HX vs. double boiler temperature stability - Page 2

[HB]

...in your opinion is the DB setup easier to reproduce since they seem to closer to flat than an HX?

As Randy said, double boilers are demonstrably easier to use. However, a skilled barista can produce equivalent results with ease using either boiler design. To use an automotive analogy, a skilled driver can drive cars with an automatic or manual transmission equally well, but may prefer one over the other.

Specifically to your question, I agree with your assertion ("the DB setup easier to reproduce") but not with the causal factor ("they seem to closer to flat than an HX"), unless of course we're talking about newbies, in which case the double boiler ease of use carries more weight than it would for an experienced barista. Finally, it's worth noting that not ALL heat exchangers require more skill to use -- the Elektra A3/Semiautomatica and Bezzera BZ07, for example, have nearly idiot proof temperature management.

[another_jim]

This is only complicated in the abstract; from a user point of view it's dead simple. If you like to pull three or four different espresso blends or SOs each day; a DB is a pain, since you can't change temperatures on the fly. If you pull just one very persnickety and precisely dialed in blend; an HX is a pain, since you can't dial it in as precisely and repeatedly for temperature. The only thing you need to know is how you like to pull shots.

[smite]

Ive been looking at the VDD DB machine. So, when brewing a shot, what does the PID show?

I can tell you based on my limited experiences and having owned both the Duetto II and VBM DD that during shot pulls, the temp displayed on the PID does change, while and after the shots are pulled. As can be expected this was more noticeable, when pulling many back to back shots. I also noticed that running the Duetto II in 20AMP mode my recovery time was slightly faster and seemed to have less intra and inter shot variability.

I also experimented slightly with measuring extraction temps via an IR thermometer as they were pulled from both machines through a naked PF and was not able to get any meaningful information.

Not that it matters much, but I no longer own either of those machines and have since moved on to an Elektra T1 and temp management with this HX machine is exceptional.

J

[Ltrain5000 (original poster)]

Awesome-- what does the PID show in regards to a swing?

Is there a temp probe on the HX machine? How stable is it?

[smite]

From what I recall on both of the DB machines the swing was pretty minimal 1-2 degree fluctuations with 1-2 shots. Others, who still have either one can probably give you specifics.

Another thing to consider is that depending on the offset, time for temp registration on the PID, may not always reflect immediate real time temp changes at the brew head. I am sure that others who are more knowledgeable on the subject can elaborate on the matter much more if needed.

In terms of the Elektra, there is a computer control that can be used to adjust temp through adjustment of the steam pressure but there is not a digital readable temp probe. I manage temps through the details of HX management. Dan posted a great update to his long standing article of HX management:

Ideal brew temperature management by HX espresso machine type which combined with help from others on this forum, provides great detail and was helpful with mastering my current machine temps and pulls.

J

[boar_d_laze]

Is there a temp probe on the HX machine? How stable is it?

No temp probe.

The Elektra T1 is extremely stable as HXs go. However, that doesn't really get to the heart of the theoretical or practical matter.

The first thing you have to address is what does "stable" mean in terms of an HX. It actually doesn't mean much. To a great extent, the beauty of the HX design is its inherent instability -- which translates as an ability to easily adjust for the desired temp on the fly.

The primary source of instability is inherent to the design. Because the HX (that is the heat-exchanger coil itself) gets its heat from the boiler into which it's set, the water in the HX will reach the same temperature as the boiler itself -- and relatively quickly at that. In other words, after a period of at most a couple of minutes, the water in the HX (held under some pressure) will be hotter than the boiling point, and will exhibit flash boiling when it emerges from the group.

To some extent, the "stability" of an HX machine might be thought of as how quickly it rebounds from the lower temp at the end of a shot (or after a sequence of shots) to it's "normal" condition of too hot -- so the temping sequence can be re-started from the flash-boiling set point.

The barista uses a "cooling flush" to get rid of some of the old, hot water, and mix it with new water and reach the desired temperature. The barista creates a reference point to gauge the length of the flush, by noting the moment that flash boiling stops. After that, the longer flush, the cooler the brew water.

