another_jim wrote:There are no bad machines in this price range, so you're likely to be happy with anything you get.

As a now verteran owner of ridiculously overpriced espresso gear, I can offer a few thoughts.

1. The other machine is always greener. Ignore that, since it never goes away, and pay very close attention to your priorities and how they match up with what's out there
2. The few hundred dollars in price differences will be meaningless after a few months of ownership. Again, ignore that and focus on your priorities.
3. Alas, priorities change with experience, so "the last machine I'll ever buy" is mostly a myth. However, the good news is that priorities change very slowly; if you pcik something that fits, it'll probably last as long as your furniture between redecorations or kitchen between remodellings.

these thoughts are great advice for the prospective buyer. they, along with your excellent review of the elektra, are greatly appreciated .

thank you

kikuchio

I have a question, and hope someone can answer or point me in the right direction.

I'm still trying to come up with an explanation of why the Elektra pulls such good shots, and have come up with a theory that explains it: laminar flow through the puck.

Laminar flow is smooth, all the water molecules moving in lockstep along the line of flow, turbulent flow is when the water eddies or has currents, so that the molecules aren't going in a straight with the flow. I'm theorizing that if the flow through the puck is laminar, the process is much more insensitive to over and under extraction, and can therefore can go well in a wide range of temperatures and pressures. If the flow is turbulent, the water is wandering through the puck, the grounds are getting jostled around, and the extraction is far less uniform. In such a case, the temperature and pressure margins for good extractions are a lot lower.

Lever machines are basket cases for pressure and temperature, yet taste great because the water path is straight down, and laminar by default. What instantly sold me on the Elektra was the clear, undistorted brewed coffee taste of the shots, similar to levers; this would also explain the Elektra's slightly lighter body (less fats and solids getting jostled into the water flow). Moreover, the double dispersion block on the Elektra group is unique, and may accomplish the same task of creating laminar flow in the puck.

So I did some research to shoot this idea down. So far it's still standing, but tottering.

The good news:
-- There is both laminar and turbulent flow through aggregates, including coffee grounds
-- The only literature on this is on making instant coffee (yech, like looking for a ring in the garbage disposal). Keeping the flow laminar in the 8 to 12 foot percolation columns is a key requirement, and requires a lot of engineering. Turbulent flow in these columns means the extraction goes way south (not by taste obviously, but by how uniformly the extraction proceeds, and how much somewhat drinkable yield they get)

The bad news:
-- the only symptom they mention of turbulent flow is channeling, narrow jets of water going through the percolation column, then creating counterflow currents along their edges. It appears to me that on a naked PF, such channeling would be obvious.

So, for the theory to work, it would need to have turbulent flow extractions in most conventional groups that do not show up as lousy naked pf shots. Anyone know if this is possible? If not, I'll have to look for another explanation.

another_jim wrote:Lever machines are basket cases for pressure and temperature, yet taste great because the water path is straight down, and laminar by default. What instantly sold me on the Elektra was the clear, undistorted brewed coffee taste of the shots, similar to levers; this would also explain the Elektra's slightly lighter body (less fats and solids getting jostled into the water flow). Moreover, the double dispersion block on the Elektra group is unique, and may accomplish the same task of creating laminar flow in the puck.

What if all shots started with laminar flow that turned turbulent over different timeframes, depending on machine. And some machines are capable of sustaining that laminar flow indefinitely or at least longer than others. So you could have hypothetically 90% laminar in manual levers, 75% in the Semiautomatica, and 50% in most pump machines? This might help explain why manual levers can produce quality doubles while your machine does better with shorter singles.

espressoperson wrote:What if all shots started with laminar flow that turned turbulent over different timeframes, depending on machine. And some machines are capable of sustaining that laminar flow indefinitely or at least longer than others. So you could have hypothetically 90% laminar in manual levers, 75% in the Semiautomatica, and 50% in most pump machines? This might help explain why manual levers can produce quality doubles while your machine does better with shorter singles.

It's an interesting idea, since the puck changes over the course of the shot, the flow speeds up, and turbulence is a function of flow speed. However, most people using pump machines report extraction problems with singles rather than doubles which would not quite fit the idea.

