Pressure Profiling Techniques for Spring Levers

A haven dedicated to manual espresso machine aficionados.
User avatar
dominico
Team HB
Posts: 1652
Joined: November 8th, 2014

Postby dominico » Feb 10, 2016, 2:50 am

The classic "lever profile" is well recognized as a desirable profile for a wide range of coffees and is loosely defined as:
  1. a low pressure preinfusion for some amount of time
  2. a ramp to 9 bar pressure (or somewhere close)
  3. a slowly consistently declining pressure throughout the rest of the shot.

Beyond this all the specifics are pretty much variable according to the machine's operating mechanisms and the barista's preferred techniques.

I'll lay out some aspects for discussion about pressure profiling in general and then describe some techniques I have been using / experimenting with to play around with different profiles. I am not claiming anything here is undeniable truth, this article is what I have gathered as a combination of things I have read, artisanal data (read: anecdotal evidence), and my own experiments and observations.

Pressure profiling observable effects:
As already described in Pressure profiling, flow profiling, and a new rule of thirds a fast pressure ramp will slow down the flow during extraction, a slow pressure ramp will cause the extraction to flow faster. This is easily observable with any machine that allows control of the initial pressure ramp, such as levers. Essentially performing a slow pressure ramp allows you to grind finer.
Pressure and flow are related: Increase the pressure and you will see an increase in flow. Drop the pressure and you will see a decrease in flow, or at least a decrease in the acceleration of the flow.

Pressure Profiling Proposed Effects (by machine manufacturers at least)
These proposed effects come from two main sources: the Slayer Instruction Manual and a paper I have called Variable Pressure Infusion Modification Results: Joint project between Mark Barnett, Synesso, and Herkimer Coffee. The Slayer effects are detailed here assuming a relationship between flow and pressure within the brew chamber, so take that as you may.
Long, soft pressure preinfusion: proposed by Slayer to increase sweetness. Synesso claims a soft preinfusion for about 4 seconds before moving to a mid range preinfusion creates the best acidity balance, but having a soft preinfusion for too long lends to an ashy taste due to more phenol extraction.
Declining pressure through the extraction: proposed by Slayer to mute acidity. Observed elsewhere to minimize bitterness as well. I find it counter-intuitive that the same declining pressure profile would both mute unwanted acidity and unwanted bitterness, but I have noted that it certainly serves to improve the flavor of the shot.
One-second initial pressure spike before dropping back down to soft preinfusion: Proposed by Slayer to enhance acidity.
Slow ramp preinfusion combined with pressure decline during extraction: Proposed by Synesso to provide the best increased texture / body and best balance between sweetness and clarity. However too slow of a ramp, according to Synesso, makes a shot where the sweetness overpowers everything else and clarity suffers.

Techniques for lever pressure profiling:
Lever preinfusion is rather flexible in that you can preinfuse as short or long as you want. You can also choose whether or not to quickly ramp to full pressure or to slowly ramp to full pressure, and then how slowly this ramp occurs.

Preinfusion:
Standard Preinfusion: Pull the lever down and preinfuse for the desired time. My personal experience is that darker coffees desire a shorter preinfusion and lighter coffees desire a longer one. I have noted though that too long of a preinfusion though can cause coffee to lose some distinctiveness. Depending on your setup you may have control over this phase. Does your group feed from boiler pressure? Does it feed from line pressure? What sort of control do you have over that? These variables might be worth playing with.
Slow ramp vs fast ramp: How quickly do you lift the lever back up? This determines the speed of the pressure ramp. A quick lift with cause a quick pressure ramp, a slow controlled lift of the lever will cause a slower pressure ramp. Observe how this affects the coffee under the same grind and dose.
Poor man's 2 stage preinfusion: The Synesso Hydra is known for its "2 stage preinfusion" concept, where there is a low pressure stage 1 preinfusion followed by a mid pressure stage 2 preinfusion before going to full pressure. This can technically be done on a lever as well. While lifting the lever stop at some point before it catches and hold it there for a few seconds. What you are really doing is another preinfusion at a pressure that is higher than your machine's standard preinfusion but lower than the full pressure of the spring. This has the very noticeable affect of increasing the body of the shot, especially for coffees with a notoriously weaker body. I have noted that for some coffees this increases the sweetness of the coffee as well. The disadvantage here is that unless you have a portafilter pressure gauge that works with a lever (i.e. needle valve to simulate water flow) you are kind of flying blind as to what this pressure really is. The advantage here however is that it is relatively repeatable. It is not difficult to hold the lever at the same angle once you find an angle that works well for the "stage 2" preinfusion.


