'Flush and go' technique suitable for smaller HX machines?

[quar]

Hi All,

Anybody have any thoughts on using the "flush and go" technique vs. the flush and wait technique on a smaller HX machine, like the Anita?

I started out using the flush and wait technique detailed in Dan's Stop Worrying and love the HX article, but after a couple of months, transitioned over to the pro-style "flush and go" flush. I seem to remember being happy with the results, but unfortunately close to a month of down time due to repairs, so I really can't remember. Since getting everything fixed up, things seem to be running a little on the sour side. I'd been blaming in on the blends that I'm trying being a little brighter than I prefer, but am starting to wonder if it's something else. After thinking about it for a while, I'm wondering if this technique is just overwhelming my Anita and the temp is dropping too low during the latter part of the extraction.

Mike

[HB]

Some advocate the "flush and go" approach, notably JonR10 and Chris Tacy (he called it the "pro's method" in early drafts of his equipment writeups), rather than the flush-rebound-go method described in the HX Love article. I've tried both methods and in most cases, they are nearly equivalent. The essential difference is the flush-and-go method is surfing to the desired brew temperature and very quickly starting the extraction, the flush-rebound-go method is surfing slightly past the desired brew temperature and rebounding back to the desired temperature. While it may seem a bit odd at first, I found that consistency for small HX machines was more easily obtained "surfing up" because it occurs more slowly (15-25 second rebound) versus "surfing down," which relies on more precise flushing and no dawdling.

Either approach would work with the Anita, which I assume works just like the Andreja. If you want a leisurely pace and a wider window of error, try this: Reduce the boiler pressure to 1.1 bar (top of cycle) and flush to the end of dance (~206F) + eight seconds (one degree / second). Then allow enough rebound time to taste, probably around 20 seconds to arrive at ~202F with a slightly humped profile. Otherwise the flush-and-go should be more like flush to the end of dance (~206F) + six seconds and no more than 10 seconds before starting the extraction. Of course these timings are just estimates... you'll need to adjust by taste.

PS: Related comments in Boiler pressure setting - influence on the cup? excerpted below:

Why would one choose to adjust the pressure?
Abe mentioned it in his review - to allow a different range of temperatures on HX machines, pressure can be adjusted. That made me curious.

Abe's review described brew temperatures ranging from 91C (195.8F) to 96C (204.8F), a range which his Giotto wasn't able to deliver at the same boiler pressure by flushing. Practically all the coffees that I work with are brewed in the narrower range of 198F (92.2C) to 204F (95.6C), with the majority in the ~201-203F range. A four degree range is well within the sweet spot of most HXs simply by adapting the flushing amount. You can add a couple more degrees to that range by switching to the flush-and-go technique ("pro's method"), but as Abe pointed out, it requires more skill to nail a precise temperature consistently, especially for the smaller HX machines.

Keep in mind though when reading these discussions that boiler pressure is only one variable that determines an HX machine's thermodynamics. The size of the HX, the percentage immersed in the water versus steam, the deliver tube lengths, group design... all of these contribute their part, so it's folly to draw precise instructions from general advice about HX machine operation. For example, the article How I Stopped Worrying and Learned to Love HXs really speaks to E61 / HX machines; the steps its advocates apply vaguely to the Cimbali Junior and not at all to the Elektra A3, despite that they are both HX machines. There's differences among prosumer HX machines too. The flush routine for the Andreja Premium would apply in principle to the Expobar Lever, but fails miserably in the specifics (the Lever has a much longer flush and it rebounds much faster at the same boiler pressure setting).

[malachi]

Keep in mind that the extraction profile will differ. So you'll likely find that one gives you a result in the cup that you prefer for some coffees.

[JonR10]

heh heh - looks like I'm chiming in late...

Very often I advocate the flush-and-go method because I feel like some folks will benefit like I have by exploring the lower end of the temperature curve. To me, it's all part of developing a familiarity with one's machine. I think this is a key (in my own experience) for making more enjoyable espresso at home.

Developing a "feel" for my machine's operating characteristics lets me make temperature adjustments "on the fly" and often without thinking about it very much at all. It starts (for many) after finding bitterness in the cup. I have found (after helping a few people) that many folks do not understand how quickly some HX machines recover and/or how much water needs to be flushed before running the shot.

If my first shots (of a new roast or blend or bean) have an edge then I'll flush more or wait less.
If the shots have any sourness then I may wait more or flush less (or both).

