Thank you very much, Gerry, for your tests without pre infusion.

I finally got my Vibiemme DD! It really is a gorgeous machine. And these tests will certainly help me get the best from this machine. It seems that it would be a waste of a fine machine not to plumb it in, since it's full potential is only achievable with the pre-infusion.

If these tests really are as factual as they seem to be, then it seems to me these results could possibly be transferable to other rotary pump machines. And it seems that when trying to measure temperature on machines with a rotary it is important take the flow into account.

Another question now is how much of the better taste in the cup is due to the lesser temperature variation, and how much of it is due to other positive effects of pre-infusion.

Any rotary-pump, E-61 machine with a manual (lever) group which is plumbed should be capable of manual preinfusion. I preinfuse EVERY extraction.

Pieter, when I was dialing in the grind for Delirium I noticed a fact that you might find useful until you get your machine plumbed in. Without preinfusion, the temperature excursion at the basket is a strong function of average shot flow rate. (That's total shot weight divided by total shot time after the basket bottom colors, which for my 0.050" gicleur starts about 10s from the beginning of the shot without mains preinfusion.) From the data shown here, it goes from 4F for a flow rate of 0.33g/s to 1F for a rate of 1.2g/s.

Incidentally, this highest flow rate at the temperature shown here is a sweet spot for Delirium, which I'm exploring at the moment. I've confirmed the 1F at 1.3g/s today with excellent gustatory results. Because of this small temperature excursion, preinfusion hasn't improved much on the flavor of the 1.2g/s shot shown here. So, if you get better taste at higher flow rates, this may be part of the reason.

Perhaps this should not be a surprise, since the recommendation on the Delirium bag suggests about 1.3-1.5g/s the way I'm measuring it. In fact, it may be my penchant for dense ristrettos that has steered me toward lower rates. Most people are probably using about 1.5g/s even for ristrettos, judging by the Favorite Espresso Blends threads. Favorite Espresso Blends 2010 Favorite Espresso Blends 2011 In this case, preinfusion may not add much. I'll explore that in coming weeks.