During the shot:
Pull back: While the lever normally has a gradually declining pressure profile during extraction due to the spring, you can reduce the pressure through the shot by pulling back on the lever. This will slow down the shot if it is flowing too fast or if you want to increase the time of the extraction.
Push through: On some levers there is a direct connection between the lever and the piston, this allows you to increase the pressure beyond the spring's capacity during the extraction by pushing up on the lever. I have found that this is almost never desirable and leads to ashy tasting espresso with more developed roasts. It does however allow you to recover from a choker by helping it along at least until a stream is established. I will do it if the grind is just slightly too fine to help the flow along, then once there is a constant trickle let the spring finish what it was doing.

Lately I have been playing with a "two stage" preinfusion, inspired by Synesso, which is usually composed of a low pressure "stage 1" and a higher pressure "stage 2" before finally ramping to full pressure. This usually involves pulling the lever down and counting to 5 or 8, lifting the lever up slowly between 5 to 8 seconds to then hold it steady at about 20 degrees above horizontal. I then either hold that for 5 to 8 seconds, or until a gram of coffee has dropped into the cup, depending on how it is pouring, then I let it all the way up and let the spring take over from there. This is similar to Synesso's suggested profile.
Synesso's suggested profile
  1. Stage 1 preinfusion: 4 seconds at 3 bar
  2. Stage 2 preinfusion: 3 seconds at 6.2 bar
  3. Full pressure (8.5 bar for them) for 8 to 11 seconds
  4. Drop-back to 6 bar for 3 to 5 seconds (simulation of a spring lever profile)

My current favorite profiles are:
  • Traditional espresso blends: 5 second preinfusion then quick ramp, pull back if the flow starts going a little too quickly.
  • Medium roasted SO's: 15 second standard preinfusion then quick ramp, pull back if the flow starts going a little too quickly.
  • Light roasted SO's: the Synesso Hydra style profile as described above - 5 second standard preinfusion, slow ramp to about a 20 degree angle above horizontal, holding for 5 to 8 seconds, then full pressure.

NB: I vary temp according to what works best for the coffee as well.
My typical dial in procedure goes in this order:
  1. Find the right dose / grind and use my best guesstimate of pressure profile based on the coffee and roast level.
  2. Adjust the temperature to balance acidity / bitterness
  3. Adjust the profile to attempt to bring out desired traits of the coffee.

What pressure profiling techniques do you use on your levers? Do you use different profiles for different coffees? How do you ensure your profiles are repeatable?
http://bit.ly/29dgjDW
Il caffè è un piacere, se non è buono che piacere è?

samuellaw178
Posts: 1797
Joined: April 10th, 2011

Postby samuellaw178 » Feb 10, 2016, 7:52 am

Thanks for putting these together for us spring lever users!

As you mentioned, the charm of a spring lever is its simplicity that works best for most coffee. So most of the time I'm using the 'in-built pressure profile'. Currently, I'm set to 2.4-2.7 bar preinfusion (flojet-line pressure) and the max pressure is somewhere 7.5-8.5 bar depending on how much air is in the system.

Here're the two profiles that I commonly use:

Standard - 2.7bar for 3-10 sec, followed by a gentle release of the lever.

Assisted push - similar preinfusion, followed by gentle pushing on the lever. I do this for the extra boost in body and it is noticable. If you overdo the push indeed it will impact the shot negatively.


One thing that really helps is getting a needle valve with pressure gauge(as you mentioned). I use it for 'practice' to gauge how much push is needed to achieve 9 bar (only a little needed, much less than I expected), and obviously use it to check my brew pressure. The 2-stage preinfusion is a new concept to me, and I tried some simulation just now using the gauge. Seems like I should be able to do 2.7 bar, followed by 4 bar (easy and consistent target) and then up to 8 bar.

