Open source espresso machine is one delicious rabbit hole inside another
arstechnica.com
arstechnica.com
I have a Delonghi bean-to-cup machine. It's quite capable of making decent coffee but also has some unfortunate software behaviours that result in inconsistent output. Most annoyingly, it gives extremely limited control over grind time and thus the quantity of grounds in each shot. If the hopper runs out of beans or you switch to a different type of beans, it gets confused and tries to adjust the grind time on it's own, usually overcompensating and resulting in either too much coffee in the shot (pump struggles to express anything) or too little (thin, bitter, over-extracted shot).
If I could control grind time precicely, it would hugely improve the consistency my coffee. I could set up a profile for each type of bean that I use and fine-tune my shots.
Has anyone tried to hack or customise software on Delonghi (or any other) bean-to-cup machine? Any projects out there to do this?
Bean origin and processing matters a lot but most people don't process beans themselves so people generally have to trust others for this step. However, you will have to grind your own beans. Grind fineness is by far the most important variable to control. The correct grind fineness is determined by achieving an appropriate flow rate given a valid pressure profile (typically 9 bars continuous). The performance of any espresso machine is a direct consequence of the fineness and consistency of the grind. No machine can make up for a poor grind. A crappy machine can often make good espresso if the grind is good. Invest heavily in a good grinder. I do not like to extra spend money for electronics that I could never maintain or fix myself, so I opted for a manual grinder with premium conical burrs. I make espresso with an Espresso Forge (https://espressoforge.com/) which effectively has no moving parts, electronics, and only food grade metal. Just a piston that you ram in by hand.
It gives precise control over the grind fineness, but tries to manage the grind time (and thus volume of coffee in the puck) itself. The software that does this isn't very smart and that's where the problems lie...
If I could control grind time precicely, it would hugely improve the consistency my coffee. I could set up a profile for each type of bean that I use and fine-tune my shots.
Bean to cup espresso machines are good for what they are, but not a good base for making great coffee, and software hacking can't fix the hardware limitations.
Anecdote: at my previous London flat, I had my bean-to-cup machine and one of my flatmates had a fancy manual machine. He knew how to use it, too, having once worked as a barista. But which machine do you think everyone actually used every morning? The manual one ended up getting put away in a cupboard most of the time! (... and my former flatmate still jokes about how much he misses my coffee machine when I see him, haha)
This isn't to say that a manual machine doesn't make better coffee than a bean to cup, in the right hands. But for most people, the difference isn't worth the time and effort required to get there.
> "software hacking can't fix the hardware limitations"
In this case, I believe it would, because the machine does make great coffee when it gets everything right. Specifically the grind time / volume of grinds in the shot. The issue is the consistency, and the software's poor behaviour in wildly changing the grind time when something throws it out of equilibrium (ie: change to different beans, or hopper runs out of beans)
Same for Espresso. A PID that keeps the brewing temperature steady is superior to a thermostat where the temperature ramps up and down constantly and your brewing temperature depends on experience or luck (people who know their machine would call this "temperature surfing" because you will have to wait for the thermostat to switch on). To which degree brewing temperature affects the result can be discussed, but this could be tested quite easily.
So if you are a coffee afficionado, the number one reason you would like to have a PID is probably convenience. One variable less to care about if you wanna get reproducably (good) results.
I'd also prefer the faders of my mixing desk to keep their levels where I set them instead of fluctuating around.
My comment came from a place of love. I’m happy to see people going the extra mile. Myself, I just want to be chill at home and aim for a balanced providing me joy without headaches.
But I’m weird. Example: I love playing vinyls not because it sounds better but because it lacks dynamics and I find that… relaxing!
Reasonable, I am similar in those regards.
I just wanted to highlight that the border between reasonable technical features and audiophile snakeoil is at times not totally clear to draw.
There are e.g. quite expensive audio cables in pro audio as well, but there you pay for reliability and not for esoteric materials.
Everything else is just chasing the final improvements, which is great but I feel like a lot of people get sold stuff they don't need. A £40 hand grinder might change your life as much as a £700 machine you got because reddit said so.
Or, as I see the mountain of stuff required to make high-quality specialty coffee at home, I may do what James Hoffman does and just do drip coffee. There is wisdom in the old, simple ways, and the $4k you can easily spend on a full setup is 1000 espressos made by a professional with bar better, more consistent equipment and tons of practice.
There are a lot of coffee shops near me. I'm surprised at how bad some of the espressos are, a lot too sour, some too bitter. Prices keep going up. You can count the break even point in terms of money, sure, but in my house, the espresso machine paid for itself almost instantly in terms of satisfaction.
But for not much more money you can get an affordable electric burr grinder (like the Delonghi KG79) which will give a good, consistent grind with much less effort than a hand grinder.
Lower grade equipment can make great espresso, but maybe you are managing good results only 1/10 instead of 5/10.
In double blind trials or audiophile I can hear the gold plating sort of way?
I think this type of comparison with espresso is comparing the differences of far different orders of magnitude. i.e. Impact of a bad grinder, poor bean distribution, or tamping force is orders of magnitude larger than the difference cable connectors makes. The difference in the end result is also (and I'm committing audiophile heresy according to some) far more apparent with espresso.
Maybe an analogy here is gold plated wires don't make any difference in a home setup (well, they make zero difference in a digital setup), but when you're AC/DC blowing the roof off an arena with your speakers turned up to 11 bad equipment can be pretty noticeable.
A friend doesn't mind, but he roasts his own beans as well. I've had great espresso from him, and I like it but I'm content with little filter
The new wisdom is grinding coarser, using less coffee, and sending water through at a lower pressure. So instead of an 18g:36g shot dripping out variably in 30 seconds, the aim is something like 15:45g in less than 20 seconds.
