Originally posted by JetBlack_Espresso
View Post
Announcement
Collapse
No announcement yet.
New Coffee Machine Advice...
Collapse
X
-
There is a restrictor in the top pipe from the hx. So it blends the too hot water from the top of the hx, with the too cool water from the bottom of the hx. Result is perfect temp brew water and no cooling flush
- Flag
Comment
-
My Musica is fine without cooling flush. But it's a bit touch and go as I don't know the start group temp.
I have a Diadema junior extra. Has Eric's group thermometer, so once it's up to temp, it's right on the money. No cooling flush needed
- Flag
Comment
-
The restrictors are called gicleurs, if these are fitted and tuned and the boiler pressure is also tuned, it shouldnt need a flush. Maybe a little purge to stabilise the temp and wet the pf...
- Flag
Comment
-
To my knowledge, that's not how the flow restriction in an Hx thermosiphon works. The water only flows one way, the rate of flow is driven by the cooling of the water in the grouphead so the more heat the head is losing the more water flows, supplying more heat. The restrictor serves to limit the flow rate so this balances near the desired temperature under normal conditions.Originally posted by Jackster View PostThere is a restrictor in the top pipe from the hx. So it blends the too hot water from the top of the hx, with the too cool water from the bottom of the hx. Result is perfect temp brew water and no cooling flush
When a shot is being pulled the water that is already in the HX comes out first, followed by water that is actively heated by passage through the HX. The water that was held in the HX will be too hot but it will be cooled by passage through the grouphead if the grouphead is at the right temperature (as above). The water that passes through the HX and the (now heated) grouphead will be at a lower temperature.
The result is a reasonable approximation of desired brewing temperature, usually declining a bit across the shot.
As I understand it, on a standard E61 Hx machine the gicleur is in circuit for the preinfusion stage but is bypassed when pulling a shot, this being what the mushroom does.Last edited by Lyrebird; 6 August 2019, 10:50 AM.
- Flag
Comment
-
Lyrebird has a firm grip on how heat exchangers work and what the gicleurs orafice is for. All heat exchanger machines will produce water temp that is too hot to brew coffee, it needs to be cooled down first. The one and only boiler in a heat exchanger machine is to produce hot water and steam. The water used for brewing comes through a separate tube/pipe that is embedded in the steam boiler, water just passes through this tube and is passively heated as it either sits in its tube or flows through it. Steam boiler temps are in the 250-270 range when the machine is sitting idle, otherwise you would not have sufficient steam pressure. As you use the machine, the ambient water entering the heat exchanger, tends to be what cools the brew water down, hence the "cooling flush". If you make several extractions consecutively, you only need to run a cooling flush for the first one since new ambient water is coming into the boiler to replace the water used for the extraction. If you hesitate between extractions long enough, the ambient water rises in temp to match that of the boiler, thereby requiring another cooling flush. If you were to lift the brew lever and just let water flow and flow and flow, its temp would drop before coffee brewing range because it didn't spend enough time in the steam boiler to be properly heated.Originally posted by Jackster View PostMy Musica is fine without cooling flush. But it's a bit touch and go as I don't know the start group temp.
I have a Diadema junior extra. Has Eric's group thermometer, so once it's up to temp, it's right on the money. No cooling flush needed
Ok the gicleur is a flow restrictor, noting more. It isn't there to mix or change the temp of brewing water. Its useful in any E61 group to slow the flow of water to a manageable level. Yes, it can have a slight cooling effect if it restricts the flow enough. It could be possible that some manufacturers, Diadema for example, try to balance the flow of brew water to the mass of the E61 group coupled with their boilers heating element's power and recovery such that brew temps are in the 200 neighborhood. However, without proper PID's all bets are off in terms of anything being at the right temp for anything. Without some way to measure the water temp and then feed that temp info back to the machine, the water temp will reach the capacity of the heating elements design and mass.
To keep costs down, a manufacturer can probably design the heat exchanger boiler heating element such that when coupled with the lifting of the brew arm, prioritizes the brew temp over the steam temp to strike a certain balance. This of course is a compromise and my guess would be that steam pressure would hover around one bar and little more in reality and probably be good enough for one steaming session before needing to stop to allow the pressure to rebuild. However, this might be ok for a more entry or even mid level priced machine.
This is why the more expensive machines will have a PID, maybe two or three in order to monitor and constantly adjust temps in the boilers as well as the E61 group, which is what I believe Bezzera is doing in one of their Duo machines that uses an electrically heated brew group, I think that might be the Duo DE IIRC.
Here's a pic that may help.. ..