One thing when you're playing with brew pressure on lever is you need to be aware of the air in the system. More air = less pressure from spring. ie - if you have very fine grind, coupled with short preinfusion, you will have more 'air' and that will impact your max brew pressure. It's worth paying attention to what angle the lever catches.

I'm not sure how important it is to fully saturate the coffee puck (indicated by appearance of coffee beads using a naked pf), but I usually preinfuse until I see beads forming (not dripping). Do you have any tips/opinion on this?

User avatar
dominico
Team HB
Posts: 1652
Joined: November 8th, 2014

Postby dominico » Feb 10, 2016, 10:24 am

The air in the group and how it affects pressure is a really good point. I have noticed that shorter preinfusions leave more air in the group and consequently the lever catches higher. Often when doing a short preinfusion for a darker roast I won't see any beads, and the lower "catch" pressure probably helps curb bitterness.

As far as whether there should or shouldn't be beads before ramping to full pressure, I wouldn't say it is necessary, plenty of my shots don't. What I will say is that if the bottom start to evenly bead across the puck during preinfusion my pours tend to look better.
http://bit.ly/29dgjDW
Il caffè è un piacere, se non è buono che piacere è?

User avatar
drgary
Team HB
Posts: 9880
Joined: February 7th, 2010

Postby drgary » Feb 10, 2016, 11:37 am

Dominick,

Your introduction is so rich with information I would love to take a few days off and experiment with my commercial lever. Oh, if I had the time right now! For the moment I can only comment on a couple of points. If you force the lever to increase pressure past what the spring produces, there are risks. With my Conti Prestina, doing this caused the upper piston gasket to yield. It didn't fail as I learned after inspecting it. Coffee was forced through the dipper tube back into the boiler and I had to do a thorough clean and rinse with coffee detergent. The Prestina is also small enough that pushing the lever can tip it backwards. That may also occur with models like the Bezzera Strega, although owners of that machine can comment on that with more authority.

About this thought:

dominico wrote:Declining pressure through the extraction: proposed by Slayer to mute acidity. Observed elsewhere to minimize bitterness as well. I find it counter-intuitive that the same declining pressure profile would both mute unwanted acidity and unwanted bitterness, but I have noted that it certainly serves to improve the flavor of the shot.

I believe that declining pressure avoids overextraction.

Dominick simultaneously posted an interesting thread on temperature management. That thread can be followed here:

Commercial Lever Temperature Management Techniques
Gary
LMWDP#308

What I WOULD do for a good cup of coffee!

EspressoForge
Posts: 1334
Joined: July 1st, 2008

Postby EspressoForge » Feb 10, 2016, 1:15 pm

dominico wrote:Declining pressure through the extraction: proposed by Slayer to mute acidity. Observed elsewhere to minimize bitterness as well. I find it counter-intuitive that the same declining pressure profile would both mute unwanted acidity and unwanted bitterness, but I have noted that it certainly serves to improve the flavor of the shot.

drgary wrote:About this thought:

I believe that declining pressure avoids overextraction.


Great info and ideas here. Like Gary stated above, I believe the same thing. But to expound, we already know in a simplistic way, there are 3 main "areas" of an extraction over time. Beginning extractions are acidity, middle sweet/caramel, end is bitters (some early bitters I think are good such as oak, spice and vanilla).

Simply put from my experience with profiles, you should increase pressure where you want those flavors to be increased (better extraction at higher pressures). This is why the simple profile of 1 to 3 bar starting, ramp up to high pressure (say 8-10 bar) then trail off or even decrease more rapidly...works well for many light roasts.

dominico wrote:What pressure profiling techniques do you use on your levers? Do you use different profiles for different coffees? How do you ensure your profiles are repeatable?


I simply modify a profile for each coffee (after tasting it on my standard profile):

Got too much bitter or bad ones? Trail off the pressure in the later part of the shot faster.
Too much acidity? Lower pressure at start of shot, or increase this time
Want more caramels (if they exist in the bean)? Increase pressure in the middle (even up to 12 bar works for me for some coffees if other things are done correctly).

dominico wrote:Pressure and flow are related: Increase the pressure and you will see an increase in flow. Drop the pressure and you will see a decrease in flow, or at least a decrease in the acceleration of the flow.


