AssafL wrote:This is pretty simple. The faster flow is due to less density in the puck.

Short PI leaves air in the puck. When pressure is applied the air compressed and puck density increases, slowing flow.

My own speculation here involves hydrophobicity--wetting properties of materials--as one factor; amongst a whole lot of them.

Coffee has a lot of hydrophobic (hard to wet) components. (Put a drop of water on a coffee bean, or grounds and you'll see that it doesn't aggressively wet.) As wetting/extraction occurs, these forces are modified by extractables.
Overcoming these forces requires work and this work is accomplished during preinfusion.

I try to minimize these forces using a spritz of lecithin in water on the puck before extraction--lecithin is a surfactant and emulsifier-- these actions work to enhance wetting and flow. Consequently, I can grind finer and achieve useable flows.

AssafL wrote:Are the group temperatures verified the same? Perhaps water temp is higher and maybe causing more CO2 to be released?

No, I haven't got a Scace. However, my observations on body and mouthfeel hold for two different Spaz S1 models (V1 and V2), over a wide range of temperature settings, coffees, grinds, doses, etc. The difference in body and mouthfeel is not a subtle phenomenon. I'd like to understand why.

Denis wrote:The vibe pump has less efficiency at high pressure compared to a rotary pump.

I believe you barely hit 3- 3.5 g/sec at 9 bar with a vibe pump, while with the rotary pump you hit more and in a shorter period. Faster flow and a more aggressive pumping means better emulsion in the puck creating that thick crema Jim is talking about.

I agree that this is a likely explanation, although the aforementioned vibe vs rotary pump studies failed to note any differences in body and mouthfeel.

doublehelix wrote:My own speculation here involves hydrophobicity--wetting properties of materials--as one factor; amongst a whole lot of them.

Coffee has a lot of hydrophobic (hard to wet) components. (Put a drop of water on a coffee bean, or grounds and you'll see that it doesn't aggressively wet.) As wetting/extraction occurs, these forces are modified by extractables.
Overcoming these forces requires work and this work is accomplished during preinfusion.

I try to minimize these forces using a spritz of lecithin in water on the puck before extraction--lecithin is a surfactant and emulsifier-- these actions work to enhance wetting and flow. Consequently, I can grind finer and achieve useable flows.

Tell us more about your experience using lecithin.

UBET...
I like to spray mineral solutions onto my pucks before extraction--this way I don't have to worry about what I put into my boiler:
Spraying mineral solutions on to pucks—a facile way to contour water minerals on a per cup basis

One ingredient, that I would never put into a boiler is lecithin. I make a 1% wt/wt in distilled water using soya lecithin granules. They're hard to disperse. When fully dispersed, your "solution" will be slightly milky-- a little heating will help. To this I add 1% overproof rum as a preservative and store in a spritz bottle in the fridge.

Simply stated, added lecithin may be acting like "greasing the water."

The crude thinking behind using lecithin is to enhance water flow around grounds by enhancing the water/ground boundary layer to increase water flow. This allows you to grind finer and still have the same flow rate.

What might be happening with added lecithin during preinfusion could be enhanced wetting of the grounds--lowering the hydraulic resistance during the extractions phase. Hard to say if the amounts I'm using are having a significant emulsifying effect???
Thanks for asking...

AssafL wrote:That is very easy to show. Grind until the no-PI stops the flow completely. Blast the puck with full pressure and wait a few seconds... If it stops - great!

Now - slowly down-ramp the pressure (not a 3-way "exhaust" - you want to keep the puck intact). Wait until pressure dissipates and now wait for it to PI.

Now ramp pressure again - flow will be as if the first "blocked" pour never occurred.

Air is a spring - and as long as air is in the puck you can play with it - and the process is reversible. Displacing the air with incompressible water is irreversible.

Not saying you're wrong....just that I'm still not quite understanding how this demonstrates trapped air as the reason. In my minds eye I visualize the effects of PI as partially negating or undoing some of the compression that happens during the tamping of the puck. Thus a finer grind is can be used to control the flow. The self loosening and resettling of the grinds during PI can also explain why it sometimes helps remedy the effects of a poorly tamped puck or one that could have benefited from WDT.

