I found its idea is quite interesting: control the temperature of these two pieces of metal, cold water goes in one end, cold water runs through internal path inside the block and absorbs temperature of the metal until it reaches the same temperature, and goes out the other end.

We can have all benefits that we want, such as 0.6mm jet, long internal tube inside the bloc, preinfusion chamber for pressure ramp up; PID, heavy grouphead for direct temperature control. But why this technology has not been in a commercial machine?

I would guess that achieving the heat capacity of a large commercial's boiler you would need a huge chunk of metal?

I imagine, that because of the performance of the thermoblock machines currently available, that to really give the stability that can beat a HX system, or even a Dual Boiler system, it would have to be one huge chunk of metal.


Personally I think that the dual boiler system is the best system available out there right now.. And adding a saturated brew group.. There is some seriously stable temps. Given that the PID is becoming commonplace on a stock setup like this, I almost think that a thermoblock would be a step backwards in temp stability, and even further backwards on temp control, meaning being able to control temps to the half-degree.

I would like to think a thermoblock would allow for a more affordable machine, but I have a feeling that it would have been done already, and it would probably be as or more expensive, given the size that that the thermoblock would be.. Im assuming.

default wrote:I found its idea is quite interesting: control the temperature of these two pieces of metal, cold water goes in one end, cold water runs through internal path inside the block and absorbs temperature of the metal until it reaches the same temperature, and goes out the other end.

We can have all benefits that we want, such as 0.6mm jet, long internal tube inside the block, preinfusion chamber for pressure ramp up; PID, heavy grouphead for direct temperature control. But why this technology has not been in a commercial machine?

Actually, it has been done in commercial machines. The machine sold here http://www.amanti.com.au/prodguide4.htm is a thermobloc design. See also the attached pic taken at the thermobloc stand at the Milan trade show.


Alan

Wall mounted lever machines from the 50s used gas fired thermoblocks/instant heaters. There are people experimenting with using the group or a small thermoblock just ahead of the group to take nearly hot enough water and regulate the temperature very precisely, maybe even profile it over time.

Often the answer to questions like this is the same as why keyboards are still qwerty: inertia.

Thank you for all the answers. Alan, that is very interesting, but the website not saying much.

This idea came up while looking at the design of Aurelia Competition machine. Hot water from lower and upper tubes of the HX is controlled by two different sizes of gicleur, and the big chunk of metal at the grouphead to help stabilize the overheated water, if there is any. Astoria +4U has dedicated heaters at each grouphead to control the temperature of the metal. Dalla Corte has combined the grouphead and boiler into one unit, thus no water travel and heat lost. Just combine them all, and voila, thermobloc.

I think as far as it is a boiler, it still is prone to temperature swing when cold water is introduced to the boiler. Often the reading probe is put at the most fluctuated point in the boiler, and the water outlet at the most stable point in the boiler, which means there is a turbulence of water inside the boiler. And not to mention the distance between boiler and grouphead that water need to travel and lose heat to the environment. Even the newest and well thought-out design, like Kees' Speedster (my dream machine), still needs some offset at the PID and the delivery. This is not to say that boiler system is not stable, but isn't it so '60s?

Would there be any drawbacks of the thermobloc? Faster scale build-up?

One disadvantage, or advantage depending on where you live, is that they work a better in places like Australia and Europe that have 240 Volt AC electric power. My understanding is that they do not work as well in the U.S.A. because of its 110 Volt AC electric power. I think is the reason why there are so few machines available in the U.S.A. with this technology. But I could be wrong.

Geoffrey

It's so easy to make generalizations like "thermoblock = BAD"

But in general, it's rarely effective or productive to over-generalize. Everything depends on the design characteristics, EVERYTHING. Just because many cheap domestic espresso machines use small, short, underpowered thermoblocks made from aluminum does NOT mean that technology couldn't be designed utilizing a similar operation principle to great effect.

In my mind (I'm a career engineer and designer), I can readily envision an extremely accurate and controlled thermoblock-type device for brewing coffee or espresso. Depending on the length of the path, materials of construction and power/arrangement/control of heating elements such a device could possibly be superior to conventional boilers in some ways.

Such a thermoblock would bear little or no resemblance to what is currently used in cheap coffeemakers.

I agree. Sunbeam Australia has come out with some home models that can produce really nice espresso.
I see no reason why the concept can no be taken to higher level ( commercial).

This is what i call a god shot coming from a dual thermoblock machine:



Also we have to keep in mind that generally speaking in terms of high end espresso the word "thermoblock" is cursed....

One Big plus of thermoblock based machines is that they can be easily descaled.

GB wrote:My understanding is that they do not work as well in the U.S.A. because of its 110 Volt AC electric power.

There is 220V available within twenty yards of where you sit reading this. I'd bet my two group on it.

We have 220V going to nearly every residence in the country, and most of the businesses. We just split it up at the panel and distribute most of it around the house as 110V.

In theory an HX machine (when brewing) is basically functioning as a thermobloc machine would... the HX aspect serves to preheat/warm the group up to operating temperature while the boiler serves to heat the cool, incoming water. Maybe an electric heater in the grouphead and a thermobloc heater would be a very valid replacement for the HX? Heater bands can be quite precise actually, which may be a really good way to set up a thermobloc (110 power is the downside to using electric heating coils/bands/etc).

