The E61 Group - Truth and Lies - Page 3

Need help with equipment usage or want to share your latest discovery?
User avatar

#21: Post by HB »

Strugs wrote:I do agree that the water in the HX is much higher than proper brew temps, but it would also appear that the group itself is overheated by this idle water as well.
Depending on how and where you measure, you will get a different result:

From E61 Group Espresso Machine: Detailed Interior Schematics and

I think we can agree that the "red zone" initial temperature varies from machine-to-machine. Specifically, of the espresso machines I've tested, the Expobar Lever requires the longest cooling flush, followed by the Giotto Premium. The Andreja Premium, Isomac Rituale, and La Valentina behave similarly with 4-6 ounce flushes. From what I've read, the Wega Lyra's flush requirement is in the Expobar's range, which may explain part of the reason behind your higher reading.

In the end though, neither a "cooling flush" or "warming flush" accurately describe what's happening. Lately I've written simply "flush" to avoid assigning it a thermal tendency. Now that I think about it, a temperature equalization flush best captures the purpose of the step.

On a related note, looking at the cross-section of a saturated group, it's easy to see how it would reach uniform equalization more readily than an E61. Whether that matters in terms of better espresso is an entirely new discussion.

"Gooseneck" extends the grouphead away from the boiler to the service area
Dan Kehn


#22: Post by k7qz »

Wow, I had to read Sean's C.Geek post twice to make sure I followed it! :shock:
HB wrote:In the end though, neither a "cooling flush" or "warming flush" accurately describe what's happening. Lately I've written simply "flush" to avoid assigning it a thermal tendency. Now that I think about it, a temperature equalization flush best captures the purpose of the step.
Or perhaps 'thermal optimization flush'? As I (currently) understand the thermodynamics expressed here, simply speaking what we're trying to accomplish is to make the blue areas on your thermal CAD diagram (Dan where do you get all these cool pictures?) yellow green or about 94 C to optimize the shot and that the 'technique' required to accomplish this may vary from machine to machine.

OK guys, tell me if I'm off base here... :lol:


User avatar
malachi (original poster)

#23: Post by malachi (original poster) »

It really depends on the specific machine.

For example, the thermosyphon setup can DRAMATICALLY alter the temp of the group. I've measured idle group temps on machines with small diameter, equal size pipe and seen noticeably low group temp and have measured on one machine that had very large diameter uneven sized pipe and seen noticeably high temp. I would suspect that this is why you see the "tuning" restrictors on machines like the Faema.

So it's not just where you measure - it's also the machine itself. Where this gets really complicated is when you get out of the "true" E61 implementation and get into the variants and modifications.
"Taste is the only morality." -- John Ruskin


#24: Post by whodat »

malachi wrote:Some other misunderstandings...

- Having an E61 style group does not guarantee that your machine offers pre-infusion.
Coming from a couple of lever machines where pre-infusion clearly makes a difference, this is an interesting statement.

How would one know if a particular E-61 machine had pre-infusion? Or is the misunderstanding the notion that there is no guarantee?

User avatar
Team HB

#25: Post by another_jim »

whodat wrote:Hmm,
How would one know if a particular E-61 machine had pre-infusion? Or is the misunderstanding the notion that there is no guarantee?
Chris was being ultra-terse here ...

The preinfusion on a lever is based on the water sitting on the puck pressurized at one bar. This is only available on levetta E-61s with rotary pumps, where the line pressure does the same when the lever is lifted half way (opening the brew path but not starting the pump). How much this affects the taste on E61s is debatable. None of the manufacturers emphasise this particular trick; it seems to have gone out of style by the end of the 60s.

The "designed in" preinfusion on the manual (levetta groups) is a spring loaded piston in the lower section of the group which gets compressed as the water enters the group under pump pressure. The effect is that the puck sees the water build up pressure from 0 to 9 bar (or final brew-pressure) in around 10 seconds. The coffee starts to flow when the pressure is close to max.

In the semi-auto "e-61ish" groups, which have a regular solenoid and no preinfusion piston, the same visual effect is achieved by having gicleur (flow restrictor) in the group and an empty space that fills up with water prior to the brewing beginning. Whether this leads to the same water behaviour over the puck is anyone's guess.