In your scenario above, would not that little bit of air trapped within the grounds still remain within the boundaries of the pressurized system even after PI? And if not (were it to get pushed out the bottom of the basket) then I'd imagine a quick ramp up on a very fine grind would also eventually push it out the bottom just the same. In the case of the quick ramp up on a very fine grind the high pressure would still continue even after the puck's eventual saturation. So maybe I am missing something, but in either scenario I imagine any presence or lack of air would still be the same. Although I could be wrong.

I suppose everyone has a different approach depending on which elements of espresso they prioritize. As for myself, grinding finer along with a nice PI is the route I prefer to take on very light roasts.

Air is a spring. And this as long as air is the dominating factor - it doesn't matter if you pressurize the puck and then Let it dissipate and then PI or if you first PI and then pressurize the puck.

This springiness goes away once fingering flows find the bottom of the basket.

Some air may stay (maybe crevices that are inaccessible to water) and hence the puck will have some residual springiness.

In regards to tamping pressure and undoing it - even 30lbs is so inconsequential compared to the 8-9 bar that it is unlikely. I'd say the majority of tamping problems is either non uniform density (ie the tamper pushes only a part of the puck) or moving it around (breaking the seal). In both cases the operator is leaving water a way out that isn't uniform.

And I have no doubt the hydrophobia has a play. Mainly during PI helping the fingering flows advance quicker. Hydrophobia may also explain why the compressed puck resists water so perfectly (remaining dry). But also the crushed pathways may have something to do with it as well... .

This "air being displaced" hypothesis came from Jake's PI thread. It was developed after the "springiness" was observed and we did a few tests to see if indeed the puck was "springy" before PI and "not springy" after PI completed.

Displaceabke air was the only candidate since anything chemical or physical like emulsion would not be reversible. (So you could PI after pressurizing the puck).

Which left non displaceable air which would remain springy even after PI which wasn't observed (actually the opposite was).

Other hypothesis (like emulsion creation) helped but none covered all the observed phenomena.

Thanks for taking the time to explain in further detail...I do appreciate it. However, I think you left out the key point of my observation that I'd invite you to consider. Please realize though that these are just my own casual observations. They're certainly not scientifically formulated by any means and could be partially or wholly inaccurate, and may not account for any differences in machine models or operation.

AssafL wrote:Air is a spring.

You're absolutely right. You could even say that once the basket is locked into the group head there is more air trapped behind the puck than within the puck itself. However, I'm just commenting that I find that springiness is inversely proportional to the flow rate. The lower the flow rate, the more the springiness. And the higher the flow rate, the less the springiness. And that the air is still trapped somewhere in the pressure chamber either way.

AssafL wrote:And this as long as air is the dominating factor - it doesn't matter if you pressurize the puck and then Let it dissipate and then PI or if you first PI and then pressurize the puck.

There is a third option here that I mentioned in my last post. It's not a very tasty one, but it was the reason behind my thought process. And that is pressurizing the puck and not letting it dissapate. By locking it up and keeping the high pressure, eventually the entire puck will saturate and begin to flow out with the tiniest trickle and pretty much stay that way throughout the shot. In this scenario the the puck is fully wetted and yet the flow rate will remain hindered.

AssafL wrote:This springiness goes away once fingering flows find the bottom of the basket.

My own experience has been that in the lock up, constant high pressure scenario, after the puck is eventually fully saturated there is a window in which it will remain springy before the flow rate increases. Probably easier to demonstrate this on a lever machine. I'd call it a 'window' of a period because after the puck is saturated you will still be able to bounce the lever (springy) several times, although with each bounce the density of the fully saturated puck is slowly loosened up (much in the same way a clogged sink gets loosened via plunger).

AssafL wrote:In regards to tamping pressure and undoing it - even 30lbs is so inconsequential compared to the 8-9 bar that it is unlikely.

Here's another part of what's led me to my thoughts about that. If I grind really super fine with no tamping whatsoever (just a fluffy, very full basket) I can still manage to get a 25-30 sec pull and a great tasting shot with crema, albeit with a little different mouthfeel. But if I were to tamp that same super fine grind I would lock it up for sure. So I've always found tamping to play a pretty big role, especially in regards to grind size.