This is sort of true on flow rate, but only in the last 2/3rd of the shot. From my experience and from what I've gathered from the pressure profiling thread, they are related, but not equivalent. The beginning 1/3rd of the shot determines flow rate for most of the shot. You can influence flow by pressure, but it's really more a function of the pressure speed increase, or the rate of pressure change. If you get pressure up to 8 bar very quickly, for the same grind you'll notice flow will drop. But get to 8 bar over say 3-5s and flow will be quite good.

You have stated this already:
dominico wrote:Essentially performing a slow pressure ramp allows you to grind finer.


Which is exactly correct, but just wanted to point out where it applies in the curve. I think some graphs would help to understand, as really the visual curve would tell you a lot. Steeper slopes in the beginning make a bigger difference than the actual pressure maximums for flow rate.

dominico wrote:Preinfusion:
Standard Preinfusion: Pull the lever down and preinfuse for the desired time. My personal experience is that darker coffees desire a shorter preinfusion and lighter coffees desire a longer one. I have noted though that too long of a preinfusion though can cause coffee to lose some distinctiveness. Depending on your setup you may have control over this phase. Does your group feed from boiler pressure? Does it feed from line pressure? What sort of control do you have over that? These variables might be worth playing with.
Slow ramp vs fast ramp: How quickly do you lift the lever? This determines the speed of the pressure ramp. A quick lift with cause a quick pressure ramp, a slow controlled lift of the lever will cause a slower pressure ramp. Observe how this affects the coffee under the same grind and dose.
Poor man's 2 stage preinfusion: The Synesso Hydra is known for its "2 stage preinfusion" concept, where there is a low pressure stage 1 preinfusion followed by a mid pressure stage 2 preinfusion before going to full pressure. This can technically be done on a lever as well. While lifting the lever stop at some point before it catches and hold it there for a few seconds. What you are really doing is another preinfusion at a pressure that is higher than your machine's standard preinfusion but lower than the full pressure of the spring. This has the very noticeable affect of increasing the body of the shot, especially for coffees with a notoriously weaker body. I have noted that for some coffees this increases the sweetness of the coffee as well. The disadvantage here is that unless you have a portafilter pressure gauge that works with a lever (i.e. needle valve to simulate water flow) you are kind of flying blind as to what this pressure really is. The advantage here however is that it is relatively repeatable. It is not difficult to hold the lever at the same angle once you find an angle that works well for the "stage 2" preinfusion.

I fully agree with the standard pre-infusion.

Slow vs. Fast. This is good info for most spring levers, but I would caution it on the pump Strega. For me this never worked well just because the pump essentially builds pressure very quickly once it starts. The dimmer mod to pump is really necessary to do real pressure profiles on the Strega I believe. But that is just my experience, maybe someone else has figured it out without that mod.

2-stage. I haven't played much with this, I'll have to check it out and report back. I have tried a very long ramp up to full pressure from pre-infusion, with a much less steep curve. This may approximate this profile, but will try it out. It sounds quite interesting.

dominico wrote:Synesso's suggested profile
  1. Stage 1 preinfusion: 4 seconds at 3 bar
  2. Stage 2 preinfusion: 3 seconds at 6.2 bar
  3. Full pressure (8.5 bar for them) for 8 to 11 seconds
  4. Drop-back to 6 bar for 3 to 5 seconds (simulation of a spring lever profile)


Will check this out too, my "curve" generally for very light roasts would have this decreased in the stage 1-2 to around 1 bar for stage 1, and 4 bar for stage 2. I suspect this is heavily coffee dependent, but will try and see what happens.

