I created a black and white conversion action for photoshop cs2, but it still requires you to make some adjustments on your own. The same basic principle applies as when you use the channel mixer for color images. The tutorial for the conversion without the action is below.
To do the black and white conversion without using the action, start out by creating a channel mixer adjustment layer, by selecting “Channel Mixer” from the layers menu.
Next, in the channel mixer dialog you want to check the box for monochrome. The default setting is for 100% red channel and 0% of green and blue. I like to have the histogram window open while I make adjustments to make sure I don’t lose any highlight or shadow detail.
There is no perfect channel setting for a black and white conversion, although I feel like equal amounts of red and green with about 20% more blue looks very similar to some traditional black and white films (don’t be too exact with this it is only a starting point, no reason to do 26.6%). Once you are happy with the results hit “OK.”
I normally do some contrast adjustments and some “dodging and burning” with curves in adjustment layers to get to an image I am really happy with. (Using levels, or the brightness contrast adjustments are fine too, there is just more control with curves.)
I also do some form of digital toning to all of my black and white images (I miss the smell of sulfur in the darkroom). An easy place to start with digital toning is using the color balance adjustment layer.
For a traditional sepia look I use 6 points of red and 6 points of yellow in the midtones, and then 3 points of red and yellow in the shadows and highlights.
You can make this more dramatic or more subtle, it is up to you. If you want a traditional archival/selenium look play with the blue channel.



Let’s see some before & afters, like in the channel mixing gig!
What was the image you were working on? It looks like the roof of a train station, like the one in DC.
good eye hud! especially since I have never published/printed/shown that image to anyone other than jerri before.
per gorjus’s request I updated the post with a before and after shot, also making Hud’s comment look silly (just an added bonus).
Is there a difference between “Flatten Image” and “Merge Visible” when combining layers? (besides the obvious difference of one only merging the layers shown)
Other than the obvious, Flatten makes the resulting layer “Background” and Merge Visible names the resulting layer the same thing as the layer you had selected when you merged.