It's almost certain that the deeper the puck, the greater the chance of channeling. Instant coffee plants try to get extraction columns as wide and shallow as possible to avoid channeling, and at the SCAA, a guy from ESI (sorry, I forgot the name) gave a talk where he stated that he'd examined a lot of the long/narrow tube shaped pucks from several brands of superauto, and found channeling signs in all of them. On the other hand, the deeper the puck, the less the channeling will show at the exit, and if the puck is very shallow, it might "melt down" by the end of the shot and produce a gusher. This could be the reason why people who mainly eyecup their shots (the majority from what I can tell), find doubles easier -- it hides the mistakes.

espressoperson wrote:What if all shots started with laminar flow that turned turbulent over different timeframes, depending on machine. And some machines are capable of sustaining that laminar flow indefinitely or at least longer than others. So you could have hypothetically 90% laminar in manual levers, 75% in the Semiautomatica, and 50% in most pump machines? This might help explain why manual levers can produce quality doubles while your machine does better with shorter singles.

So if you had an adjustable pump you could pressure profile a blend/basket combination by adjusting down your pressure as the shot progresses to compensate for the laminar to turbulent flow transition. By reducing the pressure during the shot (something I actually do with my lever machine) you could increase the length of the laminar flow and improve the cup?

From what I recall, laminar flow is what should happen over the wing of plane, a sail, the keel of a racing vessel. When the flow ceases to be laminar, turbulence and drag result. At some point when enough of the flow is detached, the foil stalls.

The detachment point varies with viscosity, velocity and chord.

Jagged corners and rough edges, like those on coffee grounds, encourage detachment.

The water path through the group is quite contorted in a relatively short distance. Each change in direction introduces turbulence.

Some bright reader should be able to calculate the necessary length required to smooth the flow after the group turbulence introduced and the maximum velocity that will maintain laminar flow for a given, and varying, particle size.

Intuitively, I would be more inclined to give credit to the lack of vibration from a pump.

cafeIKE wrote:Intuitively, I would be more inclined to give credit to the lack of vibration from a pump.

Nah, there's a couple of hypotheses falsified

1. Less vibration creates shot clarity
-- vibe and rotary pumps in the identical machine otherwise produce indistinguishable taste
-- smoothing the pump further doesn't convert it to a lever taste either
-- the semi has a vibe pump without an OPV -- even the gauge needles fall off after a while it vibrates so bad
This hypothesis is dead and buried; although there appear to be occasional memorial services

2. Declining pressure creates shot clarity (spring levers decline by about 33% over the stroke)
-- My initial experiments with a variaced vibe produced a slight improvement
-- Andy's much more high tech PIDed gear pumps also produced some improvement but not the big vibe lever difference
-- The stock Semi's singles are more lever like than any pump machine I know -- 13 to 14 bar vibe pressure throughout.
This hypothesis is code blue, but could still pull through

This is what has me looking for "and now for something completely different"

I'm with cafeIKE on this one.
While I do believe falling pressure has an effect(sometimes huge) on flavor, I would say more of an effect on mouth feel. On one of my machines the group can have rotary, or pulse-less pressure delivery. Even if I match the pressure ramp up of the rotary, and hold brew pressure steady, I see more clarity in the pulse-less.

About half the blends I use I'd say I favor a falling profile(usually not more than 20% drop), with the others being steady, but with different starting ramp times. IMO it seems the darker roasts go well with a falling profile.
So definitely not dead or buried in my mind.
Except for demonstration, I don't use a rotary pump anymore:)

another_jim wrote:This is what has me looking for "and now for something completely different"

My understanding of laminar flow is that liquids (and gases) flow in layers, and with different flow rates in these layers due to decreases in resistance towards the center of the liquid or gas stream. For example, in a pipe or tubing, the fastest flow rate is in the center of the flow. Another example is that water in a river or stream flows the quickest deep in the stream (the main reason that you should not try to cross a road that is flooded).

Once the brew water hits the coffee in the basket, it is hard for me to see how it could continue any laminar characteristics. To me, it seems that diffusion would better describe the process. In either case, channeling is the enemy.

With rotary pumps and vibe pumps with OPV, we are trying to achieve a constant pressure. However, the flow rate is a function of pressure and the resistance produced by the coffee cake.