Anyway, I found the whole thread very interesting. And my own experiences are definitely in agreement with RapidCoffee's conclusions about preinfusion and mouthfeel.

Jakuka wrote:....
You're absolutely right. You could even say that once the basket is locked into the group head there is more air trapped behind the puck than within the puck itself.

That air is pressurized to whatever the pressure above the puck is. However it does form a delay (the time it takes to fill up to create the pressure).

However, I'm just commenting that I find that springiness is inversely proportional to the flow rate. The lower the flow rate, the more the springiness. And the higher the flow rate, the less the springiness. And that the air is still trapped somewhere in the pressure chamber either way.

I can't see why air compliance (springiness) is related to flow rate of the incompressible liquid.

There is a third option here that I mentioned in my last post. It's not a very tasty one, but it was the reason behind my thought process. And that is pressurizing the puck and not letting it dissapate. By locking it up and keeping the high pressure, eventually the entire puck will saturate and begin to flow out with the tiniest trickle and pretty much stay that way throughout the shot. In this scenario the the puck is fully wetted and yet the flow rate will remain hindered.

No. If you grind fine enough there will be no flow. If you then PI it (even after pressurizing) - flow will happen.

My own experience has been that in the lock up, constant high pressure scenario, after the puck is eventually fully saturated there is a window in which it will remain springy before the flow rate increases. Probably easier to demonstrate this on a lever machine. I'd call it a 'window' of a period because after the puck is saturated you will still be able to bounce the lever (springy) several times, although with each bounce the density of the fully saturated puck is slowly loosened up (much in the same way a clogged sink gets loosened via plunger).

Yes - that is observable on any machine. Again - our hypothesis supports this due to fingering flows. When the first "fingers" find the bottom of the puck and form the first drops - there is still quite a bit of air between the fingers. So it will be springy. If you let PI continue (especially at 0 pressure - the so called Rao pause that some of us did before the DE existed) - these fingers will expand and coalesce.

Edit: More on fingering flows here: My long and rambling path to preinfusion/pressure profiling

Here's another part of what's led me to my thoughts about that. If I grind really super fine with no tamping whatsoever (just a fluffy, very full basket) I can still manage to get a 25-30 sec pull and a great tasting shot with crema, albeit with a little different mouthfeel. But if I were to tamp that same super fine grind I would lock it up for sure. So I've always found tamping to play a pretty big role, especially in regards to grind size.

I am sorry but to me it sounds like channeling... Hence the different levels of TDS/EY.

Anyway, I found the whole thread very interesting. And my own experiences are definitely in agreement with RapidCoffee's conclusions about preinfusion and mouthfeel.

+1

AssafL wrote:I can't see why air compliance (springiness) is related to flow rate of the incompressible liquid.

I'm guessing that since my example here still had a small pocket of air in front of the piston that maybe that's where it came from. As for relating to the flow, that just had to do with the amount of give in the 'surface' or puck in this case. The more flow from the puck the more give it had to reduce the sensation of springiness. Much in the same way that a rubber ball appears less bouncy if thrown against a softer surface like a wool rug or pillow.

AssafL wrote:Yes - that is observable on any machine. Again - our hypothesis supports this due to fingering flows. When the first "fingers" find the bottom of the puck and form the first drops - there is still quite a bit of air between the fingers. So it will be springy.

After a couple minutes of continuous pressing without preinfusion and only steady drips coming out I wouldn't have thought that the puck still may not have been fully saturated. After reading some of the the info in your link I'm very much inclined to take your word for it here that it likely wasn't. It seems hard to fathom initially, but it definitely could explain it.

AssafL wrote:I am sorry but to me it sounds like channeling... Hence the different levels of TDS

That's exactly what I thought would happen, although I was very much surprised. If I poured carefully enough and was slow to ramp up the pressure I didn't get strong flavors typically associated with under or over extraction. Not saying it was a 100% completely flawless shot by any means. But it was fun to try.

AssafL wrote:Edit: More on fingering flows here: My long and rambling path to preinfusion/pressure profiling

Lots of great info in that thread! Thanks for going back to look it up and share it with me. I haven't had the time to go through all of it yet but I'll definitely check it out some more later. I think I'm starting to see the light now and am able to reconcile or account for the differences I was envisioning. Thanks for bearing with me.