The E61 Group - Truth and Lies

[malachi]

Inspired by a conversation with Dan, I thought I would start a thread to allow folks to ask questions about this near-ubiquitious espresso group and to hopefully clear up some misperceptions about it.

Please - feel free to ask questions and to post your own clarifications and corrections.


I'll start with perhaps the most common misunderstanding of the E61 group.

The E61 does not "run hot". Many people take the phrase "cooling flush" to mean that the group is hotter than the target brew temp and must be cooled by running brew water through it. This is not the case. On HX machines, you do need to flush to cool your brew water - but what you are cooling is the heat exchanger, not the group head. The group, in fact, runs cool and the initial flush will actually serve to not only cool your brew water but also to heat your group.
On a dual boiler machine with an E61 group you would thus need to do a "warming flush" (as you would with an auto Marzocco Linea).

[barry]

not all e-61 heads are true e-61 heads. the patent has run out and there are several aftermarket and non-faema versions, some w/o pre-infusion, and, iirc, some that aren't even thermosyphons.

[malachi]

Good point!

At some point we should try and group the various E61 groups (original, variant, imitation) and categorize them.

[Teme]

I recall someone claiming that Faema and Vibiemme are the only ones with the "original" groups while all others are imitations. Can anyone confirm this? I'll try and find the comment and post a link...

Br,
Teme

[malachi]

I think it would be worthwhile to try and "classify" the various groups.

Just because a group is an "imitation" doesn't guarantee that it's inferior. I'm predicting that there are probably variants that are equal to or perhaps even superior to the "true" E61 while there are others that are inferior.

So rather than "true" versus "imitation" perhaps we should focus on classifying based on functionality and performance.

[HB]

I'm predicting that there are probably variants that are equal to or perhaps even superior to the "true" E61 while there are others that are inferior.

What does a "true" E61 mean anyway? For those who are joining late, the discussion E61 Group Espresso Machine: Is its reputation justified? included the original E61 patent diagram, enlarged below:


From patent 3,230,974, "Alternately Seating Valves"

Below is Lino's carefully CAD-recreated image measured directly from a modern version:


Courtesy of Verna Design Inc (please do not copy)

In terms of mechanisms, they look to be identical. However chamber #3, which contains the water that circulates from the boiler, is much larger in the original than the newer model. The patent doesn't mention the boiler design, so I can't say whether the espresso machines of that day were mostly HXs or had dedicated brew boilers. If it were circulating steam boiler water, no doubt the grouphead would have the potential to heat up much more, explaining why thermosyphon flow restrictors were standard issue for the Faema E61:


Flow restrictor of the Faema E61 Legend, courtesy Ninth Street Espresso

I recognize that many attribute nearly hallowed status to the E61, but ironically it's usually in lauding its thermal stability, which isn't even mentioned in the patent (read it above) versus a "solenoid type" E61 (i.e., without expansion chamber like La Valentina)? There is indeed a measureable difference, as noted in Pressure profiles, preinfusion and the forgiveness factor:


Comparisons of Expobar Brewtus (E61 "lever type"), Elektra A3 (rotary), and La Valentina (E61 "solenoid type")

In all fairness, I haven't dedicated time exclusively to discerning the "forgiveness factor" between the lever and solenoid type E61s. My practical usage suggests that if it exists, it isn't a critical distinguishing feature.

So rather than "true" versus "imitation" perhaps we should focus on classifying based on functionality and performance.

And finally I arrive at my point: Classifying based on performance will have to take into account the whole machine, not just whether it is a "genuine" or "imitation" E61. Speaking more generally, I believe many consumers and critics naturally want to explain why one espresso machine presumably out-performs another, and they look to design characteristics (e.g., temperature stability, preinfusion) that support what they observe and taste. While this intrigues the armchair engineers among us, the conversation needs to return to the practical matter of improving espresso, not just admiring hardware for the sake of admiration.

(My apologies if I rambled a bit, I was interrupted several times while writing this...)

[k7qz]

On HX machines, you do need to flush to cool your brew water - but what you are cooling is the heat exchanger, not the group head. The group, in fact, runs cool and the initial flush will actually serve to not only cool your brew water but also to heat your group.

Great thread!

Dan has been kind enough recently to answer my e-mail questions exactly along these lines. I'm using a Cimbali Junior D (for 2 or 3 months now) and have found that the "warm up" recommendations Dan posted in his 'Bench' article for this machine to be perfectly in accord with Malachi's statement above. (Recently verified with my shiny new Scace thermofilter! Very nice toy BTW!!)

OTOH, the Junior does seem to behave a little differently than the other HX machines that I have used (take this as an observation from a novice ), again probably due to variation in group design, as alluded to earlier in this thread.

