seconds comes out as 560.44 seconds rather than Bond's 560 seconds. Feel free to pick your personal number of significant digits.

Here's a post with this equation and derived variables for the Bond data for the five films he tested in the PhotoTechniques article that Tom Kershaw linked to: (there was a url link here which no longer exists)

Lee

 

[M.A.Longmore]

Perfect Information !
Just what I needed.

Thanks,


Ron

From The Long Island Of New York, and the
Long Island @ Large Format Group, right here on APUG
.

 

[Lee L]

Perfect Information !

Well, I wouldn't claim that. I'd say "much more appropriate and much more accurate" than the old generic plots. But the real credit goes to Howard Bond and Patrick Gainer. Bond for doing the heavy lifting to get good data on modern films, and Gainer for his follow up article in PhotoTechniques on "Reciprocity Misbehavior". All I do, as shown in the post I linked to, is let the exponent from Gainer's equation float, where he uses a fixed exponent, and run the regressions, earlier with QtiPlot and more recently with SciDAVis (a fork from QtiPlot).

It's more a question of GIGO, the Ilford data being sadly generic and outdated (and giving rise to a very misleading equation based on bad data and posted on the internet), and Bond's being well done testing with modern materials.

Lee