If you really want to accelerate the learning curve for your machine, then take some temperature measurements.

• Measure the temp of the flush itself.
• Measure a "blank" shot pulled immediately after an 8-ounce flush (from idle).
• Measure the water temp 20 seconds after an 8-ounce flush (starting from idle).
• Measure the water temp 40 seconds after an 8-ounce flush (starting from idle).
• Measure the water temp after flushing 8 ounces, then waiting 30 seconds, then running a blank shot then wait a full minute then flush 2 ounces then run the blank shot and (you get the idea)

It's not a prescription. Just take a few measurements at different conditions to develop a feel for how your machine behaves and then adjust for taste.

It works for me....

(and now I need to go paste this in the Pallo Tool thread!)

[HB]

Developing a "feel" for my machine's operating characteristics lets me make temperature adjustments "on the fly" and often without thinking about it very much at all.

Thanks Jon for chiming in, you bring up a very good point: Although all this minutia may sound daunting, it really isn't that difficult to develop the skills for excellent HX temperature control. Initially a newbie may be all over the map, getting within no better than a 4F range of their desired brew temperature, but most will halve that after a week or two of practice. I believe it's fair to say that someone who's mastered the routine is within a 1.5F range. Thanks to the jumpstart the thermofilter gave me while evaluating the Elektra A3, I was routinely within 0.5F of the target temperature - when I took the trouble to measure (admittedly that machine makes it easier to nail precise temperatures).

BTW, I was talking to AndyS recently and he commented that he thought I favored HX machines over double-boilers. Given the amount of time that I spend writing and talking about them, I see how he would come to that conclusion, but actually I like both for different reasons. No argument that temperature management of a double-boiler is easier. However, I don't think HX machines are the poor design compromise some would argue. I can twiddle temperatures shot-to-shot on the fly to suite a particular blend without delay, plus I genuinely believe some blends are more tasty with the "humped" temperature profile (it seems that lighter, fruity, aromatic blends fall in this category). That's consistent with Chris Tacy's and Jim Schulman's comments, but to be quite honest, I don't have their depth of experience to speak conclusively (and besides, I don't like speaking conclusively ).

[quar]

Thanks for your replies everybody. Glad to know that I'm probably not overwhelming the machine. After finally managing to get my autofill problem sorted out, I guess I'm still being a little overly cautious that there aren't other problems lurking somewhere. Mostly paranoia that I managed to damage the element before noticing that my boiler wasn't refilling. Overtemp thermo never popped or anything though and it comes up to pressure OK.

Sounds like it is the brightness of the blends that I've been using that I'm tasting and not actually an underextracted sourness. Went back over my order history and I was tasting this in these blends before having any issues. I'll have to try a different flush profile with these blends to see if I can mute this somewhat. Once I get back in town, I'll have to take some temp measurements as well.

Mike

[HB]

I'll have to try a different flush profile with these blends to see if I can mute this somewhat.

Although my preference for brighter blends has grown the last few months, I would try adding an extra 5-10 seconds to the rebound to attenuate the brightness (a "high hump" temperature profile). I believe that a rebound of 25-30 seconds is the upper boundary for the Andreja / Anita before it will scorch the puck at practically any boiler pressure setting.

[malachi]

If you're finding blends to be too bright (not too sour, but rather exhibiting more acidity than you normally like) I would suggest that the best option is to change coffees.
If this is not an option for some reason, I've had far better results with technique changes than with temp changes. Upping the brew temp (even for a percentage of the extraction profile) with the brighter coffees tends to lead to a thin and often harsh flavour note. You might, instead, want to try going with a larger dose and a lower extraction volume.

[quar]

If you're finding blends to be too bright (not too sour, but rather exhibiting more acidity than you normally like) I would suggest that the best option is to change coffees.

Yeah, I'm starting to think that I just managed get a couple of brighter blends. The first is a 50-50 post blend of Harrar Horse and DP Sumatra Mandheling. The other is Espresso Elite from coffeewholesalers. Both are new to me as I usually drink Espresso Monkey Blend or MBG. Thanks for the advice on updosing, I'll give it a shot.