From Dan's Junior article:

"To get a quick feel for the performance characteristics of a new machine, I do a temperature profile test nicknamed "drop off." It involves pulling a shot and continuing to run the pump until the grouphead water temperature falls off, providing a general idea of the HX size and temperature stability without the need to disassemble the machine. In Junior's case, the results were incredible: After the temperature leveled around 12 seconds into the shot, it varied little more than 0.5 degree for another forty-five seconds (the pump auto-shutoff stopped the test at one minute). No prosumer HX machine that I've tested is capable of such lengthy tail-end stability."

Perhaps this accounts for the "difference" or "feel" that I'm noticing between Junior and the other HX machines that I've used?

I searched for a cutaway diagram of the Cimbali group to compare to Lino's beautiful CAD image but couldn't find one. Thought it would be interesting to compare them side by side. Anybody have one for the Cimbali?

-Mike

[Teme]

I think it would be worthwhile to try and "classify" the various groups.

Just because a group is an "imitation" doesn't guarantee that it's inferior. I'm predicting that there are probably variants that are equal to or perhaps even superior to the "true" E61 while there are others that are inferior.

So rather than "true" versus "imitation" perhaps we should focus on classifying based on functionality and performance.

And finally I arrive at my point: Classifying based on performance will have to take into account the whole machine, not just whether it is a "genuine" or "imitation" E61. Speaking more generally, I believe many consumers and critics naturally want to explain why one espresso machine presumably out-performs another, and they look to design characteristics (e.g., temperature stability, preinfusion) that support what they observe and taste. While this intrigues the armchair engineers among us, the conversation needs to return to the practical matter of improving espresso, not just admiring hardware for the sake of admiration.

I agree that the performance is naturally key. I think it would be quite difficult to isolate the specific characteristics of the different E61 groups from the other variables / characteristics of individual machines - that is unless someone would happen to have access to several different E61 groupheads and be willing to try them all out on a single "reference" machine. Even this would potentially have problems, wouldn't it - if a particular machine is designed to suit that particular E61 group, the other groups might not be as well suited to another machine? One could of course go one step further and swap groupheads between different machines but I really do not think this is a realistic proposition...

With regards classifying the different E61 groups, this looks to me as being equally challenging. At least I have trouble separating facts from fiction when reading comments on this. Just as an example below are three quotes (I can add the names if you think that's appropriate):

Here's the story, as told to me by a Faema Toronto employee, where they also sell many other makes (new and used), including ECM. The authentic E61 is used only by Faema and ECM...they sell it to ECM, because the guy who owns ECM, is a former employee of faema (and obviously somehow in the good books with Faema while also being a competitor).

The Only Original E61 Faema design is in the VIBIEMME machine, which they own the patent. I heard the same story form them, but the design of the group ( look at the top part of it you'll see the difference) speaks for itself

Apparently, there are at least three different replicas or knockoffs of the original E61 head. The almost exact copies found on ECM and Isomac machines, as well as those with the angled shank found on Wegas and Grimacs are excellent, whereas the ones made by Ariete/Futurmat in Spain, and found on the earlier Domobars (like in the link I gave) are not very good. These are the ones with the angles on the "flying saucer" section. The newer Domobars may have moved to the Wega head.

Which one is true? They cannot all be true? Or is none of them true? And what exactly are the differences between the designs?

I guess someone could spend quite some time researching this and end up writing a book. In the end it might be feasible just to compare the merits of one whole machine against another - and even here I am leaning towards an opinion that the differences in the performance of E61 HX consumer machines are likely to be small (not taking into account the vibe vs rotary pumps or the Brewtus with its pid and dual boilers). I.e. the main differences are in some of the features they offer and/or the durability of their components (e.g. vacuum breakers, brew pressure gauges, electronics, internal layout etc). But I may be wrong

Br,
Teme

[HB]

Perhaps this accounts for the "difference" or "feel" that I'm noticing between Junior and the other HX machines that I've used?

I searched for a cutaway diagram of the Cimbali group to compare to Lino's beautiful CAD image but couldn't find one. Thought it would be interesting to compare them side by side. Anybody have one for the Cimbali?

The best I found was straight from the owner's manual:


Cimbali Junior hydraulics diagram

While my last post argues against using design diagrams to infer quality / superiority, it is instructive here to explain the behavior you noted: The long trunk connecting the grouphead and large HX likely contributes to Junior's lengthy "thermal memory." Compare this to your typical prosumer E61 HX:


E61 group hydraulics diagram

The red / blue arrows indicate the thermosyphon loop, composed of two copper tubes of varying length and diameters. Espresso machines like the Expobar Lever have a lengthier loop than say the Andreja Premium, and it shows dramatically in the flush amount. One "advantage" of this increased distance from boiler to grouphead in the second diagram compared to Junior's direct-convection design depicted in the first is the short reset time between the consequences of the flush. I say "advantage" in quotes because the shorter-term thermal memory helps non-professionals who may unknowlingly over/under flush from time-to-time. With your average prosumer machine, all is forgiven in a couple minutes. Junior will still be ticked off 10 minutes later.

