vit wrote:I think that measuring the pressure is not relevant. The goal is not to exceed 100°C, so as long as you are a few degrees lower, the pressure is very close to atmospheric pressure, only several mbars above it, being equal to the height difference between the water in the tank and water in the basket/spout + some resistance of the coffee to water flow.

You have repeatedly made this claim about the pressure without any evidence. Suppose the puck is a few degrees below 100°C. At 97°C the saturated vapor pressure of water is 0.9 bar. In a Moka Express 3 cup the head space volume at the onset of strombolian will be approximately 1/7th the initial volume. So the partial air pressure will be ~0.2 bar and the total pressure will be 1.1 bar. But that assumes that the puck temperature is the same as the vapor temperature. Navarini found the latter to be quite a bit higher.

It's also clear from Navarini's work that the resistance of the puck is significant. You could argue that he ground too finely. He only says that he used "coarsely ground powder for stove-top coffee makers".

I did an experiment where I forced hot water through the grounds in my moka pot with a pump. I aimed to produce a flow rate that was consistent with actual moka pot brews and I measured the resultant pressure. I used a range of grinds, from very fine to very coarse. Even the coarsest grind I used resulted in a gauge pressure which started at 0.1 bar and rose to 0.2 bar. With the grind I typically use in a moka pot the pressure started at 0.5 bar and rose to 0.6 bar. The finest grind drove the gauge pressure above 1 bar.

Measuring the pressure isn't relevant to you. I suspect measuring anything in a coffee pot seems silly to the vast majority of people. It certainly is not necessary, and maybe not even that helpful, if the goal is to make good coffee. But I think understanding the physics has intrinsic value.

Well, his measured pressure remained at 1 bar flat (meaning overpressure close to 0) until TI2 reached around 95°C (fig 7). Then it started raising towards 2 bar (overpressure 1 bar) as the temperature went towards 120°C which is as expected
I'm also fine about your measured overpressure 0.1-0.2 bar or even more ... it would happen if the temperature goes several degrees over 100°C
So I'm not sure where is our disagreement

That figure isn't very detailed. If you look at figure 3-b you can see that the pressure began rising above atmospheric when the tank water temperature (TI2) was still below 90°C. When the water reached 97°C the pressure was a little less than 1.2 bar.

That's the tank water temperature. Navarini didn't measure the temperature of the grounds, only the water, vapor, and the temperature of the liquid extract just above the puck. So we can only speculate about the actual brewing temperature. The liquid extract temperature (TC10) remained under 100°C at strombolian and the tank pressure at that point was a little over 1.2 bar. It's probable that there is a temperature gradient in the puck which narrows as the brew progresses. So the bottom of the grounds could have been above 100°C. We just don't know.

We also don't know if a brief exposure to above 100°C is the culprit in nasty, acrid, burnt moka pot coffee. Navarini didn't provide us with tasting notes. Maybe it has to go higher than 100°C.

I have a 2-cup Bialetti Brikka. If you believe Gabór's measurements, in the final moments the coffee grounds exceed 120°C. In the hundreds of times I've brewed with my Brikka I have never had that burnt, bad moka pot taste. Sometimes the coffee is over- or under-extracted and I have to adjust the grind and/or brewing time, but it's never burnt. And I grind fine and usually tamp an extra gram or two of coffee into the funnel. So it's likely seeing some pretty hot water/steam at the end and yet it produces delicious coffee.

As about Brikka, it seems that short exposure of the coffee to higher temperatures isn't problematic. They make instant coffee at around 140°C

Sure, we can only speculate about brewing temperatures in case of Navarini. But it's safe to assume that at the bottom of the basket it was somewhere between tank temperature (calculating also the time that water needed to reach the basket through the pipe - in the meantime, water in the tank heated up a bit) and not above 100°C at the top of the basket, because that side is open to atmosphere, so pressure is equal to atmospheric - if the coffee is hotter, it starts boiling so the temperature lowers (it is a main source of the foam in Brikka). There is definitively some temperature gradient in the puck during brewing even if temperature doesn't reach boiling

Didn't get what you meant by sentence "In a Moka Express 3 cup the head space volume at the onset of strombolian will be approximately 1/7th the initial volume. So the partial air pressure will be ~0.2 bar and the total pressure will be 1.1 bar". Can you explain what do you mean by "head space"- Pressure in the moka pot depends on both the temperature of mixture air/water and on amount of water that flowed from the tank into basket at particular moment

dominico wrote:E vabbe', mi son trovato con un po' di tempo libero ora, quindi eccoci qua:

English translations of Lucio's blog posts above:
Part 2:
At this point you could simply screw on the top but I wanted to dust off one of my old techniques that has gained many enthusiastic adopters across the ocean: the addition of a paper filter.

For those who don't know what it is, I suggest you click this link. (goes to a page I have yet to translate into English, sorry, but at any rate links back to this HB thread: Bialetti Sneeze)

This time I cut the filter with a bit larger circumference, which, by the way, doesn't have any negative impact on the effectiveness of the rubber seal.