- Flag
Comment
-
This is almost ... completely incorrect.Originally posted by Martino View PostLyrebird has a firm grip on how heat exchangers work and what the gicleurs orafice is for. All heat exchanger machines will produce water temp that is too hot to brew coffee, it needs to be cooled down first. The one and only boiler in a heat exchanger machine is to produce hot water and steam. The water used for brewing comes through a separate tube/pipe that is embedded in the steam boiler, water just passes through this tube and is passively heated as it either sits in its tube or flows through it. Steam boiler temps are in the 250-270 range when the machine is sitting idle, otherwise you would not have sufficient steam pressure. As you use the machine, the ambient water entering the heat exchanger, tends to be what cools the brew water down, hence the "cooling flush". If you make several extractions consecutively, you only need to run a cooling flush for the first one since new ambient water is coming into the boiler to replace the water used for the extraction. If you hesitate between extractions long enough, the ambient water rises in temp to match that of the boiler, thereby requiring another cooling flush. If you were to lift the brew lever and just let water flow and flow and flow, its temp would drop before coffee brewing range because it didn't spend enough time in the steam boiler to be properly heated.
Ok the gicleur is a flow restrictor, noting more. It isn't there to mix or change the temp of brewing water. Its useful in any E61 group to slow the flow of water to a manageable level. Yes, it can have a slight cooling effect if it restricts the flow enough. It could be possible that some manufacturers, Diadema for example, try to balance the flow of brew water to the mass of the E61 group coupled with their boilers heating element's power and recovery such that brew temps are in the 200 neighborhood. However, without proper PID's all bets are off in terms of anything being at the right temp for anything. Without some way to measure the water temp and then feed that temp info back to the machine, the water temp will reach the capacity of the heating elements design and mass.
To keep costs down, a manufacturer can probably design the heat exchanger boiler heating element such that when coupled with the lifting of the brew arm, prioritizes the brew temp over the steam temp to strike a certain balance. This of course is a compromise and my guess would be that steam pressure would hover around one bar and little more in reality and probably be good enough for one steaming session before needing to stop to allow the pressure to rebuild. However, this might be ok for a more entry or even mid level priced machine.
This is why the more expensive machines will have a PID, maybe two or three in order to monitor and constantly adjust temps in the boilers as well as the E61 group, which is what I believe Bezzera is doing in one of their Duo machines that uses an electrically heated brew group, I think that might be the Duo DE IIRC.
Here's a pic that may help.. ..
[ATTACH=CONFIG]22973[/ATTACH]
Charlie
- Flag
Comment
-
The short version: The water in the HX circuit reaches an equilibrium temperature (ET) that is below the boiler temp and this ET is set by the design of the whole HX/thermosyphon/E61 system. It is possible to achieve a ET that is anywhere from 80 degrees C to well over 100 C. When it's over 100 C it is obvious because the group head will hiss and spit steam if the group head has not been used for some time ("some time" is also a function of the design parameters of the HX system). One factor which may help explain how this is possible - consider that the water constantly circulates through the HX (or boiler in a DB/SB E61) and group head in an E61 equipped machine and the E61 is a heat radiator (effectively cooling down the brew water). It's this cooling that causes the brew water to be below the temperature of the water in the boiler.
A properly designed HX-E61 machine requires no cooling flush and the brew water temperature will be right where it needs to be from first shot to last shot (as long as the machine has had enough time to reach it's ET).
charlie
- Flag
Comment
-
It is physically impossible for the water in a thermosyphon to be in equilibrium when the thermosyphon is working. The disequilibrium is the driving force.
The thermosyphon reaches its operating point when the heat loss from the group head balances the heat gain from the flow of water through the Hx. At this point the water coming from the Hx is hotter than that returning from the group head. If we draw the system boundary around the HX / grouphead loop that system is indeed in equilibrium with its surroundings, perhaps this is what you meant.
If we now force water out of the grouphead by pumping in to the bottom of the Hx, the water at the group will change temperature as it flows. The degree of variation will depend on the flow rates but it can never be zero so the water cannot be at a constant temperature.
Consider the following: Call the flow rate through the thermosyphon Ut and the flow rate when pulling a shot Us. If Ut does not equal Us then the heat loss from the Hx to the grouphead will be different between the two conditions, so the temperature of the water will change as the shot is pulled.
If Ut does equal Us then the heat provided by the Hx will vary because the makeup water from the pump is colder than the return water from the thermosyphon. In this condition the water temperature will again change as the shot runs: the flow rates are equal so for a finite heat exchange capacity the final temperature will also be lower.Last edited by Lyrebird; 6 August 2019, 12:08 PM.
- Flag
Comment
-
Not to take anything from the above posts, but my observation has a slightly different insight.
The gicluer will do little to the flow of the thermosyphon, as this flow will be quite slow.
But when brewing, the pump pressurised both the upper and lower hx pipes equally, so the flow is blended with reverse flowed water from the lower pipe (that has already cooled the group) and the water that is coming out of the hx inside the boiler (that is hotter than the group temp).
