Flair Espresso news - Page 6

Read about current and future offerings from the site sponsors.

#51: Post by vit »

They actually have a probe in the tip of metal pipe, so they are not that "instant" as he calls them, simply because some time is needed until the heat is transferred through the metal and small amount of surrounding air to the probe inside it. You can make it instant by removing the pipe (on some models you can simply unscrew the whole bottom part with the pipe and expose the probe

On Classic I also measured some difference between the top and bottom of water column (with probe, so considerably more precise location wise). Even plastic piston, if not preheated, introduced some osculations at the start of the pull in measured temperature (up to about 2°C), and sometimes measured temperature actually started rising again after end of the pull for 20-30s for almost 1°C... but generally, these differences were much smaller. On his video, thermometer is showing 98°C at the top of the cylinder at one moment, while temperature measured at the bottom (which obviously isn't just the tip but 1-2cm of the tip) seems to be much lower

Maybe the metal sleeve around the bottom of the cylinder and portafilter indeed draw much heat. On Classic, it's plastic + very thin metal, so not much heat consumed there ....


#52: Post by jpender »

It looks like he is using this thermometer. I have a very similar probe. I'll bet at a near boiling temperature it's response time isn't the issue. It certainly wouldn't explain why the temperature drops so much.

Probably the reason is exactly as the guy said. It's because the chamber was open on top. Evaporative heat loss would likely dominate. So his graph of the temperature declines isn't very useful.

BPlus: turning your coffee spirit
Sponsored by BPlus

#53: Post by renatoa »

The smaller Flair group, if left open at top, without a warming sleeve, keeps the temperature a lot better, so I doubt this explanation.
5 degrees in 35 seconds simply can't be explained as heat loss, just think this is the average rate of temperature change during coffee roasting !


#54: Post by vit »

Even my csezve that I use for preheating Classic isn't cooling down this fast with similar amount of water and it has less than 1/5 of the thermal mass of Flair 58 cylinder, no insulation, no heater and several times bigger water surface where the water is evaporating ...

Also, with Classic, first simulations of this type performed by some of our members together with boiling preheat returned less temperature drop than I measured in real conditions, because they didn't take into account real quantity of water (which is decreasing during the pull, so is the thermal mass that cools down). Here, it is much worse than that despite additional heater + decreasing of water quantity during the pull also wasn't taken into account

I'm still hoping that there was something wrong with measurements, but ... to me, is still looks like too much heat was lost through the metal disk on the bottom of the cylinder which didn't exist on previous versions and 10min might not be enough to make it hot ...


#55: Post by jpender »

renatoa wrote:The smaller Flair group, if left open at top, without a warming sleeve, keeps the temperature a lot better, so I doubt this explanation.
5 degrees in 35 seconds simply can't be explained as heat loss, just think this is the average rate of temperature change during coffee roasting !
If all that metal in the Flair is really at temperature then I agree with you, it's too fast a decline. But is it? We don't really know how hot the metal is. In any case, to say that it's dropping so fast because the thermometer has a sluggish response doesn't make sense either.

So it seems we have to wait a little longer to find out.

User avatar
Supporter ♡

#56: Post by Brewzologist »

With my PRO2, I steam preheat both the group head and basket which helps with temperature stability. I wonder if the shower screen and PF are acting as even more of a heat sink on the 58, and if the heater elements are higher in the group head it may be more challenging to get heat down to them? Agreed we'll have to wait for more carefully controlled temperature measurement tests to find out.


#57: Post by vit »

I doubt that PF had much influence in this measurement, as it's not in direct contact with the cylinder. Similar to other machines, rim of the basket is pressing onto the seal that is not a good heat conductor and here it is in the metal disc that is at the bottom of the cylinder (is it a separate piece or part of the cylinder is unknown ATM), so heat transfer from the cylinder through the seal to basket and portafilter shouldn't be very fast. As I can see from videos, PF locks onto the stand so it's not in direct contact with cylinder either. However, that metal disc is sitting on the stand so it is transferring the heat from the cylinder to the stand through the part of the surface. Higher temperature loss on the bottom of the cylinder might explain why the water temperature on the bottom is considerably lower than on the top

I don't believe that shower screen is much heavier than in previous models - its weight is probably 1-2% of the mass of the cylinder - so its influence should be similar like in previous models (small)

There is actually no need to make "intrusive" measurements of the brew water by drilling the elements of the machine. Putting the probe onto the center of the basket bottom (from below) would give enough information about temperature stability - on Classic it's something like 2°C lower than temperature in the cylinder towards the end of extraction and not that much different in other machines (although it depends on the flow and other things)

Espresso Vision: the perfect cup of coffee starts with understanding your roast
Sponsored by Espresso Vision
User avatar

#58: Post by instantkamera » replying to vit »

This has been covered elsewhere, but Xris' readings are low because blank baskets on a bottomless portafilter is a huge exposed surface that is easy to sink heat away from through ambient air. The top of the cylinder would have shown much higher temps.

I agree on measuring stream temp, and in that regard I was never able to get above ~90c on the pro, whereas the 58 stream is above 90c and peeks at ~93. Temp is but one aspect of extraction improvement; another being the relative grind fineness. Pro would have to be significantly down-dosed to match, in my experience. The ability to change baskets is obviously key here too.


#59: Post by vit »

Blank basket would explain big heat loss if there was no shower screen. However - at least on Classic, if you plug the spout of empty portafilter, put the shower screen on it, mount the cylinder and pour the water, it will not fill the basket. There will be only a few millimeters of water at the bottom, while air from the basket won't go up through the small holes in the shower screen. My assumption was that the same would be on Flair 58, and since there is no direct contact of water above the shower screen and much colder water in the blank basket, there won't be much heat exchange either

However, looking again at his video, it seems that, unlike in Classic, lots of bubbles is coming from below, which might be a sign that the basket was actually filled with water and shower screen came in direct contact with much colder water, so that measurement was quite different from real conditions while pulling the espresso, hence useless

As I understand, you actually tested Flair 58 and measured stream temperature up to 93°C, which is similar to my measurements on Classic (see previous page). In that case, everything is fine ...

User avatar

#60: Post by instantkamera » replying to vit »

I can tell you that with my classic and the silicone edged screens I was often able to "pour through" and break the negative pressure that keeps the air trapped between the puck and the screen. That was before I started pushing the screen down to be a "contact screen", which is how I've used it since ~2018 probably.

I saw the same thing with my 58 and shower screen; sometimes it traps air, but not always. I actually entirely removed it (the pressfit shower screen) though, so I for sure filled the blank basket when I did the testing. I prefer this config with the mesh puck screen for cleanup as well as puck integrity with the Lagom + unimodals.