For an example of where this technology is already used and perfected... Anyone really familiar with plastic injection moulding, think of the heater bands around the screw... those are PID controlled too (and really that is a truly ideal application of a PID (at least according to our controls guy from cincinnati milacron))! All that to say, this could be a great way to get into true temperature profiling. Just my two cents.

I guess one way to tackle this would be to calculate how much energy the machine should be capable to put out. If we are talking about a machine with 4 groups, 2 steam pipes and hot water outlet, the maximal thermal energy output is huge. To be capable of that, there either needs to be huge heating capacity or huge energy reservoir. With a prosumer machine demands are smaller.

My gut feeling is that metal would not be very good for storing thermal energy, unlike water. Am I right? But if it is not crucial, then thermoblock would be a one good option. And I agree that boilers are very oldschool and clumsy.

If I was to design an espresso machine, it would have a boiler similar to a HX machine, but instead of HX it would mix hot and cold water to produce what ever is needed. You would need a boiler for steam in any case.

Actually I don't think heat storing that is actually the issue, it is actually heat transfer... water has a very high specific heat, especially compared to metals... like 8 times higher on average compared to steel. Thus, the real challenge is getting the water warm to be ready to be delivered to the espresso (i.e. it takes a heck of a lot more energy to increase a unit of water one degree F than it does steel one degree F)... up till now it seemed easiest to just store hot water that way it can be heated slowly and it large quantity... but an efficient thermoblock could possibly heat the water quickly and thus reduce the need for the boiler. And this is the thermodynamic/fluid dynamic issue that I think is the real heart of the OP question.

Point being, its not storage its delivery capacity, and a boiler is not necessarily the only way to do that... there are many thermosystems which do a similar process without having needed to transfer electric energy to heat energy well before the delivery is actually needed, thus I think the question is quite reasonable and a fairly good idea to explore.

Induction heaters enclosing a ferrous thermoblock?

Non-pressurized PID'd kettle to bring brew water and steaming water to a stable temperature, say 185F, and then it goes through either the brew thermoblock or the steam thermoblock.

Pysd

The point I was trying to make is that in Australia domestic electric power is 240 Volts. Which means power outlets (called power points there) and lights are 240 Volt. Sunbeam in Australia has taken advantage of the high power available due to the higher voltage to make by many accounts a decent domestic thermoblock espresso machine. And there are commercial machines that use the same technology as show in an earlier post.

You are correct, 220 Volt is available at the breaker box in my house in the U.S.A. But typically is only distributed to devices like driers and ovens. Almost all other distribution is 110 V to power outlets and lights etc. Without having a special circuit wired to my kitchen I cannot power up the Sunbeam. Although it may not be a bad idea?

Cheers
Geoffrey

HB wrote:That proves nothing. I've seen plenty of highly photogenic pours that tasted anywhere from exceptional to average to awful.

A beautiful poured shot might not be everything but it is surely a good start.
I have had thermoblock machines in the past and none have been able to pour a shot that remotely resembled the one from the Sunbeam.

Keep in mind that this thread is just discussing if thermoblock machines could be used in a commercial setting.

By the way, I have only had one good commercial shot out of hundreds. The place is Catalina in Houston.
Commercial espresso is usually bad IMHO.

this has been an interesting discussion that makes me want to tinker. Ever since i PID'd my first espresso machine, I've been trying to figure out a method for better temperature control. Conventional thinking is that stability = control. Chasing stability is still very worthwhile: current systems are starting to get close (or there, depending on your point of view) to true stability.

However, as the proponents of humped temperature profiles will tell you, temperature stability is only the ability to control the temperature in a crude manner. Stability is actually the enemy of true temperature profiling. The ability to control the temperature throughout the extraction pushes the boundaries and future of espresso. Perhaps a flat profile is best - but maybe for a particular coffee, or even a particular person. The ability to profile temperature enhances control - and more control is better (unless you want to factor in price, but let's stay theoretical, OK?)

So, I currently see two routes for temperature profiling - a low mass/precise heat solution (thermobloc falls into this category) or a mixing solution (mixing of hot and cold temperature boilers). There are advantages and disadvantages to each approach. Just to hit on a few: The low/mass precise heat approach is probably cheaper for a crude solution (perhaps a separate grouphead heater so you can trail the temp up or down through the shot) but more expensive for precision - precisely adding heat as the water flows through the thermobloc/grouphead is gonna be difficult. The mixing solution requires additional thermocouples and flowmeters for precision, but seems to have a fixed engineering solution that is achievable with todays technology. It's more expensive from the start, but easier to perfect.

All that is just my thoughts. I actually experimented a little with these concepts when i had my zaffiro working - i encased the E-61 in heater strips and ran a separate PID loop to keep the grouphead around 200. So i could start with my boiler water at 202 and my grouphead at 198, to tail the temperature up during the shot, or boiler at 198 and grouphead at 202 to tail the temp down. Ultimately, as i mention above, this approach (grouphead as thermobloc) is rather crude, and without great temperature stability in either the boiler or grouphead (probably +/- 1 degree each) and only the ability for simple rising or declining profiles, i never really got much out of it.

I think i'll have to get to work on a mixing solution.