Is body of espresso on the DE1 thinner?

June 7th, 2019, 11:50 am

smite wrote: I will say that while I can pull great shots on the DE1 I do notice a difference in the body of the shots on the DE1 overall versus other machines, so it might be worth taking that into consideration. Additionally, the promise of software updates and feature improvements is nice but in reality since I have received the machine they have dramatically slowed down to a serious crawl on the DE1 over the past year.

If you have additional specific questions please let me know.

Point of clarification, the body of the shots on the DE1 are thinner? If so, any theories as to why?

...split from Profitec Pro 700 or DE1? by moderator...

#2: Post by **CwD** » June 7th, 2019, 12:03 pm

The DE still works as a simple and reliable way to make good espresso. You can set it to behave just like the Profitec would and only rarely change it to make slight tweaks to difficult coffees, or never change it at all. Even if you don't change anything ever and keep it as an e61 mimic, it's a better machine. You still get a warmup time so fast it's done heating up before you're ready to use it (and built in timing if so fast it's hot before you're dressed isn't enough). You still get both volumetric and gravimetric shot ending. You still get linearly heating timed steam that can hit exactly your desired temperature based on the input weight. You still get the ability to change shot temperature between two different back to back shots with no waiting or flushing routines. You still get graphs that can tell you if any invisible channeling happened, that can be turned off if you just find them distracting. And, hey, if you ever do find yourself thinking "I wonder how this might taste for Mike on his Londinium", you can set it to mimic that for a little while.

thirdcrackfourthwave wrote:Point of clarification, the body of the shots on the DE1 are thinner? If so, any theories as to why?

The same reason tube amps totally produce richer sound than solid state amps + software tube mimicry. They don't unless you get to look at the tubes glowing. (Now, choice of baskets, profile, etc will make a difference. But you can throw any 58mm basket in the thing. I am also testing a method of objectively measuring body to see if it can be used to pinpoint things that change body.)

#3: Post by **smite** » June 8th, 2019, 12:24 am

thirdcrackfourthwave wrote:Point of clarification, the body of the shots on the DE1 are thinner? If so, any theories as to why?

There are some ideas as to why this is the case but I want to be really clear that by no means is my comment intended to imply that the shots are not consistently good with proper puck prep.

As to why, there has been some speculation on the Decent forums that the large distance to the coffee in the basket and the resulting puck integrity during the shot pull has some impact so reducing that distance either by using a shallow basket and more coffee seems to help. Additionally some people with the help from the Decent team have developed and 3d printed spacers that reduce that distance also with some improvement.

#4: Post by **NelisB** » June 8th, 2019, 2:40 am

CwD wrote:The same reason tube amps totally produce richer sound than solid state amps + software tube mimicry. They don't unless you get to look at the tubes glowing.

That's a bad example. Tubes give even harmonic distortion. Human ears love those. They sound natural. Those harmonics can be measured with analysing equipment. No magic.

Subaru WRX vs. Lamborghini Aventador might be a better comparison. Both fast cars but experience is somewhat different.

June 8th, 2019, 5:12 am

You can create the same sounds tubes make in software. I'm not saying tubes aren't different or tubes don't sound better to some people. I'm saying mimicry is equally good at the worst, and better since you can adjust it. (That's not to say tubes aren't interesting and fun, I actually love vintage audio gear for that, but they're straight up worse than mimicry for performance and I've no interest in tricking myself into thinking there's any advantage outside fun)

It's the same kind of thing with the DE. It can be an e61, if you love what e61s make. But it can also be tweaked if you ever feel the desire to.

#6: Post by **NelisB** » June 8th, 2019, 8:07 am

Thanks Luca for your post.

What do you think of the body the DE1 produces? Do you experience it as thinner?

June 8th, 2019, 9:10 am

Good question; never really thought about it and I'm probably not a good person to ask! I always try to maximise sweetness, flavour and aroma and minimise bitterness; I'm not really fussed about body/mouthfeel, since it's almost always good enough for me. I haven't played around with the lever machine profiles on the DE at all yet, which I'm sure is what people would do if they were really into body.