Note that the cooling flush not only cools the brew water in the HX, but cools the rest of the group -- which tends to become overheated. The Elektra uses a "thermosyphon" type group (the Elektra is not an E-61, but damn well like an E-61) which helps to keep the group temp more stable than it might otherwise be.

Whether HX, DB or lever, the entire group plays a strong role in intra-shot temp stability (that is, what the temp curve looks like during the shot). Typically, the best way to stabilize the group itself is to run a lot of water though it; which is why most espresso machines work better after the second or third shot in a series when they're running as fast as can they go and still maintain temp. The idea is not only to control the temp of the brew water but to equilibrate the group.

From an engineering standpoint, that takes us back to thermosyphon and other group stability strategies. From a barista standpoint, it either means lots of water or an extremely clever machine. But no matter how clever the machine, if I'm the one pulling the shots, I'm still going to run some water if the machine has been idling for more than a minute. Because, why not?

Getting back to the Elektra T1 specifically, it is one of the few, friendliest and most consistent single-group HXs. There are other single groups which do as good a job, but most are considerably more expensive. The only real competitors of which I'm aware are the La Cimbali M21 "Casa" and the Nuova Simonelli Appia.

Once you've figured out how to temp your particular HX -- which is pretty frikkin' easy -- there aren't many drawbacks to an HX design itself as compared to DB design.

TMI?

BDL

[Peppersass]

If you like to pull three or four different espresso blends or SOs each day; a DB is a pain, since you can't change temperatures on the fly.

What's painful about changing the PID target temperature, flushing a little water, and allowing the boiler to come up to the target temperature? That takes less than 30 seconds on my GS/3. I don't find that burdensome for three or four shots a day.

[HB]

If you raise the temperature, the boiler may well reach the target temperature in 30 seconds, but the rest of the group takes time to catch up. It takes even longer if you're lowering the brew temperature. For the GS/3, I flush at one minute and give it another two minutes to stabilize if increasing the temperature; for lowering the temperature, I do a long flush, wait a couple minutes, flush again, then wait a couple minutes. This routine is based on Scace thermofilter measurements I did years ago, so I certainly welcome corrections.

Jim's point is that you can't force a La Marzocco GS/3 to bounce up/down a couple of degrees every other shot like a skilled barista could on an HX espresso machine by manipulating the flush/rebound time. This should not come as a surprise to anyone, since one of the GS/3's raisin d'être is to hold the same temperature independent of flush routine.

This returns us to Jim's closing comment:

The only thing you need to know is how you like to pull shots.

[boar_d_laze]

What's painful about changing the PID target temperature, flushing a little water, and allowing the boiler to come up to the target temperature? That takes less than 30 seconds on my GS/3. I don't find that burdensome for three or four shots a day.

For one thing, not every DB is as quickly responsive as GS/3. >$5K single groups breathe a rarefied atmosphere; and the GS/3 isn't exactly typical.

For another, what Dan said -- which tracks what I said -- equilibrating the group of any machine takes time, or water, and preferably both.

For a third, I knew the GS/3 brew boiler could heat up quickly, but wasn't aware it could shed heat as fast. Not saying it can't do it, just saying I either forgot or never knew.

Frankly, I'm not sure that I can get repeatable temp changes in 30 seconds even with my La Cimbali -- which does temp changes about as fast as they can be done -- because it takes about a minute after pulling a double to rebound to the flashboil baseline.

BDL

[coffeesnob1]

One thing that I don't see mentioned is how the environment should be taken into factor which machine is right for you. In my experience at least, I've noticed that DB's in the E-61 class and the GS/3 tend to radiate ridiculous amounts of heat especially so if you use them in a cooler basement for instance. I didn't realize what a huge amount of heat I was losing until I figured out that locking in the portafilter (already packed with fresh grounds) and waiting 30 - 45 second prior to pulling the shot made a huge difference in taste. The hotter group on a h/x may give it an advantage here but definitely not a deal breaker, just something to consider. Of course a basement (65F year round) is definitely not the ideal spot for either machine, unless you have a compulsive tendency to pop up in the middle if the night periodically as I do craving the Divine goodness and don't want to wake anyone (I'm one of those rare people that it has a relaxing effect on.)