Since you've asked, this is something we've spend a ton of time on...
I've been looking for the perfect tamper for quite some time as I don't have a C and C machine to make my own. Basically I want the tamper to match the shape of the diffusion screen as closely as possible. The reason for this is simple: to allow even expansion at the initiation of flow. This means that if all things are operating properly (the screen is clean and in good shape-not dented or flattened from use) the water column will contact the entire surface of the coffee all at one time. This allows the coffee to expand as a whole unit and rise to meet the screen uniformly. A small amount of room for expansion allows the surface area (grind) to do the work of providing resistance against flow. 1 to 1.5mm is all that is necessary. More than that -(2mm or greater)- allows too much rise resulting in a large quantity of particle migration (fines dropping to the bottom) causing overextraction and interupted flow. No room for expansion (an overpacked basket) requires a coarser grind to allow any flow to occur resulting in either an underextracted shot or any overly long flow duration (to balance sugar and colloids with acid) which causes the sugars to fall apart making the texture and aftertaste suffer.
Different infusion methods will yield slightly different results. Full pressure infusion (8.5 to 9 bar) from beginning to end requires a perfectly level tamped surface that matches the shape of the screen to avoid channeling. Low pressure initial infusion (4 bar) followed by full pressure is more forgiving on tamping variations by allowing the coffee to expand slower but still will show directional flow if the surface was angled. Low pressure followed by ramped or staged pressure is a small amount better in the forgiving department as the expansion progresses as the pressure builds. Even though pressure variation helps minimize the detriments of poor tamping it does not eliminate them. This means using a tamper that does not match your screen will allow for a path of lesser resistance for the water to flow toward before allowing infusion causing an uneven expansion. Flat screen with a round tamper equals an uneven rise in the center of the coffee whereas a flat tamper on a rounded screen equals an uneven rise on the outer edges of the basket.
Long story short, disparate screen and tamper shapes cause expansion problems and channelling which make the coffee suffer. Any path of lesser resisitance equals underextraction through that path or channel. Underextraction means acid out of balance even in small amount caused by a slight discrepancy. To eliminate that possibility your variables have to match. Flat tamper with a flat screen. Round tamper with convex screen. If the curves don't match you will create problems so take care in matching the radius of the screen to the tamper head.
We found the Caffelat 58mm Curved Tamper to be the closest match to the screens we are using on our Synesso machines. This is the same screen used on alot of machines including La Marzocco. Take your screen with you when checking out tampers and you should be able to find one that matches.
Hope that helps,
Haven't played with a C-Ripple yet (need to) but I can tell you that I most certainly prefer a US Curve (aka- slight curve) to anything else, for the same reason that Jason mentioned.
I guess my thing with the matching of the piston shape to the screen is... who's to say the water flows from the screen in the same shape as the screen? Pretty sure it doesn't. Also, I get no channeling (like ever) from my US curve against LM's flat screen. I just don't think the puck expands as much as people think. They see the impression of the screen on the puck and don't assume that perhaps its from the giant, sucking, 3-way pulling the top layer against the screen. Am I wrong about this? I truly don't know.
Sorry to derail this thread...
Sorry to derail this thread...