I don't think I really have any good information that's enlightening on this. If you care about body, the roasts you buy are probably more towards the nutty to pyrolytic bit of the SCA flavour wheel in terms of your roast preferences and are probably most at risk of overdeveloped roasts, whereas I tend to buy stuff more in the grassy to nutty part and am most at risk of underdeveloped roasts. These are very different coffees.

Others have certainly observed they feel other machines are better and I think I read that one of the things that Decent are working on is updating the software to enable higher flow rates, which some speculated might be a way to get higher body. So if it's something that's important to a prospective buyer, they should certainly read the reviews with this in mind.

Tomorrow morning, I might see if I can drop the temp down, fine the grind a touch and hit a good extraction at 1.8-2:1 ratio and see if that sheds any light on the subject. Is that the sort of ratio that high body seeking folk like? And higher doses?

#8: Post by **NelisB** » June 8th, 2019, 10:25 am

luca wrote:Others have certainly observed they feel other machines are better and I think I read that one of the things that Decent are working on is updating the software to enable higher flow rates, which some speculated might be a way to get higher body. So if it's something that's important to a prospective buyer, they should certainly read the reviews with this in mind.

Tomorrow morning, I might see if I can drop the temp down, fine the grind a touch and hit a good extraction at 1.8-2:1 ratio and see if that sheds any light on the subject. Is that the sort of ratio that high body seeking folk like? And higher doses?

I don't think a high flow rate is needed for a big body. On this video DenisB extracts on a Synesso which I believe is not a typical high flow machine. https://www.instagram.com/kafatek/p/Bvo ... lkyorc0l0b

#9: Post by **luca** » June 8th, 2019, 5:58 pm

I have another post for y'all to read. The bad news is that it's not very useful. The good news is that at least there's a lot of it.

I am reporting in on my breakfast shot (DE1Pro + EK43). In summary, I think it's probably less viscous than what the roastery that sells this particular SOE would make (Robur 83mm + Linea PB AV), but it probably has more flavour and intensity.

It's worth pausing to reflect on what we mean by "body". I think that people commonly mean something like viscosity; physical texture. People usually don't mean strength, which I guess is intensity. I can't measure viscosity, but I can measure strength and my shot is at about 11.0%, so I suspect that's plenty strong. This particular coffee is a new crop Kenyan and the roast level is pretty great. It is roasted and sold for espresso, but it's not crazy bitter and full of nasty pyrolytic flavours like most "espresso" roasts in Melbourne; it's kind of just pretty well developed. My breakfast shot was sort of falling over itself with berry type aromas and flavours; blackberry, blackcurrant and a little bit of citrus; probably a little more orange than grapefruit and I didn't have to wince at a roasty aftertaste. The coffee itself is probably a legit 88 points as a minimum and might even crack 90. If I get a chance to stumble by the roastery, I'll see if they have it on and I'll see if I can grab an espresso and report back.

I now realise that none of this is very useful, since I haven't really done much to try to maximise body; it might well be that with a profile that mimics a machine that produces lots of body does a better job, but I don't really want to experiment with that, since I'm unlikely to prefer the results and it's different enough that I think there's a good chance that I'll have to throw away a shot or two to dial it in. Sorry! However, some of this is kind of useful in terms of my thought process and approach, which might help to illustrate how you can use the machine.

So, as I posted yesterday, my objective was to find out what a shorter shot tastes like and I wanted to use 20g so that I ended up with enough volume to top it up with milk for my first coffee of the day. I had been using less coffee with paper filter above and below, so the grind setting that I was on wasn't really going to cut it. I had been experimenting and had saved a lot of data on this coffee using the "god shot" function, so I went into that and found the grind setting that I had previously used for 20g on this flow profile and I went a little bit finer. Then I went into the profile editor and reduced the target flow rate from 2.2ml/s to 1.8ml/s and reduced the brew temperature 0.5c at each step of the profile except the soak (where I have a drastic undershoot programmed). I then pulled the shot and used an espresso cup to catch about the last 10ml after I hit the 2:1 ratio (ie. 40g). So I extracted 50g total from 20g of grounds.

