Pressure Profiling Techniques for Spring Levers - Page 5

A haven dedicated to manual espresso machine aficionados.
bakafish
Posts: 629
Joined: 11 years ago

#41: Post by bakafish »

dominico wrote:I have the exact same group, how are you measuring the flow rate? and roughly how many "turns" from fully closed is that?
I put a cup and scale under the group. Tared it and pulled the lever for 10 seconds and weigh the water, not precise but useful. I don't know how many turns. I just pulled the lever and turned the valve to almost stop the water. This video clip shows about 10 ml a second.
nickw wrote:Cool you can play with the valve.

I noticed with boiler level PI there wasn't much difference, but it may also be grinder dependant.

You may also like getting some freshly roasted coffee. I find nitrogen flushed coffee (like Illy) does okay for the first day or so, then drops off fast.
I have no idea if it relates to the grinder and coffee. My grinder is Huie-71, a custom made Versalab M3 style electric grinder with Robur's 71mm burr set and adjustable 1-120 rpm. I set it at 45 rpm, much slower than any espresso grinder, because the taste is total different than it does at 120 rpm.

Today I used a new opened illy Monoarabica Ethiopia, VST 18g basket with 18g coffee beans. Again, with the 20 ml/sec (10 seconds wetting / preinfusion) and 4.5 ml/sec (30 seconds wetting / preinfusion) flow rate and slowly lifted the lever for 3 seconds to its catch point, it pulled almost the same shots (I mean the extraction flow rate). Then I pulled shots with the 20 ml/sec and 4.5 ml/sec, lifted the lever right after the group was filled (of course the fill / wetting puck time is different), took 3 seconds to reach the catch point, and again, the 2 shots were the same. In the third batch tests I took just 1 second to reach the catch point, and the 2 shots were still the same, but they almost choked the machine at the beginning of extraction. That's the biggest difference than the 1st and 2nd batch tests.

I think that wetting the puck at 0 bar and preinfusion at 1 bar boiler pressure are similar. They affect the extraction flow rate not that much. The extraction flow rate is mainly affected by the extraction pressure ramp up speed. That reminds me the discussions of water debit many years ago. The water debit, i.e., the flow rate without the portafilter, affects the pressure ramp up speed on pump machines, but on spring lever machines, the pressure ramp up speed is affected by the time from lifting the lever to the catch point, not the water flow rate.

User avatar
vberch
Posts: 596
Joined: 14 years ago

#42: Post by vberch »

Great discussion, guys! Thanks Dominick for starting it!

User avatar
nickw
Posts: 559
Joined: 11 years ago

#43: Post by nickw »

bakafish wrote:I think that wetting the puck at 0 bar and preinfusion at 1 bar boiler pressure are similar. They affect the extraction flow rate not that much. The extraction flow rate is mainly affected by the extraction pressure ramp up speed. That reminds me the discussions of water debit many years ago. The water debit, i.e., the flow rate without the portafilter, affects the pressure ramp up speed on pump machines, but on spring lever machines, the pressure ramp up speed is affected by the time from lifting the lever to the catch point, not the water flow rate.
Yes, I got similar results with going from 0-1.2 bar, both quickly and slowly. I think because it's only able to push the water so deep into the puck. Using a find grind, lighter tamp, and quick PI gave me the best and most consistent results.



I am also still thinking:
Flow rate is based on saturation speed - Which for many is connected to ramp speed, but not necessarily.

As an example (which is just my own theory) I'm still thinking, whether you go from:
0 - 3 bar, with full saturation in 20 seconds,
will taste similar to,
0-6 bar with full saturation in 20 seconds.

Even though the speed/incline of the ramp is twice as much. It's the full saturation which matters, not the speed of the ramp.

Note: Please also forgive the fact - that in actuality - the ramp is logarithmic, not linear.

User avatar
radudanutco
Posts: 147
Joined: 11 years ago

#44: Post by radudanutco »

nickw wrote:...
- Slowly wetting the puck at 1.5ml a second on a Slayer, and letting the pressure slowly climb*, until
the pack saturates is very different than
- quickly filling a lever group (to a set pressure based on the lever**) and having that column of water push down on the coffee.
...
Hybrids like the Strega are variable and can go higher. Toggling the pump on/off, or modding with a dimmer allows for a lot of flexibilty.
yes, the stock Strega starts with a 5.5-6 sec of practically zero bar preinfusion, as the group fills at about 10 ml/sec; at the end of this wetting stage, the pressure ramps up slowly enough to stop it at any value, especially on the lower part (1-4 bar);

all these are part of the pump preinfusion, and they could motivate adding a pressure gauge on the pump output, as I did;
and if the pump is dimmed, the length of the "zero pressure stage" as I work with, is 12-15 secs;
then, the increase in pressure on ramp is much slower (the slope is controllable by the dimmer) and the pump could be stopped at any value of reached pressure;
or, at two different pressure values, making kind of a slow 'gravity' soaking, followed by a two stage PI on the pressure ramp (10s zero bar PI - 5s ramp to 3 bar - 2s pump again to ramp to 6 bar (or watever is the maximum pressure for the dimmed pump and some flow already initiated) - some pause at 6 bar, then on springs...)

