What is the purpose of long HX flushes? - Page 2

[HB]

Another variable to consider is the length of the HX injector. It controls the mixing of overheated HX water and cool incoming water. The shorter the injector, the faster the recovery and longer the idle flushes. According to Michael Teahan, Italian roasters even manipulate the HX injector length to produce different "humps" in the temperature profile for a given espresso blend.

[cafeIKE]

How hard would it be to measure the temp of water in the thermosyphon as well as the GH temp and come up with some sort of calculation regarding heat loss?

These numbers are only applicable to the Vibiemme Domobar Super. Machines with a different TS volume and geometry, group mass and geometry and few other variables respond differently.

On the Vibiemme prior to PID, TS inlet ~230°F and outlet ~213° . Group ~205°F after about an hour. Boiler an uncalibrated 1.1bar. External boiler temperature ~245°F.

Post PID inlet ~211°F, outlet ~195°F, boiler 232.7°F, group ~198°F after an hour. However, changing the PID parameters for a rock solid boiler temperature of 232.5°F setpoint idles the group at ~196°F.

As you can see, the I/O delta changes from ~17° to ~16°F, the inlet drops ~19°F, but the group only drops ~8°F.

These numbers are not absolute and a completely different set will be obtained depending on whether the bay window is open or the air conditioning is heating or cooling.

The numbers vary slightly lower with no PF in the group and slightly higher with a wet puck left in the basket.

[cafeIKE]

Another variable to consider is the length of the HX injector. It controls the mixing of overheated HX water and cool incoming water. The shorter the injector, the faster the recovery and longer the idle flushes. According to Michael Teahan, Italian roasters even manipulate the HX injector length to produce different "humps" in the temperature profile for a given espresso blend.

They vary not just the length, but also add holes along the length of the injector as well.

[RapidCoffee]

Another variable to consider is the length of the HX injector. It controls the mixing of overheated HX water and cool incoming water. The shorter the injector, the faster the recovery and longer the idle flushes. According to Michael Teahan, Italian roasters even manipulate the HX injector length to produce different "humps" in the temperature profile for a given espresso blend.

Yet another variable: position of the HX tube within the boiler (submerged in water or steam). Neither of these seem easy to modify on prosumer E61 machines.

I've always been intrigued by Faema's adjustable thermosyphon restrictor. In light of the preceding discussion, it probably should be a standard fixture on E61 machines.

[erics]

I've always been intrigued by Faema's adjustable thermosyphon restrictor. In light of the preceding discussion, it probably should be a standard fixture on E61 machines.

I am likewise intrigued. This valve obviously(?) controls the flow in the thermosyphon system which I figured (on the back of an envelope, naturally) to be around 3 ml/sec. Ironic that it is about the same as typical espresso flowrates. The valve would also "control" the mixing of brew water (once the shot commenced) from the upper and lower hx lines. Naturally, on the 2-group Faema Legend, there is a valve for each group. Just guessing here but I can imagine that the valve would be cranked open for the "morning rush hour" in a cafe, tweaked down as the day progressed, and opened wide again for the evening crowd. Just this AM I ordered one of these valves to play around with.

Controlling the flow in the thermosyphon was successfully done by Greg Scace on his Astra:

http://tinyurl.com/3dogo6

The above TinyURL gives you the results of a search for "thermosyphon" on GoogleGroups Alt.Coffee. A potential negative would be the time it takes for the grouphead to recover after pulling a shot.

[RegulatorJohnson]

A potential negative would be the time it takes for the grouphead to recover after pulling a shot.

and could this also be considered a positive? if they can turn it up in the morning and then down in the slow part of the day then back up in the evening rush. they are controlling not only the group temp but also the intershot recovery duration.

with thermosyphon control you can adjust brew temps by adjusting the grouptemp using the thermosyphon flow rate, you dont alter steam power. without thermosyphon control, you control brew temp using the pressure stat, this also will change your steam power. if we use the e-61 with a "missing" part in the system, the adjustable thermosyphon valve, we then lose independent control of steam power and brew temp.

could the missing valve be the reason why our HX's need cooling flushes? did faema design the e-61 to be used with the valve, but not "advertise" the need for this thermosyphon to be controlled?

i can tell you that the pulser makes a great shot either with or without the restrictor. but with the restrictor in there it could make shots with less fartin around and wasting tank water, feeling self conscious when i have to just sit there and pump out water because they don't do that in the real cafe, blah blah, lots of steam and mellow brew temps, 2 minute rebound/brew temp range duration. (how long it would stay in the range of brew temps, without a flush) with out the restrictor it would be 15-20 seconds.

