LevaMayBe wrote:Thanks to all for the great discussion and facts. Some of these points I knew, and some were new to me. Poison's comments were right on the nose for me. I am a hybrid. I like the experience, instinct, reliability, and feel, but I also want the data. I have tried both a grouphead thermometer and a flow control valve and manometer set up on my HX machine. That was why I tried the Decent and really did like it, although I would not be too addicted to the graphs. I think on balance the ACS is calling to me. Love the Londinium, Brian Quan has a good video on it, but I really value the data and control that the ACS offers, the ability to stop the shot, the LSM group, and the faster heat up time. Dave Corbey has some great videos of it and of ACS. I have to say its steam and hot water wands are stunning...though there may be technical objections to their length. I did look at the Nurri - fantastic - but I cannot stretch the budget that far. I look forward to more interaction with the group. Happy pulling!

ACS Vostok is way to go.

I've tried different preinfusion pressures on my manual lever machines and spring machines.I never tasted any difference at all. None.
Temperature and pressure do give a different taste.
And of course the beans.

daveyb wrote:Just for the record, I am not saying that the ability to adjust the pressure is a bad thing and has no effect on a shot and how it tastes. What I am saying, is that on the LR24 this is the only adjustment that you can make. Other owners of machines that allow you to control temperature of the group and twin boilers via Pid's report back that the ability to tweak these settings has a far larger effect on the shot. Surely that is just common sense?

It's not clear to me that having multiple temp controls of group and twin boilers via PID is a benefit or will help get more out of a bean vs the single control of PI pressure on the Compressa. What I see is more complexity, expense and potential points of failure. These controls on a lever may also negate one of the inherent characteristics of a lever - the naturally declining temperature profile.

I'm not saying a PID based machine can't make great espresso - clearly they can.

My goals have evolved over the past 10-15 years of making espresso. Now that I have a deeper understanding of the core principals of making espresso, my goal is to find the simplest machine and the simplest workflow that will reliably and consistently give me excellent shots for any kind of bean I might come across. And for me, now, that is the Compressa. I'm not aware of a simpler machine that does what it does, nor a more complex machine that does it better.

I'm not saying the Compressa should be somebody's first or second machine; perhaps everyone needs to go through this evolution on their own. I do see value in machines like the Decent early in the journey so you can internalize the impact all the various shot attributes have on the flavor of a shot. But, I suspect, that many will yearn for a simpler system once they gain this insight.

I guess the point I'm trying to make, and apparently not very well, is that there are many ways to skin a cat (as the saying goes), the cat in this case being presenting a dose of water to the puck with a sufficiently consistent temp and pressure profile to make excellent espresso, reliably. And, that in my view, the simpler the machine that can do this, the better.

I don't believe it is necessary to precisely measure and control the temperature at one or more points of the brew path in order to achieve a sufficiently consistent temp profile. At least during the use phase of a machines lifecycle. During its development, then I imagine lots of measurement and control points would be used to characterize the machine. But in my mind, the most elegant design would be able to remove all of that complexity and have the "analog" architecture of the machine achieve the same result.

I've not used a lever machine with a PID-controlled group temperature, so I can't speak from experience. But I expect it would either attempt to keep a constant (I.e. non-declining) brew temp, or re-create the declining brew temp occurring naturally with a traditional lever. Both of which seem kind of silly to me for those looking for a lever machine, but I admit I may be missing something here and am open to being educated.

As far as single vs double spring - I've not used a double spring machine, so again no first hand knowledge. As long as you have enough control of the brew temp and pressure in the proper range to adjust for different beans, and the machine architecture can deliver this consistently, then the machine can make excellent espresso. After that, for me anyway, it is a mater of refinement to eliminate complexity in the machine and the workflow.

I would love to see a comparison of a Compressa and a LM Leva X, with each machine operated by a competent barista, pulling shots of beans at a variety of roast levels, and then doing some blind a/b testing. If the far more expensive and complex LM consistently pulls better tasting shots, I'm willing to eat my words! And start saving for an upgrade

CoffeeMac wrote:I guess the point I'm trying to make, and apparently not very well, is that there are many ways to skin a cat (as the saying goes), the cat in this case being presenting a dose of water to the puck with a sufficiently consistent temp and pressure profile to make excellent espresso, reliably. And, that in my view, the simpler the machine that can do this, the better.