On a related note, I was pleased with the Elektra A3 because it combined the best of these attributes: Very short term memory (i.e., forgives your mistakes), rock solid temperature stability (i.e., hitting and maintaining the same temperature shot-after-shot was a breeze). With all due respect to rightfully popular E61-based espresso machines, this experience prompted me to re-examine my potential bias towards such "flagship designs," and to adopt a more fully holistic approach to evaluations.

(my roundabout way of saying "buy based on results from actual use, not on specs")

[malachi]

The red / blue arrows indicate the thermosyphon loop, composed of two copper tubes of varying length and diameters.

This is something of great interest to me.
I've noticed that a ton, in fact possibly the majority, of home E61 machines have equal diameter tubes. This seems to be a serious (perhaps even fatal) flaw.

[k7qz]

The best I found was straight from the owner's manual

Owners manual, who reads the owners manual ? ....

I bet I spent 20 minutes doing an internet search for a diagram like that. Dan the man, what else can I say!

-Mike

[lennoncs]

The red / blue arrows indicate the thermosyphon loop, composed of two copper tubes of varying length and diameters.

This is something of great interest to me.
I've noticed that a ton, in fact possibly the majority, of home E61 machines have equal diameter tubes. This seems to be a serious (perhaps even fatal) flaw.

It takes time and money to tune a system properly, and that tuning will change with configuration.

I agree that it is a flaw of home E61 designs and most likely many commercial designs also; but when a manufacturer is trying to stay in business on low production volumes as seen in this market, it doesn't surprise me that they just slap in tubing that is "close" and send it to market.

Analog system design is not a common skill found in the modern industrial base anymore, with increased reliance on digital controls to counter basic design deficiencies I wouldn't count on seeing any elegant system design practices anytime soon.

The whole concept of digital control is not a bad thing either...we all love our 1/32 DIN PID controllers and many times It can take a marginal system and make it world-class with little time invested.


Sean

[HB]

This is something of great interest to me.
I've noticed that a ton, in fact possibly the majority, of home E61 machines have equal diameter tubes.

Indeed, all the ones I've seen are equal diameter too, though the lengths vary considerably. Referring to the differing diameters, I was thinking of flow restrictors mentioned earlier. BTW, if you really want some good brain hurt on this subject, check out Greg Scace's Results from real time measurements of e-61group temperature and More tuning of thermosyphon groups. They go into a lot more detail.

[espressobsessed]

Dan, how is the e61 commercial restrictor different from a gicleur?

[HB]

Lino provided a very good close-up in E61 Group Espresso Machine: Detailed Interior Schematics; the gicleur is shown in yellow near the bright green section:

Courtesy of Verna Design (please do not copy)

As you can see, the gicleur gates the water's entry into the brew chamber (blue). Tweaking the size of its inlet allows for slower pressure ramp-up. The thermosyphon restrictor on the Legend has a completely different purpose: To control the speed with which water circulates. It is installed along the pathway of the "blue" (return) arrow shown earlier.

It's not rocket science once you have a clear diagram. Many thanks Lino.

[malachi]

Gicleur (in essence) tunes pre-infusion.
Restrictor (in essence) tunes group temp.

[espressobsessed]

does anyone have a picture of a gicleur?

[barry]

does anyone have a picture of a gicleur?

it's a plug with a tiny hole drilled through. depending upon manufacturer, it can be a purposed machined part, or simply a socket head hex set screw with a tiny hole drilled through. nothing fancy.

[malachi]

Some other misunderstandings...

- The E61 group is not a "saturated group."
- Having an E61 style group does not guarantee that your machine offers pre-infusion.

[Strugs]

The E61 does not "run hot". Many people take the phrase "cooling flush" to mean that the group is hotter than the target brew temp and must be cooled by running brew water through it. This is not the case. On HX machines, you do need to flush to cool your brew water - but what you are cooling is the heat exchanger, not the group head. The group, in fact, runs cool and the initial flush will actually serve to not only cool your brew water but also to heat your group.

Malachi (if that is your REAL name....),

Up until a month or so ago, I would have agreed with your statement. However, after installing a TC probe in my E61 group directly below the dispersion screw, I have actually found that the group does in fact run hot (at least on my machine). After the machine has been idle for some time (15 + minutes), my TC probe reads 225 F). Unlike some others that have installed their E61 probes in the easily accessible E61 bolt location, my probe is not affected by water inside the HX. It's location is not saturated with water unless the pump is engaged, so the TC probe is only be affected by the heat of the actual group itself.

See my post on CG that chronicled my science project. There are several pictures in that topic that show the actual location of the probe.

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.