Mike

[hperry]

I've been lurking a fair amount during these discussions and am awed by the temperature stability that some are able to obtain with HX machines. I have a Bezerra BZ40 - which is a larger machine, but shares a strong family resemblance to the Pasquini Livia. Using the Scarce device I can repeatedly hit a given start temperature using a variety of the techniques to maintain a given temperature. From there, however, during a 25 second pull there is an 8 to 15 degree temperature variation from start to finish. I watched the Elektra (I think) movie where there was only a minute variation during the shot. Settings are 1.1 and 9 bar. I suspect this indicates a fundamental flaw in my methodology - but also wonder if the machine itself may be flawed.

Developing a "feel" for my machine's operating characteristics lets me make temperature adjustments "on the fly" and often without thinking about it very much at all.

Thanks Jon for chiming in, you bring up a very good point: Although all this minutia may sound daunting, it really isn't that difficult to develop the skills for excellent HX temperature control. Initially a newbie may be all over the map, getting within no better than a 4F range of their desired brew temperature, but most will halve that after a week or two of practice. I believe it's fair to say that someone who's mastered the routine is within a 1.5F range. Thanks to the jumpstart the thermofilter gave me while evaluating the Elektra A3, I was routinely within 0.5F of the target temperature - when I took the trouble to measure (admittedly that machine makes it easier to nail precise temperatures).

BTW, I was talking to AndyS recently and he commented that he thought I favored HX machines over double-boilers. Given the amount of time that I spend writing and talking about them, I see how he would come to that conclusion, but actually I like both for different reasons. No argument that temperature management of a double-boiler is easier. However, I don't think HX machines are the poor design compromise some would argue. I can twiddle temperatures shot-to-shot on the fly to suite a particular blend without delay, plus I genuinely believe some blends are more tasty with the "humped" temperature profile (it seems that lighter, fruity, aromatic blends fall in this category). That's consistent with Chris Tacy's and Jim Schulman's comments, but to be quite honest, I don't have their depth of experience to speak conclusively (and besides, I don't like speaking conclusively ).

[malachi]

8-15 degree range!
Intra-shot!

Woah!

Ummm... yeah, that seems very wrong to me.
As some have pointed out - I have limited experience with home machines, but still...

Are you flushing and then building your shot (rebound method) or building your shot and the flushing (flush and go)?

[hperry]

I guess the pattern is flush - build - flush, which may be the problem. Typically the machine warms up a couple of hours. I do an initial 6 oz flush - 4 to quiet the water dance and 2 additional. I'm using a Versalab grinder and, grinding into the portafiler, it takes me 45 seconds to a minute to grind, distribute and tamp. By that time the water dance is back. The best stability I've got so far is to flush again - this time just until the dance ends. This usually starts me out around 205 degrees with a steady temperature drop of about eight degrees over the course of a 20 - 25 second pull. I have tried not doing the second flush, but typically we then start at 202 or 203 and climb 4 to 6 degrees.

8-15 degree range!
intra-shot!

Woah!

Ummm... yeah, that seems very wrong to me.
As some have pointed out - I have limited experience with home machines, but still...

Are you flushing and then building your shot (rebound method) or building your shot and the flushing (flush and go)?

[malachi]

Try build, flush, pull shot.
See what you get then.
Then try flush, build, pull shot and see what you get.


If you don't get way better results... well... then something is wrong.

[cannonfodder]

The flushing sequence is almost an art, each machine is little different just as each persons tastes are a little different. Unfortunately, there is no 'golden rule' to flushing. You just have to practice and make note of what you did so when you hit the sweet spot, you can reproduce it.

I do a big cooling flush after the initial heat up, about 10oz, then start my routine. I grind, flush, dose, flush, lock the PF in, dump the water out of my very hot cup and dry it, hold another 5 seconds and go. That put me right at 200f, hold another 5 and I get 201.5, no waiting after I dry my cup gets me around 198.

[HB]

The best stability I've got so far is to flush again - this time just until the dance ends.

Most mornings I move pretty leisurely behind the machine, so it's not unusual that I do a second "mini flush" before locking it. Not to get all Zen on you, but there's a definite intuition for the particular machine. An eight degree drop sounds like a lot, I would expect more like four degrees F. Let me toss out this idea: Let it rebound a second time, but only for 10 seconds. Remember that the HX is fairly small -- on the order of 110ml -- and easily exhausted. Specificially I'm proposing the slow man's routine:

1. Flush to usual brew temperature (~six ounces)
2. Leisurely prepare your basket in around a minute
3. Flush barely two ounces, just a teenie bit past the water dance end
4. Lock in the portafilter. Face towards Seattle and say a short prayer to the Espresso Gods (10 seconds have passed since step 3)
5. Go.