Any group whatsoever can be adjusted for "dwell-time" (the time from turning on the pump to seeing the first drops) by varying the size of the gicleur or putting an adjustable needle valve in the pump line. Again, how far this qualifies as "true pre-infusion" is anyone's guess.

My take, unless I get taste evidence to the contrary, is that dwell-time is the factor of interest, and that any method for getting it comes to the same effect. In a side by side of various E-61ish machines 2 years ago in Milwaukee; none of the participants could detect a systematic difference between manual and semi-auto e-61 machines. There were obvious differences between them and the other machines at the side by side.

In other words, the term "true-pre-infusion" is truly meaningless.


#26: Post by mteahan »

Not meaningless. Even vibe E61's have true pre-infusion just like rotary machines, with the PI pressure set by the spring preload of the lower chamber, not by line pressure. Because the spring acts like a pressure regulator, it's PI effect is constant until the chamber is full, just like a lever machine. Different springs rate mean different PI rates and pressures.

As far as what 'true pre-infusion' is, it's a matter of how the extraction process is handled. If PI wasn't an issue, why did Marzocco patent a saturated version of their group with pre-infusion back in the 90's. In part, it was because Starbucks was roaming Italy looking for a solution to tossing 2000 pounds of coffee a day at a shift change. It probably was the rational behind developing the Swift grinder--though I can only speculate.

As far as temp stability of E61 groups (or any for that matter) against a saturated group, you have to consider the heat transfer capability of a static water column versus brass. My inclination is that copper and brass are more efficient. Idling saturated groups run several degrees colder while non saturated group run several degrees hotter. Considering that temperature inside a boiler can vary from top to bottom, and stainless steel boilers exhibit even greater temperature differentials, the traditional system might be more stable.

Perhaps the reason Marzocco never implemented the saturated PI system was the same reason they DON'T want anyone to pre-flush before each shot; decreased temperature stability due to the extra demand placed on the element.

Just more to consider, I suppose.

Michael Teahan
analogue | coffee


#27: Post by mteahan »


The fundamental design and implementation of the e61 group is pretty much the same for all machines, with the sole rational given for original vs. non original being the weight of the group and its inherent ability to retain and disperse temperature from the exchanger. Most modern groups are 4.2 kg vs. 5 kg for the original.

Which a couple of pounds more than a Marzocco. At least one without water.
Michael Teahan
analogue | coffee


#28: Post by poison »

A little OT here, but you say stainless boilers are less temp stable than copper/brass? Are you saying stainless is an inferior material to use for a boiler?


#29: Post by mteahan »

If you are relying upon the SS to transfer heat energy, then yes. If water is moving through the system, then water is the primary conduit using flow as the mechanism and the material is less important.

For example: the Espressa uses stainless steel for transfer pipes to the group and steam pipes, but a copper pipe to the hot water valve. Without the copper, the first 3 ounces of water would be warm at best--it is the copper that transfers heat energy to the valve to keep the water column hot. The water alone is not enough. The brew group on the early LM's were nickel plated brass as opposed to the new ones which are stainless (apparently) and are probably less stable than the original design.

LM steam boilers are less affected by the material because the distance between the surface of the water and the steam pipe is less than traditional machines. Stainless steel boilers in some machines I have worked with had perhaps a 20 drop from surface temp to boiler apex. Wet steam was the result. I think stainless boilers in the brew system probably are part of the reason idle temps drop on the LM after 30 minutes.

So yes, copper is better than stainless for espresso machines with the current state of technology. SS requires a new approach to replace the loss of characteristics associated with copper and brass. It can be worked around, but you have to actively look for solutions to address the problem.

And copper costs more than stainless, especially if you have to plate it. Why, then, would anyone do it unless it works better?

Michael Teahan
analogue | coffee


#30: Post by poison »

The reason I ask is because I've been eyeing the Astra Pro machine, which has a stainless boiler. It is reported to be very temp stable, and I'm not sure how it's stainless boiler plays into that. I understand it could just heat-cycle more often in a smaller range.