Joe, the water bath doesn't increase the contrast. It reduces it.
Let's say you have a negative that's so contrasty that it doesn't print well on grade 2. What does that mean? It means that if you choose an exposure time long enough to get good highlights, the shadows will be too dark. If you choose an exposure that gives good shadows, the highlights will be too light. The difference between the densities of the highlights and shadows on the negative is higher than the exposure range of the paper, and as a consequence you cannot find one single exposure time to give good shadows and highlights at the same time. Normally you would either use a lower grade of paper, or a lower contrast filter, but with Lodima you have no such options (assuming you're already using grade 2).
What you need is a method that gives you good highlights, but without allowing the shadows to get too dark. That's the water bath technique. It works very well with amidol because it's an extremely powerful developing agent.
In order to get good highlights, you must give them enough exposure. There's no way around that. So, you expose for the highlights. Then insert the paper into the developer and carefully watch the progress of the image. When the shadows get close to the desired density, quickly remove the sheet of paper from the developer, immerse it into a tray of plain water, and leave it there without agitation for the rest of the development time. The image will continue to develop proportionally to the amount of developer adsorbed in the emulsion. The shadows exhaust the developer very quickly, so they will soon stop developing and won't get too dark. In the lighter areas however, there's still plenty of active developer that continues to work until the highlights are fully developed.
Now you understand why I said on the other forum that a plain water "stop" bath won't actually
stop anything if you use amidol.
That's a rough description. You have to experiment to find a proper exposure time (it's not necessarily the exact same time that gives good highlights with normal development), to learn how to assess the progress of the development so you know exactly when to snatch the print from the developer, and so on. You'll have to waste some paper first. And you shouldn't expect miracles; it works, but only to a limited degree.
I don't know what more I could say about it. I've learned it from Michael and Paula, so I can't really tell you anything that's not already in their articles.