the pressure gauge will give a valid indication as long as the water inlet in group is open;
this could be the pressure profile of a full ramp extraction with dimmer set to max and min (pump flowrate):


okmed wrote:Wouldn't the thermostatically controlled heaters in the group head on the Strega prevent the water temperature from cooling?
the group heaters does not have such a rapid effect with their 100W heating power;
IMO, they are not meant to react efficiently for group thermal recovery in those 40-70 secs of shot time (PI + extraction);
important also, they are not reacting to temperature changes in the active area of the group (lower cylinder); the thermostat and the resistors are far in the upper back part of the group neck, and they seems to be efficient in dissipating a constant amount of heat, which as nickw wrote, is important for the idle thermal management at 82-83,5C (180-182,3F) in pt. #1 as erics;

during the shot, the temperature profile of the group is a result af thermal contributions of the water and of the cylinder walls in the active area;
this is the temperature profile of the group wall, measured in pct. #1, as refered above (lower front of the cylinder) during a shot:



and here comes the motivation for a third mod for Strega: PID-ing the group temperature!
with the group kept cool enough, it is possible to work comfortably without flushes; and of course, to manage higher or lower extraction temperatures during spring action;

it seems that with a Strega with a full/dimmed pump, with a PI manometer and with a PID on group heaters there are endless possibilities to experiment different pressure profiles;
but consistently, only in preinfusion;

it would be interesting to know such profiles, pressure vs. time, for Slayer, or any other slow PI pressure profiling machine; I do not expect to duplicate the results, but to see the particular Strega ones;
bakafish wrote:...I think that wetting the puck at 0 bar and preinfusion at 1 bar boiler pressure are similar. They affect the extraction flow rate not that much.
... but on spring lever machines, the pressure ramp up speed is affected by the time from lifting the lever to the catch point, not the water flow rate.
aren't they different if the time is the same, hence different degree of saturation, or excessive preinfusion?

as pressure ramp slope, suppose I have 2 consecutive shots with identical (...) preinfusion "protocol";
now, if I am lifting slowly the lever, in, say, 4 seconds, I have a slower ramp, than in the case I am releasing the lever abruptly, in one second; and I have then different flow rates in extraction;
mostly if the drops are comming only at the very end of the preinfusion (saturation);

the pressure ramp with Strega is rising slower at the begining (up to 3-4 bar), then faster, that is, more like an exponential function; as with pump machines which have an increasing flow rate as the puck is losing more solubiles; in this case, from drops at the end of saturation, to full stripped cone flow

well, if a constant flowrate is desirable, one could retain the lever;

I am speculate that this (lever rising speed) has something to do with the fines: during wetting, they are already fixed between the larger particles, but if the ramp is an abrupt one, the water is disrupting some of them, increasing the thickness of the fines bed at the bottom of the basket;
of course, this is an ideal case, as a water pulse could cause some channeling too (?)

User avatar
dominico (original poster)
Team HB
Posts: 2006
Joined: 9 years ago

#45: Post by dominico (original poster) »

bakafish wrote:I put a cup and scale under the group. Tared it and pulled the lever for 10 seconds and weigh the water, not precise but useful. I don't know how many turns. I just pulled the lever and turned the valve to almost stop the water. This video clip shows about 10 ml a second.
The most enlightening part about this for me specifically is that I had no idea you could take that lower cap off with the machine on. For some reason I had thought you had to adjust it with the machine off. I tweaked the flow for both of my groups now and the flow from my shower screens have improved as well. Thanks!
https://bit.ly/3N1bhPR
Il caffè è un piacere, se non è buono che piacere è?

bakafish
Posts: 629
Joined: 11 years ago

#46: Post by bakafish »

dominico wrote:For some reason I had thought you had to adjust it with the machine off.
When my Lambro arrived, I found water leaked from this valve with the cap on during test. I disassembled this valve and took the brass cylinder
out,

and found the o-ring on it was broken.

After I replaced the o-ring and realized the valve mechanism, I knew it can be adjusted on the fly. :D

chris_n
Posts: 389
Joined: 11 years ago

#47: Post by chris_n »

EspressoForge wrote:This quote is from the Open Source Lever thread, again not too sure how adjustable it is, but if it can be in fact totally closed when all the way down, I bet you could crack it 1/2 to 1 full turn and get a very restricted flow. Once I can find the thread specs or figure them out and connect it to my line pressure I'll do some measurements.

I know we've been going off on flow restricting/profiling topic, but I think they are inter-related enough to make sense in this thread.
anyone know if i can fit something like this to the group bolt on a kvdw mirage idro?


User avatar
TomC
Team HB
Posts: 10535
Joined: 13 years ago

#48: Post by TomC »

chris_n wrote:anyone know if i can fit something like this to the group bolt on a kvdw mirage idro?

<image>

Please start a separate thread for such an issue. This thread should stay on topic.
Join us and support Artisan Roasting Software=https://artisan-scope.org/donate/

User avatar
dominico (original poster)
Team HB
Posts: 2006
Joined: 9 years ago

#49: Post by dominico (original poster) »

So I assembled one of these over the weekend:




For my particular spring, assuming no air pockets in the chamber, it maxes out at 8 bar;
holding the lever at about 30 degrees above horizontal gives me a 4 bar preinfusion.

I'll post a video later.
https://bit.ly/3N1bhPR
Il caffè è un piacere, se non è buono che piacere è?

User avatar
vberch
Posts: 596
Joined: 14 years ago

#50: Post by vberch »

Very cool, Dominick! Good job!
dominico wrote:So I assembled one of these over the weekend:
<image>


For my particular spring, assuming no air pockets in the chamber, it maxes out at 8 bar;
holding the lever at about 30 degrees above horizontal gives me a 4 bar preinfusion.

I'll post a video later.