On my new lever build, I'm hoping to try to apply these same techniques. So much appreciated in articulating them. I also have been thinking about a mod to get a brew pressure gauge inside, but I really don't want to drill my brand new group! I feel that a pressure gauge is fairly critical to repeating a profile, because especially with a lever mechanical advantage, very small pressure on the lever can change the internal pressure quite a lot, which Sam said well here:
samuellaw178 wrote:One thing that really helps is getting a needle valve with pressure gauge(as you mentioned). I use it for 'practice' to gauge how much push is needed to achieve 9 bar (only a little needed, much less than I expected), and obviously use it to check my brew pressure. The 2-stage preinfusion is a new concept to me, and I tried some simulation just now using the gauge. Seems like I should be able to do 2.7 bar, followed by 4 bar (easy and consistent target) and then up to 8 bar.



dominico wrote:NB: I vary temp according to what works best for the coffee as well.
My typical dial in goes in this order:
  1. Find the right dose / grind and use my best guestimate of pressure profile based on the coffee and roast level.
  2. Adjust the temperature to balance acidity / bitterness
  3. Adjust the profile to attempt to bring out desired traits of the coffee.


I think this is a good order as well, grind/dose is always the start, but keeping in mind the pressure profile does influence my choice there. Temp adjustment is the coarse acid/bitter balance, and pressure profile is the fine adjustment. I think you said the same thing, but just another way to look at it.

User avatar
drgary
Team HB
Posts: 9880
Joined: February 7th, 2010

Postby drgary » Feb 10, 2016, 1:29 pm

Off topic for a moment: If you know the temperature performance of your machine pretty well, you can arrive at a good starting temp via pourover brewing. If carefully assessing them I like to dial in coffees by brewing first, then making them as espresso. Okay, back to topic.
Gary
LMWDP#308

What I WOULD do for a good cup of coffee!

User avatar
another_jim
Team HB
Posts: 10295
Joined: May 5th, 2005

Postby another_jim » Feb 10, 2016, 3:06 pm

A couple of notes:

-- A declining pressure profile == a more constant flow, since the resistance of the puck decreases as the fines migrate into the cup. (IMO: even flow improves extraction, and the bulk of pressure profiling benefits late in the shot amounts to this)

-- Air in the lever group: Astoria derived groups have a relief valve (except the Strega's, which uses the cavity for the heater), many others do not. With longer PI times, more air escapes through the puck. Figuring the amount of air and water in the group for a particular shot, prior to lifting the lever, is often complicated. But if the lever rises too high, indicating excessive air in the group; you can increase the water amount by resetting the lever down again, waiting, and letting the air bleed through the puck.

-- Changing maximal preinfusion pressure: The Strega has a pump, so it's easy to get up to any PI level you like. For sippers or mains/HX groups, the maximal nominal pressure is 1 or 3 bars ... but, repeated pumps of the spring lever (similar to the Fellini move) can raise the PI pressure (Vlad has done this with pressure monitoring gear in place on an old Faema, a conventional Astoria, and various home groups, YMMV)

pizzaman383
Posts: 707
Joined: May 14th, 2011

Postby pizzaman383 » Feb 10, 2016, 8:46 pm

If you measure the commercial lever group temperature (especially near the location where incoming water flows) you'll see that the length of preinfusion time can have a big (2-5 degree F) difference in group head temperature. This means that while you may be adjusting the pressure during the preinfusion the difference in temperature could make a significant impact on the shot.
Curtis
LMWDP #551

day
Posts: 1252
Joined: December 24th, 2014

Postby day » Feb 10, 2016, 9:32 pm

Been playing with these concepts as well for a little while but I think you have a nice well though out well laid out post that has helped me to put together some thoughts. Testing some it out specifically right now.
Yes, i you per this on an iPhone

User avatar
nickw
Posts: 519
Joined: July 15th, 2013

Postby nickw » Feb 10, 2016, 10:28 pm

I agree with much of what you're saying.

I still find people confuse pressure and flow profiling. We need to think about what the coffee puck sees:

- 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.

The slower you expose the puck to water, and the slower you saturate it, the finer you can go.

Thats where the slow flow rate of a Slayer, along with the slow rising pressure allows for both slow and full saturation. Thus a very fine grind.

* Max pressure achieved during PI depends on: pre-brew flow rate, set machine pressure, coffee used, your grinder, degree of fineness ground etc..

** Max pressure depends on lever type: Boiler level, or plumb-in/line-level.
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.

 
Sponsored by intelligentsiacoffee.com
www.intelligentsiacoffee.com: home of the black cat project