This image is copyright ~LadyEruname and used here (with much thanks) only for educational purposes. You can see the original, uncropped version here.

So here's how I got from the image on the left to the one on the right.

Step 1. Sharpening. I used a third-party plug in to do the sharpening in this, but Photoshop's own built-in Smart Sharpen or even Unsharp Mask (with careful attention to the settings) will work.

Step 2. Greyscale conversion. Since there's a lot of red in the skin tones, I figured I could get some good contrast out of the green channel (the opposite of red, and therefore where most of the contrast is bound to be). I fired up the Calculations filter under the Image menu, and used the following settings:

Source 1: Green Channel
Source 2: Green Channel
Blending: Overlay 100%
Result: New Channel

That got me a nice, contrasty image, but some of the highlights seemed just a little too bright still, so I ran the filter again with these settings:

Source 1: Red channel
Source 2: Red channel
Blending: Multiply 100%
Result: New Channel

I now had two new channels, one with a lot of contrast all the way through, and the other less contrasty, but with slightly darker highlights. I ran the filter one more time to merge the two:

Source 1: Alpha 1
Source 2: Alpha 2
Blending: Darken 100%
Result: New Channel

And there it was, a nice, richly toned and contrasting image. I copied the channel and pasted it as a new layer above the original. Then I moved the original layer above the greyscale version, changed the layer mode to Color and reduced the opacity to 30%, so that the b&w image would take on a little of the tone of the original.

Step 3. Tweaking. Finally, I cropped the image in close to his face to remove some of the background and really focus the eye on the skin tones. Then I carefully burned a few corners and edges to accentuate the craggy, sunken features. I did this on the greyscale layer, not the colour layer, so as not to mess up the hue.

So, will this work for your image?

Well, maybe. The thing to keep in mind is that each image is different. This method works in this case because the old man’s face is rather reddish, the shot is well-exposed with fairly even lighting overall, and there is sufficient detail already in the image that sharpening it doesn’t introduce any weird artefacts. You’ll also notice that I’ve cropped this pretty tightly compared to the original photo. If your own portrait is brighter lit (say, under a bright, noontime sun, for instance) there may not be enough detail in any of the channels to create a nicely-toned greyscale image. The best thing to do is to try every combination of merging two channels with every mode (I find Multiply, Overlay and Hard Light tend to work the best) to give you the best possible tonal range overall. If you can get close, but just can’t get the shadows looking right, for instance, try it again with another combination that gives you good shadows, then merge your two results using either Darken or Lighten, depending on what gives you the best-looking result.

Bottom line: experiment with it, try a variety of things, and remember that to get a good overall tonal range, you may need to try merging channels several times, then merge the results.