I have just had a Diadema junior extra on the bench. It had a scaled hx, and the gicleur was near blocked with scale. It would not come up at the group head (no more than 180f), and the brew temp was way to low. It was taking it's brew water from the bottom cooler pipe and almost none through the blocked gicleur.
I diagnosed the fault by temperature gauges on the top and bottom pipes, and could clearly see the flows by the temperature change. With the top temperature stalled by the blockage, the bottom pipe temperature dropped also, as the cold water into the bottom of the hx was going straight out of the hx and into the bottom hx pipe, up to the group
Obviously, if I removed the gicluer, then the machine would need a cooling flush, and the brew temp would be acquired by blending freshly drawn cold tank water into the hx and using it before it gets too hot.
With the gicleur cleared, and brew selected, the top pipe increases in temp with the additional flow from the top of the hx, and the bottom pipe reduces in temp.
The orifice size balances the hotter/cooler mixing by the ratio of the flows.
My Musica has 2 gicleur, one in the top hx pipe, and one in the bottom pipe.
There was a link that Mal gave but it wasnt directly about hx temp control. But it had a awesome engineering view of the hx temp control. They were basically bagging the dual boiler setup as being too easy, and removing the romance/engineering from machine design. It was one of the very expensive machine websites- van deer Kees, maybe...
- Flag
Comment
-
I think we're now arguing over semantics. There is a thermal gradient in the HX system which drives the thermosyphon. And after a sufficient heat up period, the brew water won't get hotter even if you leave the machine on all day. This is the temp I'm calling ET. Design the system so this temp is, say, 93 degrees and you have a HX which is a breeze to use and enjoy. No cooling flushes required.Originally posted by Lyrebird View PostIt is physically impossible for the water in a thermosyphon to be in equilibrium when the thermosyphon is working. The disequilibrium is the driving force.
The thermosyphon reaches its operating point when the heat loss from the group head balances the heat gain from the flow of water through the Hx. At this point the water coming from the Hx is hotter than that returning from the group head.
charlie
- Flag
Comment
-
Ok, it was Kees, im going to rip it straight from them, and post the link. I hate that in 10yrs time the links don't work, and the info is gone...
Im a bit wrong in my post, but close.. Kees explains very nicely.
Link:
https://keesvanderwesten.com/mirage/construction/
THERMO-SYPHON LOOP AT WORK
The water in the HX is heated by the hot water and steam surrounding the HX in the large boiler. This is an extremely quick way to heat water, thanks to the massive energy in the 125 degrees C. steam boiler and the fast heat transferring capabilities of the copper. It is far more efficient than a heating element. The capacity of the HX therefore may remain relatively small. We use HX`s of 450cc per group.
Heat always wants to go up: the heated water becomes lighter and moves upwards. This flow of hot water leaves the HX, enters the thick tube attached to the top of the HX, then runs towards the group. The group is installed slightly higher than the top of the HX. The group is a massive piece of almost 4 kg of brass. Brass being a very good heat transferring material, soaks up heat from the water when this enters the 35cc chamber at the rear of the group.
The group becomes hot, the water losing much of its heat becoming heavier and starts to sink through the lower thick tube, running from rear of group to the bottom of the HX. Arriving there, the water re-enters the HX, heats up again and all starts anew. The temperature inside the HX is much higher than the normal operating temperature for the group. Capacity to heat the group is thus ensured very well.
This thermo-syphon loop, travelling up from HX to group, then returning to bottom of HX, is driven solely by the heat inside the large boiler. There is no pump or any other device. When the machine is on and at idle (no coffee being made), the thermo-syphon loop is flowing constantly all by itself. As heating the water inside a closed system also means pressure will rise, there is an expansion valve installed. This is adjusted at the workshop to start releasing drops at 11-12 bar. The group solenoid valve has a rating of 15 bar and may act as a secondary expansion valve for extra safety.
SYSTEM AT IDLE
The group sitting at the very end of the thermo-syphon loop is a vitally important stabilizing factor. It is after all the last part water travels through before hitting the coffee, during brewing. Its temperature therefore is important and should be very close to the desired temperature of the brewing water.
The temperature of the group body can be adjusted simply by having more or less hot water flow through its rear chamber. The temperature inside the HX being quite a bit higher than the desired temperature for the group, the thermo-syphon loop needs to slow down, needs to be restricted.
With the Mirage we adjust the boiler to 125 degrees Celsius. To get the group body temperature at low nineties we install a standard restrictor at the top of the HX with a passage of 1.5mm.
With different restrictors installed it is quite possible to operate one machine with temperatures adjusted differently from group to group, for example to accommodate for the use of lighter and darker roasted coffees. As an option we offer adjustable restrictors.
One also may attempt to change the group temperature by changing the boiler temperature, but this method is limited because fairly large boiler temperature changes are needed to realize just a little effect at the group. Also the steam or hot water capacity of the boiler may be affected that way. With the proper restrictor one can for example easily combine the extremes of 1.5 bar high steam power with a low 88 degrees C. brew temperature.