I use an Aeropress filter, that is already round (and in some cases has the correct diameter).

Edit:
I now have read in some other thread you do that as well. (It is kind of obvious to use those I guess )

vit wrote:As about Brikka, it seems that short exposure of the coffee to higher temperatures isn't problematic. They make instant coffee at around 140°C

Are you really using instant coffee as an example? :-)
Bottom line is that we really don't know what the time/temperature limits are. We only know that too much of too hot is bad.

vit wrote:if the coffee is hotter, it starts boiling so the temperature lowers (it is a main source of the foam in Brikka).

In the Navarini experiments the average maximum pre-strombolian temperature at the top of the puck (TC10) remained below 100°C (table 1). And boiling does not result in a drop in temperature. Where did you get that idea?

vit wrote:Didn't get what you meant by sentence "In a Moka Express 3 cup the head space volume at the onset of strombolian will be approximately 1/7th the initial volume. So the partial air pressure will be ~0.2 bar and the total pressure will be 1.1 bar". Can you explain what do you mean by "head space"- Pressure in the moka pot depends on both the temperature of mixture air/water and on amount of water that flowed from the tank into basket at particular moment

From Navarini, initial air volume in the tank was 20cm3. At the onset of strombolian 120g of water had flowed. So the volume at that point would be approximately 140cm3, or 7 times the initial volume. The initial air temperature was about 25°C and at strombolian was about 97°C. So at strombolian, P1 = P0 * (V0/V1) * (T1/T0) = (1bar) * (20cc/140cc) * (370K/298K) = 0.18 bar.

-------------------
EDIT: As many times as I've looked at the Navarini paper there's something I should have caught but I always glossed over it. I treated all the figures as coming from the same experiment. But it's actually fairly clear that figure 3 and all the other figures have a different pressure curve. Even figures 6 and 7 appear to be from different runs. So my calculation above, based on figure 3 but assuming that strombolian occurred at the same time as in figure 7, actually applies to an earlier point in the brew represented by figure 3. Navarini doesn't tell us when strombolian began for the figure 3 brew but the air/vapor pressure and temperature must have been higher at that point.

Going back to figure 7-b, at strombolian the air temperature isn't shown but is probably approaching the water temperature (117°C) at that point. So the air partial pressure would still be about the same (0.2 bar) and the contribution from the water vapor would be 1.7-1.8 bar, or nearly 2 bar total, which is what he shows as the measured value.

I'll bet his coffee tasted bad. :-)

tohenk2 wrote:I use an Aeropress filter, that is already round (and in some cases has the correct diameter).

Edit:
I now have read in some other thread you do that as well. (It is kind of obvious to use those I guess )

I think opinions vary. I personally like the fine particulates in moka coffee. I think it is one of its distinguishing features and a paper filter would reduce it somewhat. Would you paper filter French press? Or an espresso?

Of course if you're getting a lot of larger particles that sink to the bottom of the cup that's a different story.

Instant coffee doesn't taste good of course, but it doesn't actually taste burnt either, that's why I used it as an example. So I assume we agree that shorter exposure to "higher than optimal" temperatures isn't very bad

In the basket of moka pot, there is a pressure gradient during extraction. At the lower screen, pressure equals to the pressure in the pot, which can be above atmospheric, while at the upper screen (TC10), it's equal to atmospheric, because it's open (ok, add 0.005 bar for the height of the water column). So within the basket, coffee can have a temperature above the boiling point, while at the top it can't. So if it is hotter towards the end of extraction, it starts boiling while flowing upwards, to avoid being superheated, which reduces the temperature of fluid phase and produces some foam. In Brikka, there is a lot of it, however it doesn't last because bubbles are filled mostly with vapor, unlike in espresso where they are filled mostly with CO2 ....

Yes, I got what you meant by head space in the meantime (you wrote 1/7 instead of 7x, which confused me)

Here's the table with summary of my measurements at beginning of strombolian phase (without the one with wet basket) and Navarini's measurement in the last column, with the (over)pressure in the pot calculated. I also pasted formulas that I used. I assumed that temperature in the pot was 2-3°C higher than measured near the bottom of the basket (there is some delay before water from the pot reach the thermometer, which also has some delay). Results look within expectation to me

vit wrote:Instant coffee doesn't taste good of course, but it doesn't actually taste burnt either, that's why I used it as an example. So I assume we agree that shorter exposure to "higher than optimal" temperatures isn't very bad

I can't say with any certainty just how long and how hot. And to me most instant coffee does taste kind of burnt. The processes used to produce instant coffee so little resemble what we do at home that it's hard to use this as an example.

vit wrote:Here's the table with summary of my measurements...

Looks good.

So I guess we're done. I think so. I hope so. I keep getting that warning about starting a new thread. :-)


In the end I think it shows that the technique Lvx/Lucio proposed has merit.