I don't believe it is necessary to precisely measure and control the temperature at one or more points of the brew path in order to achieve a sufficiently consistent temp profile. At least during the use phase of a machines lifecycle. During its development, then I imagine lots of measurement and control points would be used to characterize the machine. But in my mind, the most elegant design would be able to remove all of that complexity and have the "analog" architecture of the machine achieve the same result.

I've not used a lever machine with a PID-controlled group temperature, so I can't speak from experience. But I expect it would either attempt to keep a constant (I.e. non-declining) brew temp, or re-create the declining brew temp occurring naturally with a traditional lever. Both of which seem kind of silly to me for those looking for a lever machine, but I admit I may be missing something here and am open to being educated.

As far as single vs double spring - I've not used a double spring machine, so again no first hand knowledge. As long as you have enough control of the brew temp and pressure in the proper range to adjust for different beans, and the machine architecture can deliver this consistently, then the machine can make excellent espresso. After that, for me anyway, it is a mater of refinement to eliminate complexity in the machine and the workflow.

I would love to see a comparison of a Compressa and a LM Leva X, with each machine operated by a competent barista, pulling shots of beans at a variety of roast levels, and then doing some blind a/b testing. If the far more expensive and complex LM consistently pulls better tasting shots, I'm willing to eat my words! And start saving for an upgrade

Thanks for your reply. I want to start by saying that I respect your choice of the compressa and I am sure it makes great tasting espresso. What I am wanting to do is not to suggest that everyone should always choose a hybrid lever over a lr24 for example, but I want to explain the rationale for why a hybrid lever is in my view a very attractive proposition that should at least be seriously considered by anyone wanting more than what a trad lever offers especially if they use light roasts often.

I agree firstly with your point on simplicity. But what is simplicity? To me it is the easiest and most direct way and most measurable way (so that we know we have achieved it) to achieve a certain objective. We have a common objective which is to vary the temperature on command in order to make it more suitable for different roast levels and different beans. Isn't the easiest and most direct way and most measurable way to achieve this objective is to simply use a PID (a most common and robust hardware) and a thermocouple sensor to regulate the temperature of the grouphead, and boilers independently? But if we then adopt the methodology to vary preinfusion pressure in order to vary temperature in the grouphead, we are using a more difficult, indirect and not measurable way (because there is no temp gauge is installed in the machine) to vary the temperature. How would the approach of the lr24 be simpler then? With respect, it seems to be unnecessarily complicating the situation by varying a completely diff variable instead of the variable of temperature that is being targeted...

Secondly if you don't have the measurement of the actual temperature at any given time of the components used to brew the coffee, then how do u know that the objective is being achieved in your machine? If that measurement is only done on tbe prototype during the development phase, why should the results for that prototype during initial testing be necessarily applicable to every machine off the assembly line that sits at the homes of their users? And how would we know if there is no ability to measure the temp in every machine to give the user an idea of how much to change the pressure so as to change the temp to the level he wants?

I agree that the best approach is to have analogue presentation. That's what u get with the nurri leva - no app, no touchscreens. Just the pid and two paddles and one pump control dial, two.manometers, and the lever. But a wealth of abilities made thus available like the palate of a painter, subject only to his imagination. The workflow is a pleasure to use.

IMHO the main point of the pid controlled grouphead is not to choose between declining profile or constant temp, although both options are obviously available (you could.even do ascending remp profile if you wan with a hybrid lever...). The primary objective is to be able to set on command whatever temp is best for the beans that u r using. That ability is what the hybrid levers primarily offer and it depends on cartridge heaters and PIDs - hardly cutting edge technologies which are robust and have been used for a long time.

No need to go all the way to a LEVA X's prodigious cost. The hybrid levers like the Nurri leva aren't much more expensive than the Londinium and they offer the capabilities that I am mentioning above.