BREWING
The challenge of the thermo-syphon system is to get water of low nineties temperature to the coffee, provided by a HX that stores water at 125 degrees. The extra heat is needed to keep the group at its perfect temperature, the remaining heat is needed to deliver brewing water of the correct temperature. To start the brew process the group solenoid valve opens the path from the thermo-syphon system towards the coffee in the filter holder. At the same moment the pump starts to run, keeping the pressure at a steady 9 bar, everywhere between pump and coffee.
The sudden exit, provided by the open solenoid valve, has a drastic result. All water will now flow to that exit, pushed out by the pump pressure. The flow in the lower tube changes direction, it flows upwards to the group, instead of downwards to the bottom of the HX. The flow in the upper tube remains the same. The crucially important second function of the restrictor now becomes clear. The flow through the upper tube towards the group is severely restricted by the1.5mm orifice of the restrictor. The lower tube does not have any restriction, it has an inside diameter of 8mm. This full-flow 8mm, compared to the 1.5mm, means 99% of the water reaching the group arrives via the lower tube.
This suits the requirements very well, since this lower tube contains the cooler water (generally 88-92 degrees, compared to the upper tube 115-125 degrees). To get the most out of this, we install a very long lower tube in the Mirage, containing 30cc. The remaining water comes directly from the HX. All water leaving the HX is replaced simultaneously by fresh cold water. The hot water leaves the HX through the lower tube, from bottom of HX. The cold water enters the HX also through the bottom. The cross flow of outgoing hot and incoming cold water is used to help each other. As there is just a 1mm copper wall between these 2 flows, the outgoing hot water is cooled down considerably, losing its heat to the incoming cold which starts to heat up right away. The cold water is forced to travel to the top of the HX via a long injection tube.
When leaving the top of the injection tube it needs to travel down again to the bottom of the HX, during brewing. This ensures the heating capability of the HX is used to its full capacity. Naturally, the cooling down of the water from HX to lower tube by the crossflow is not a very precise process, as the temperature of the cold water can vary because of climate, location, season, etc. It will however be a major drop to well below 100 C. Travelling upwards through the long lower tube offers a good start to further stabilise. When entering the big brass group body the temperature of the water will be easily stabilized to the correct temperature of the group.
To facilitate this temperature stabilization process we changed the insides of the group to increase the contact of water with the group body as much as possible. During brewing the water first passes the 35cc chamber in the rear of the group, then runs to the solenoid valve from where it re-enters the group. It fills the infusion chamber, passes through the passages towards dispersion fitting and finally is pressed through the group membrane- type screen.
The stabilizing ability of the mass of 4kg brass is impressive. Even after taking a bucket of hot water from the main boiler, causing this to cool down for a short while during re-filling with cold water, the energy stored within the brass ensures enough shots of the correct temperature to allow the boiler to regain its standard temperature. When starting a shot, we recommend to draw a short flush: 30-45cc. Never flush more, because the water runs much faster through the entire system during a flush. When flushing longer the inherent stability of the system can easily be ruined, requiring at least 5 minutes to completely restore.
CONCLUSION
55 Years after its creation the E61 thermo-syphon system can still very much be considered a technically highly elegant system. Especially the thermal efficiency impresses and easily surpasses the popular multi-boiler systems. In comparison, the double, or multi boiler system, can technically be regarded as rather run of the mill.
It is a great pity the Italian espresso machine industry, although widely adopting the thermo-syphon idea, did for about 40 years not take any initiative to keep on developing. Instead, it finally took the easy way, went back in time and grabbed the previous old idea of separate boilers for steam and coffee production. As a tiny company we could only follow. We do our best to offer the finest of either system.
- Flag
Comment
-
Jackster, your text is very interesting and I agree with your conclusion, HX is a great technology and it could be improved.Originally posted by Jackster View Post
CONCLUSION
55 Years after its creation the E61 thermo-syphon system can still very much be considered a technically highly elegant system. Especially the thermal efficiency impresses and easily surpasses the popular multi-boiler systems. In comparison, the double, or multi boiler system, can technically be regarded as rather run of the mill.
It is a great pity the Italian espresso machine industry, although widely adopting the thermo-syphon idea, did for about 40 years not take any initiative to keep on developing. Instead, it finally took the easy way, went back in time and grabbed the previous old idea of separate boilers for steam and coffee production. As a tiny company we could only follow. We do our best to offer the finest of either system.
But I believe that the Italian espresso machines for shops, such as Victoria Arduino, Simonelli, La Marzocco are HX, could you confirm that? or these companies are selling DB machines for shops too?
Althought, I'm sure that DB is a trend in the prosumer segment.
- Flag
Comment

Comment