I don't recall the exact numbers, but during the extraction we are removing between 10-15% of solids by weight. Regardless of the exact number, as we dissolve material from the ground coffee during the extraction we are increasing the porosity of the grounds and the coffee bed. If the pressure is kept constant, the reduced back pressure will increase the flow rate. This will increase the chances of channeling and probably lead to over-extraction.

The spring loaded lever machines produce a declining pressure profile. The manual lever folks comment that they reduce the pressure towards the end of the extraction. Both of these techniques serve to reduce flow rate, perhaps keeping it constant.

More speculation..... The increased body and mouth feel of the pump versus lever might also be due to the increased flow rate towards the end of the extraction. The body is probably due to an increase in the extraction of the fats in the coffee. Fats are insoluble in water, hence they are there in the form of an emulsion. Emulsions are created by rapid and turbulent mixing between aqueous and oil phases. Perhaps the increased flow rate towards the end of the extraction account for the differences in body? Collecting three or four portions for the extraction would shed light on this theory.

Perhaps the ideal machine will not only control pressure but flow rate as well.

another_jim wrote:

cafeIKE wrote:Intuitively, I would be more inclined to give credit to the lack of vibration from a pump.

Nah, there's a couple of hypotheses falsified

1. Less vibration creates shot clarity
-- vibe and rotary pumps in the identical machine otherwise produce indistinguishable taste
-- smoothing the pump further doesn't convert it to a lever taste either
-- the semi has a vibe pump without an OPV -- even the gauge needles fall off after a while it vibrates so bad
-- The stock Semi's singles are more lever like than any pump machine I know -- 13 to 14 bar vibe pressure throughout.
This is what has me looking for "and now for something completely different"

I have been meaning to chime in for a few weeks, so finally the time;

I think the high pressure means that the semi's vibe pulses are a lower part of the overall pressure. In other words at 8 bar the on-off pulses produce big changes on the overall pressure whereas at 13-14 bar they are making less of a change. So it would be more like a lever or rotary pump machine at the higher 13-14 bar pressure.

But here are my 2 new points;

A)

I think the reason singles taste better than doubles is the shape of the basket. They are all shaped pretty much the same from maker to maker and I think the reason is that the shape makes the best tasting espresso. It is obvious that a shorter version of the double basket would be cheaper and easier to make.

In a single basket the coffee is forced together at the outlet and that mixing may be the secret. Also since it is conical shaped it has to have some venturi effect with the water flow accelerating as it moves through the coffee grounds.

It is obvious to me that the double basket is the cheapest way to double coffee production and was never meant to be considered the equivalent of 2 singles. The trade-off in taste for the doubling of production in a cafe environment would be totally justified.

So those who think a double is where it's at need to re-think. If your singles don't have a more nuanced flavor compared to your doubles you need to work on that. There's a reason single malt scotches cost more than blends and generally have less "body".

B)

I think the reason lever machines and the Semi make better coffee is that the water path is straight down through the coffee with all the coffee getting the same pressure and water flow. On E61 (and similar) machines the water enters at an angle and I can clearly see on my Isomac that there is flow closest to the water inlet before there is flow across from it. This happens on both the single and the double baskets. So there is not even pressure throughout prior to flow.

Between my points A and B above I think you will find the answer to the Semi's better coffee production and why a quality company like Elektra would put out such a machine (which defies all logic according to the coffee-tweaks) year after year.

Elektra may have adopted a sales approach similar to Apples old (1980-1997) policy. Us Mac owners were always incensed that year after year they would just keep saying "it's better" instead of laying out the benefits one by one (with their new policy they have taken a little bit more aggressive approach).

Elektra may have the policy of "if you can't perceive the improvement, then maybe you shouldn't buy it" rather than spelling out the reasons why it works better. Also, I can see that Elektra does like to file for patents to protect their technology. But if they know they can't get a patent on their process then keeping their mouth shut is the next best thing and that may be why they don't say "see, here's how/why we do it".

Anyway, when friends ask why I will be selling my Millennium and getting a Semi, I point out that it looks like the espresso machine Flash Gordon would use and that that's enough reason for me .

M./