i whole heartedly support this direction. i have experienced what changes can occur. i want my group to idle @ 206-208° at idle. on the vetrano its 210-212 (stock pulser i believe was very high cant recall) i have been thinkin of slipping a extra pulser restrictor into the vetrano. but dont want to "void" my warranty. i dont think it would blow the machine up. i am not an expert but i also dont think it would damage anything by being in there. chris' service dept warranty claims might have a different opinion.

in some ways like miss the pulser, more specifically the little white disc inside it.

jon

[RapidCoffee]

So to summarize: long HX flushes on prosumer machines are due primarily to
1) overactive thermosyphons, which cause the grouphead to run hot
2) injector dynamics, that cause mixing (as opposed to replacement) of overheated HX water with cool incoming water

Anything else I might have missed?

[erics]

My opinion is that it is very difficult to summarize why hx equipped machines require long flushes but a sincere "thank you" for your attempt to do same.

A real quick way to reduce the cooling flush requirement is simply to dial down the pstat to the lowest level that will still provide satisfactory steaming capability - of course you may also want to throw some limits on the number of espresso loving dinner guests.

A. Several months ago, I attempted to simulate (and I may have posted about it) Bob Yellin's "quick & easy" flush regimen on his Andreja Premium. His procedure was simply to flush 6 oz, wait 45 seconds, and pull the shot. At 1.10 bar on my Anita, this resulted in a pretty stable 200 F shot.

B. Over the past 10 months, I have done long flushes and started the shot when the grouphead temperature was in the 197-198 range. This takes a couple of minutes but also results in a pretty stable 200 F shot.

C. Just this evening, for the heck of it, I flushed the group until the temperature read 200 F and then almost immediately pulled a pretty stable 200 F shot.

D. When I started reading & learning from this and other espresso forums, there was a lot of talk about so-called sink shots. What if you simply pulled a shot after the 1.10-1.15 bar machine had been on for about an hour. Obviously, it would be waaay too hot but what about the next (within seconds) 25 second adventure?

A, B, & C were all done with max pstat settings of 1.10 to 1.15 bar and serve to illustrate that the same results can be achieved in a variety of ways. I'll do "D" over the weekend.

[knewmans]

Another variable to consider is the length of the HX injector. It controls the mixing of overheated HX water and cool incoming water. The shorter the injector, the faster the recovery and longer the idle flushes. According to Michael Teahan, Italian roasters even manipulate the HX injector length to produce different "humps" in the temperature profile for a given espresso blend.

My carimali has an adjustable injector to alter brew temperature. It needs to be dismantled to adjust so not something you can change on the fly but straightforward. The group head is also electrically heated so no thermosyphon overheating.

Using a styrofoam cup and thermocouple, as far a I can see, back to back double shot flushes are the same temp.

[erics]

A potential solution to "long" hx machine flushes - this is the infamous Faema adjustable thermosyphon valve.

[jesawdy]

A potential solution to "long" hx machine flushes - this is the infamous Faema adjustable thermosyphon valve.

Eric, did you order a complete assembly? What did it it set you back?

[erics]

Jeff -

About $127 but it is a rather nicely machined valve. Unfortunately, the only way the end user can buy this valve is as an assembly.

[Randy G.]

My naivete notwithstanding in this area, but it would seem to me that it would be a simple matter for the manufacturer to design in a wax pellet type thermostat (as in a car) or possibly a bimetallic spring that opens and closes a small shutter door in the brewhead to regulate the thermosyphon flow. An external screw would make it adjustable as well. This seems a better solution since it would be sensing the brewhead temperature itself.

[shadowfax]

Randy, that idea has really been intriguing me. It seems like you really ought to have your cake and eat it too, with an HX, with a valve on the thermosiphon that simply almost closes as the water in the pipe gets too hot, but opens all the way when the water is cool. This could, I would think, easily be designed so that your recovery is fast (the disadvantage to a restrictor), but also such that your grouphead idles as brew temperature, keeping the flushes short.

Honestly, though, my irritation with wasting water on big flushes notwithstanding, I care a lot less than someone that doesn't have a plumbed in driptray...

[mhoy]

Connect a bimetallic spring (or perhaps a PID) to the copper of the HX loop. Use this to enable a relay to open/close a solenoid on the HX loop. Not as nifty as a saturated group or an actively heated group head, but there are a lot of thermosiphon systems out there.

Mark

[Randy G.]

Honestly, though, my irritation with wasting water on big flushes notwithstanding, I care a lot less than someone that doesn't have a plumbed in driptray...

I don't need a plumbed drip tray. I have a wife. ..I hope she doesn't read this forum!