Here is what the shot looked like physically:

Here is what the shot looked like in DE data:

To read the graphs, you need to know that blue = flow rate, green = pressure, pale = reference shot (the one I mentioned above that I used to guess at the grind setting) and solid = current shot (ie. the one this post is about). The pressure peak and drop off after the soak phase is a shape that's about right for this grinder and this pressure profile, so I more or less got the grind in the zone and probably didn't screw up the puck prep that much. Usually, the pressure peak height tells me how bitter and astringent a shot is going to be (higher = worse/more). On this coffee, the higher 12 bar peak of the reference shot was OK in that regard, so I probably have a little scope to grind finer if I want. Finer grinds tend to have a lower pressure drop off towards the end, which is to say that they maintain the higher pressure for longer, and that the derivative of the pressure after the peak is probably a smaller negative number than coarser shots. It's kind of like a "ROR" curve in roasting, so I look forward to someone coming up with another jargon acronym that shouldn't exist that we can learn instead of just calling it the pressure derivative (or the bean temp derivative or dy/dx of whatever). I think at the moment you can enable dy/dx graphing for flow, but pressure is in the works. Since I am usually using flow profiling, dy/dx of flow is not useful, so I haven't experimented it. Maybe I'll look at doing that for when I switch to a pressure profile for darker roasts. Anyway, all of that is a bit of a digression, but it kind of illustrates that the graphs get to be useful after you have stared at them for a while and correlated them with the results in the cup.

Here is the refractometer working out for the main shot (ie. the 40.2 g of 2:1 ratio):

Here is the refractometer working out for for the extra bit caught in the espresso cup:

Now the next bit of taste information that you need to know is that I tasted the extra 11mL that would represent the extra .5 to take it from a 2:1 extraction to a 2.5:1 extraction and ... it was still great! If you've split shots up into multiple bits, you've probably gotten to the stage where you realise that if you keep on going, it doesn't tend to be all that bitter or anything, but it is pretty flavourless and dilutes your espresso. You can see that this had a strength of 3%, so about double, maybe a little more, the strength of a filter coffee brew. It still had a fair whack of blackcurrant and blackberry and negligible bitterness.

All of this illustrates the tension between flavour, strength and viscosity: It's pretty easy to go to a lower ratio extraction and get more of the latter two at the expense of less of the former. I didn't actually measure the 2.5:1 shot resulting from combining the two, but my back-of-envelope maths puts it at 9.3% strength instead of 11%, but 23.7% extraction instead of 22.1% extraction ... every percentage of which was delicious.

So this illustrates the tradeoff: would you sacrifice ~15% of strength for ~7% more flavour? For this coffee, I would every day of the week: the minimum 88 pt score that I reckon a Q-grader would give the green is largely due to the flavour and not really body. With this coffee, that is what you are paying for. But what if what you really want is a thick, rich, viscous shot? That's a fine personal flavour preference to have, but don't buy this coffee for it: buy something more suitable. If you don't mind the dried hay of a monsooned malabar or the rubbery aftertaste of a robusta, go for that. If you do, grab a nice natural or pulped natural brazil or similar. I don't often buy these sorts of coffees, but if someone wants to get some to me, I'm happy to pull a bunch of shots to try to optimise them and report back. As always, having the right coffee for the job is probably the most important thing!

Sorry, 99% of that post wasn't really that responsive to the question, but I hope it was otherwise helpful to people, or at least interesting!

#10: Post by **another_jim** » June 8th, 2019, 7:45 pm

I think viscosity is a thick tongue coating quality of the crema, not body per se. Really viscous shots remind me of chocolate truffles, which are a kind of foam. This kind of viscosity is one of the reasons people fall in love with espresso in the first place; and it's been getting short shrift in both barista competitions and in third wave coffee thinking.

It would be nice to have a good taste based definition for viscosity; so that independent tasters could sort a set of shots identically on a scale for viscosity. Without that, there's not much chance of getting either reliable recipes or a good science for viscous shots. For now, the best I could do for a high viscosity recipe is 50% DP Brazil, 20% Robusta, with 30% of whatever type of coffee you like, pulled espresso-porn style. And while that will be viscous; it's not going to be very good in any other department.