Oh, with all this flush minutia, I forgot to ask: How's the espresso taste?

[hperry]

With the Scace device I have to measure over a period of time rather than actually dosing because it has a flow restrictor that emulates the coffee. What I do discover is that there is an interval after the first flush (perhaps 40-50 seconds) where the temperature stays within 2 degrees plus or minus of a set point between 194 and 202 depending on how long you wait. I can get that at an interval after the first flush. However, after that I will get the ascending or descending pattern. I suspect that it has to do with the amount (or lack of it) of new water in the boiler whether the temperature climbs or descends. I appreciate you taking the time on this as I know this is very old territory for those of you who already have worked these things out.

Try build, flush, pull shot.
See what you get then.
Then try flush, build, pull shot and see what you get.


If you don't get way better results... well... then something is wrong.

[hperry]

Thanks. An initial 8 oz flush takes me down to 180. It comes back to 194 and hovers within 2 degrees of that for a 20 second pull (seems to be optimum time with the Versalab grinder). Because I pull 2 shots for the second drink I cannot follow the same routine for the second shot because the machine has cooled down some. Seems probably I should be concentrating on the second shot - it appears that I can exercise more control on the first than I thought I could.

The flushing sequence is almost an art, each machine is little different just as each persons tastes are a little different. Unfortunately, there is no 'golden rule' to flushing. You just have to practice and make note of what you did so when you hit the sweet spot, you can reproduce it.

I do a big cooling flush after the initial heat up, about 10oz, then start my routine. I grind, flush, dose, flush, lock the PF in, dump the water out of my very hot cup and dry it, hold another 5 seconds and go. That put me right at 200f, hold another 5 and I get 201.5, no waiting after I dry my cup gets me around 198.

[hperry]

Dan,

Thanks! Your routine is very close to the timing I need. Dosing directly into the portafilter with the Versalab, distributing and tamping takes about a minute. Using your protocol I was able to keep things within about 2 degrees at approximately 198 +-1. I think I need to "edge" that up a couple of degrees. At that temperature the espresso is slightly sour.

I pull 2 ristretto shots using the double filter (about 1 oz) for my morning latte. After I get the first shot right (maybe a slightly shorter flush?) then I have to figure the second one out - at that point the machine has not been resting so the flush will be different.

The espresso? There are mornings when I'm really happy - and I achieve a really rich, sweet cup. More often than not I'm a bit on one side or the other of where I'd like it to be. I suspect ultimately a machine with temperature controlled in degrees fahrenheit is in my future. Over the last six months I've improved my tools and learned to manage many of the other variables (dosing, tamping, fresh coffee, grind) pretty well. I'm inclined to believe temperature is the major issue left to get more repeatable results.

Thanks for spending time going over what I know is old territory for you. It helped move things ahead. I really do like a good latte in the morning.

The best stability I've got so far is to flush again - this time just until the dance ends.

1. Flush to usual brew temperature (~six ounces)
2. Leisurely prepare your basket in around a minute
3. Flush barely two ounces, just a teenie bit past the water dance end
4. Lock in the portafilter. Face towards Seattle and say a short prayer to the Espresso Gods (10 seconds have passed since step 3)
5. Go.

Oh, with all this flush minutia, I forgot to ask: How's the espresso taste?

[quar]

Well, looks like it was 1/2 of one and 6 of the other. My suspisions were confirmed that I had a heat issue. The more I thought about it, the more convinced I was that despite the boiler temp, the HX wasn't operating up to par. I descaled and everything seems hotter. Boiler flushes last longer, the group is hotter, and the flush and go tastes good again.

This particular machine has only been in service since Sept and my water hardness is 5 grains according to the test strips Chris included. Starting to think that they were wrong....Need to get a better test kit and a water softener.

Mike

[firedog]

I saw Jim's great video on his HX flushing technique. Thanks to him for it.

My question: does anyone know where I can get detailed instructions on how to attach a temperature sensor to my PF or my grouphead? I saw this in his video and I would really like the feedback so I can get the shot temperature right.

My machine in an Splendora Unica with an e61 type group.

I'm not very technical minded, I will probably need help to do this, but any detailed info about what kind of thermometer/sensor is needed and how to install/attach it would be much appreciated.

Thanks,
Danny