Wow, if this thread had been about airplanes, the Wright brothers would have given up hearing time after time, "Every flying machine I've ever seen or heard of was either a joke or a disaster."

Plenty of industrial machines flash heat large amounts of material and pump it under pressure; injection molding machines and extruders come to mind. So it seems very possible to build a steam generator without a boiler. In fact, if you search that, you can go buy some, though they're a little large for your countertop. On the extreme end of the scale are nuclear fired ones used for power generation

So the question comes down to "Why isn't there a good thermoblock-style steam generator for espresso machines?"

I think the question comes down to peak power consumption, especially in home machines.

In order to generate the steam in a thermoblock, you need to deliver the power to the element as the water/steam is flowing. In a boiler, you can provide it also when the steam isn't being used. Even with a 50/50 duty cycle of steaming and not, that could reduce the current draw at a given supply voltage by a factor of two.

Here's a concrete example. My Anita draws 1450 W when hot and heating, nearly the limit of a domestic 15 A circuit). From other measurements, I know it "idles" at about 120 W on average (mainly the power required to balance heat loss from the boiler and group). So, here's the experiment:

• When the green light (pstat high-pressure contact closes) comes on, open the steam wand full
• Heating element turns immediately on due to pressure drop
• Leave steam flowing for 30 seconds and close the valve
• Measure the total time until the green light comes on again

For my Anita, retrofitted with the Chris Coffee no-burn arm and lever valve, it takes 48 seconds (total) to recover the heat lost in 30 seconds of steaming. If that energy were to be concentrated in the steaming time, it would be about (48/30) * (1450-120) ~ 2200 W.

That's a pretty hefty load, significantly more than a typical home appliance could draw.

So as not to discourage prospective developers, there is probably something clever that could be done with a big enough brew boiler to pre-heat the water (the Anita draws from an internal tank of what ends up being warm water), though I believe that the heat of vaporization (539 cal/g) dominates over the heating of the water to boiling (say 100 C - 25 C ~ 75 C at 1 cal/g/C ~ 75 cal/g).

In theory there is no difference between theory and practice. In practice there is. - Yogi Berra or somebody else

The most reliable answers about Silvano steaming will be from experienced users who have had direct experience on the (now 1200w) version of the machine. Otherwise the Chris Coffee video would seem to be the next best indicator.

I wonder how the Anita's and Silvano's would compare with say a Vivaldi db at 120/15A and 120/20A.

The little known secret is that the brew boiler can be very very small. The steam boiler must have some volume.

Folks with big boiler SBDU's and boiler for brew, thermoblock for steam have it backwards.

ampguy wrote:The little known secret is that the brew boiler can be very very small. The steam boiler must have some volume..

A poorly kept secret.

For commercial dual boiler machines, both boilers tend to be quite large to deal with the sort of frequency of use in a shop, particularly with multiple-group machines. In the home DB machines (not that there are a lot of them to compare) they mostly all follow this design principle. There are many SBDU machines that show great potential for quality espresso creation, and these mostly all have small boilers. Unfortunately, this potential is rarely achieved out of the box, mainly due to the use of button thermostats.

I thought the advantage of a small boiler for brewing is quick re-heat ability. But how does the small boiler help or hinder temperature stability (assuming a PID controlled unit)?

Lower thermal mass and quicker response can be regulated well, IF you have a PID and good sensor making small power decisions precisely. OTOH if the heater power has traditional non-pid switching with more hysteresis, you need larger thermal mass to keep things more stable... like a flywheel on a rough idling engine. If the engine idled smoothly and consistently and was less massive, you would need less flywheel.

innermusic wrote:Is it just about the nuances? Does the TB do what the discrete boiler does, but with less control over the variables? Is that it? Is there more to it?

I think the Silvano is very similar to a dual boiler, but from my understanding the water used for the steam/thermoblock comes from the main boiler, right?
So you use a bit of the hot water in there, which could affect the temps, however I think the boiler is large enough that it retains heat well, and the PID seems to work to keep the temps pretty stable, it tends to go down a few degrees after steaming for a bit, but seems to rise back fairly quickly in my experience... well at least based on what it displays on the PID.

I'm curious to test what the actual temps are like at the group head, if it matches the PID display (and test brew pressure gauge as well for accuracy.)

cpreston wrote:The most reliable answers about Silvano steaming will be from experienced users who have had direct experience on the (now 1200w) version of the machine. Otherwise the Chris Coffee video would seem to be the next best indicator.

The 0.7L brass boiler is 600 watts, the steaming thermoblock is 1000 watts.

Benjammer wrote:I think the Silvano is very similar to a dual boiler, but from my understanding the water used for the steam/thermoblock comes from the main boiler, right?

The Crossland CC1 works the way you describe. The Silvano has a separate pump drawing water from the water tank to feed the large thermoblock.

Benjammer wrote:I'm curious to test what the actual temps are like at the group head, if it matches the PID display (and test brew pressure gauge as well for accuracy.)

This short video from Chris Coffee will answer that temp question:

http://vimeo.com/15671852

Oh nice, EricBNC so you're saying it takes the water directly from the reservoir and it hits the thermoblock right away? I guess that's why there's two separate water intake hoses in the water reservoir, I thought one was a backup one in case the other was clogged

Oh nice vid too, looks like it was set to 200f and stayed pretty close to 200 once it heated up, I would have thought with the rapidly falling #s on the PID it would be lower, so that's pretty awesome.

Benjammer wrote:Oh nice, EricBNC so you're saying it takes the water directly from the reservoir and it hits the thermoblock right away? I guess that's why there's two separate water intake hoses in the water reservoir, I thought one was a backup one in case the other was clogged

It is nice. You can heat milk for hot chocolate after only a couple minutes if that is all you want and not espresso - it also lets you steam milk and pull a shot at the same time. Pretty nice for "entry level"

Benjammer wrote:Oh nice vid too, looks like it was set to 200f and stayed pretty close to 200 once it heated up, I would have thought with the rapidly falling #s on the PID it would be lower, so that's pretty awesome.

It does keep the water heated through the shot - recovery is quick too - faster than I am when weighing beans, grinding and filling the basket. It has limitations on speed (if pulling shot after shot) but for personal use production is more than adequate.