If you have a thermosyphon loop of a certain capability (diameter pipes, pipe length, etc...) is it true to say that any boiler size would have the same ability to keep the group head heated as long as the element power in the boiler that is heating to water can keep up with the heat loss from the lever group?

The reason I ask is that in a few threads it has been commented that a certain boiler size would not be large enough to keep a commercial lever group hot, but the Strega's active heating is only 100w. So it seems to me that even a small Rancilio Silvia boiler (around the 300ml) would be able to keep up with the heat loss on a commercial lever group.

Comments?

The heating element is no problem. But the typical thermosyphon is based on a heat exchanger inside a 120C boiler. If you modify the Silvia to to drive a thermosyphon lever group like the Rancilio one, it would be a 100C brew boiler thermosyphon. That pushes the limits on E61 single and double machines; and may be too little for such lever groups. Not sure how you could calculate it.

I imagine that if the boiler were too small, the temperature would drop precipitously each time you pull a shot, leading to a thermosyphon stall. The steam boiler on an HX espresso machine is large enough that the temperature drops only a few degrees and the thermosyphon re-establishes itself very quickly after the pump stops (around 30 seconds (?)).

On a related note, the Quickmill Alexia has an E61 group, a 750ml brew boiler, and a 1400W (!) heating element. It works quite well if allowed sufficient recover time, but I believe it's safe to say it would not keep up with even a leisurely home barista if the boiler were half its current size.

Thanks for the replies,
Assume I have a tight PID band and a whopping big element in a tiny boiler so that the temp is stable even after pulling a shot and injecting cold water. It seems that the problem of whether the group will be warm enough is down to the temp differential between the boiler and group for the thermosyphon flow.

I'm not sure how to calculate it either. Can anyone tell me which HX thermosyphon machines (lever or E61) are real firebreathers before installing a restrictor? If I can get some idea of the temp differential between boiler and group on such a machine without restrictors, then it would indicate what the minimal temp differential would be to make the thermosyphon work/flow.

Basically I have a 450ml boiler and a lever group. I want to marry them together into shot machine (no steam). I am trying to figure out if a thermosyphon system will work with the boiler sub 100c, ie not under pressure.

dane5431 wrote:Basically I have a 450ml boiler and a lever group. I want to marry them together into shot machine (no steam). I am trying to figure out if a thermosyphon system will work with the boiler sub 100c, ie not under pressure.

An e61 group idles about 25-30°F / 14-16°C below the boiler temp. The thermosyphon works because the group loses heat from the water, creating a temperature differential. A boiler below 100°C may struggle to keep the group at 90°C

cafeIKE wrote:14-16°C below the boiler temp

Thanks for that. Do you know if this is commonly the case in a thermosyphon loop with a restrictor installed?

My Vibiemme HX PID has a flow restrictor and the DoubleDomo doesn't. For the same group idle temperature, the HX boiler is about 5°F hotter than the DD. When I tried a flow restrictor in the DD as an experiment, IIRC, the group idle temperature dropped a few degrees.

dane5431 wrote:If you have a thermosyphon loop of a certain capability (diameter pipes, pipe length, etc...) is it true to say that any boiler size would have the same ability to keep the group head heated as long as the element power in the boiler that is heating to water can keep up with the heat loss from the lever group?

The reason I ask is that in a few threads it has been commented that a certain boiler size would not be large enough to keep a commercial lever group hot, but the Strega's active heating is only 100w. So it seems to me that even a small Rancilio Silvia boiler (around the 300ml) would be able to keep up with the heat loss on a commercial lever group.

Comments?

Hi there:

There have been successful brewing-only e-61-based machines, although the need for the thermosyphon decreases when building a brew-only device. I think the Isomac Zafiro is a good example.

Your idea will work. You'll have to tinker with the length of the plumbing to the group from the boiler. Ideally a line from the top of the boiler goes into the top of the group and the return goes to the bottom of the boiler in some way. Flow to the boiler when brewing is through both pipes and there is art in how the cold and hot sides mix so that the temperature profile you achieve is what you want. Expect to tinker.

For a brew-only machine there is no fundamental reason why you couldn't use a small boiler. In practice, temperature in small boilers is hard to control if the heating element is large. This is because there is a lot of energy stored inside the element if you blast full current to the element, and this ends up getting dissipated to the relatively small volume of water, causing significant overshoot if you are trying for quick recovery times. Since the only cooling is via thermosyphon loop, you'll have to wait for the system to cool back down. A feedwater preheat scheme such as using a point-of-use hot water dispenser downstream of the pump (an old Marzocco trick) will help. Beware of pressure specs on the dispenser.

-Greg

gscace wrote:In practice, temperature in small boilers is hard to control if the heating element is large.

The plan at the moment is to have two elements heating the boiler. The first will be on all the time, the second will be connected to a PID to compensate for when cold water enters the boiler. I need to test to make sure the first element provides just enough heat to set the group and boiler in a state of equilibilrium or a little less. Therefore the cycling on the second element will be infrequent and I can tighten the tolerance considerably without the loss in element life.

Secondly, as the second element is in addtion to the first, it can be a smaller element than a standard boiler, because the combination of the two elements will dictate the warmup time.

I can't really see any dissadvantage to this setup, apart from the extra cost of a second element.

Hi:

the element for maintaining temperature is gonna wind up being very small, or have very low current running through it. Maintaining temperature doesn't require much power at all. It's recovery that does. You'll still be dealing with the stored heat issue if you are looking at fast recovery times using a second element. So I don't see why two elements is better than one at all, particularly with PID. Larger volumes of water win out. Since there is more water, the temperature drop when you introduce cold feedwater is less, and the temperature overshoot is also less.

-Greg