[0]: https://strivefortone.com/2020/09/19/low-pressure-turbo-shot...
I guess I should have made it clearer that I really was talking about coffee in general, not just espresso.
Roast level is also a big deal. Dark roasts have a much wider sweet spot on the grinding dial. The beans are brittle so they tend to shatter and produce a lot of fines. As long as you don’t grind too fine that the machine chokes during the shot, you’re going to be able to find the sweet spot.
The lighter you go in roast, the more challenging everything becomes. Lighter roasts are less brittle so the grinder tends to cut rather than shatter. This often allows you to grind way finer on the dial. It also makes the coffee much more difficult to extract, requiring higher temperatures and a narrow grind sweet spot. Loads of people complain of sour, weak extractions with no body. This is where you can get into spending loads of money on a high-end flat burr grinder and a pressure profiling machine with PID temperature controls (and even temperature profiling during the shot).
Light roasted single origin coffee beans can smell absolutely amazing, so people know there is potential there for unbelievably good coffee. All of the disappointment in the cup is what leads to the big spending (or giving up and going back to darker roasts).
Fine as in drinkable? Sure. But if you want delicious coffee, with great aroma, crema, etc, there is a big difference between fresh roasted beans and something that has been sitting on a shelf for a few months.
I have yet to see any assertions in the coffee world that approach the sheer amount of overt dishonesty and grift that characterizes high-end audio.
I don't know if it's a grift but it is absurd
No one is arguing -- or at least I haven't seen it argued -- that a $200 basket, or a $100 distribution tool, or a super-expensive knock box make better coffee that the basic models.
There's just a certain subset of home espresso people who want fancier things, or prettier things, for reasons external to efficacy. How does it look on the counter when I'm not using it? How pleasant is it to hold? Does it tickle that "finely machined object" thing in my brain?
For example, lots of people use the Weiss Distribution Technique on their espresso grounds. Simply put, you stir the coffee in the portafilter using a tool with several fine needles on it to reduce clumping and more evenly distribute the grounds. This has been experimentally proven to produce improvements in the shot.
BUT, critically, it doesn't matter how expensive the tool is. The DIY crowd buys acupuncture needles on Amazon and sticks 'em into a wine cork, which costs almost nothing. That works every bit as well as a $150 fancy one with a hand-tooled wooden handle and matching base, and no one argues differently. It's just that some people value aesthetics enough to pony up for the fancy option.
The question isn't whether it makes a marginally better espresso (it probably does); the question is whether it's the best bang for your buck.
For most people, it probably isn't.
For people who have Weber EG-1s on their coutertops (AFAICT, it was Weber with their Unifilter who started this wave of super-expensive baskets/portafilters?) and La Marzoccos / Decents? It might be!
(I prefer Hoffmann, and he definitely does the same thing -- ie, point out that some given object might be pretty, and it might do a fine job, but not enough of a BETTER job to justify its premium.)
Oh this is easy. - Starbucks, as an org, in whole. Brutal dishonesty in much of their marketing, naming, and Brewing methods. Their Coffee flavor is intended to distort your expectations of 'quality coffee'. - The "Roast Date" disappearance on high-end coffee to sell "Quality" rather than freshness. - Keurig's marketing of quality or that they can make 'Espresso' (it can't), their product is designed for lock-in. - Nespresso Vertuo's fake "Froth"
Nobody who cares enough about coffee to grind their own beans is fooled.
Again, not in the same realm.
The VAST majority of Americans are "Fooled" and would say "Starbucks" or "Nespresso" is "Luxury Coffee". I would argue that the coffee industry is where the Audiophile headphone industry was at around the mid 2000s (Monster Headphones, Beats, ETC) where the public was so fooled and only those who knew what good sound was like, knew it was way off.
Espresso is a lot easier to understand too, it's an entirely physical process, and everyone has had bitter, bad tasting espresso, so it's easier to know what "good" tastes like, vs some arbitrary chart or subjective POV in the audiophile world talking about how your house needs rewired because it puts noise into your setup.
i order green coffee from sweetmarias.com and it's $5-8/lb which makes a good deal more than what's in the $15-25+ 10oz bags from 3rd wave roasters.
plus i learn a new skill and can roast to my preference every time!
Having switched to a local shop's medium roast, I very rarely have to adjust my grinder after buying a new batch of beans.
* Breville Bambino
* Niche Zero
* IMS 20mm baskets for Breville portafilters
I can get within spitting distance of most third-wave coffee shops most days.
Still a lot to learn but I can tell I'm nearing the point of rapidly diminishing returns. I'll probably look to upgrade the Bambino when it dies, but I'm not really jonesing for much more at this point.
https://www.reddit.com/r/FlairEspresso/comments/j53vy0/fello...
Sibling suggested heating the chamber with the kettle, in my experience that has caused the silicon wrapping to deteriorate.
I like these threads a lot, but dream of the open source bread maker for sourdough bread. I’ve been collecting bread makers from junk stores and have a fleet of ESPs. I have no idea what I’m doing but it’s getting dangerously close to having to do it myself.
This is paired with a Breville Barista Express. Unlike most espresso people I am not tempted to upgrade this because I am so used to the workflow and it is quite good for a consumer grade machine (except for its grinder). With the niche I can make something 90% of the fidelity of a good cafe.
https://www.home-barista.com/blog/niche-zero-grinder-review-...
The niche is interesting, and a far saner choice. My ‘70s orange isn’t for everyone, it’s like a traffic cone.
I have a Lelit MaraX and a Niche Zero. I really want to get a lever machine next, but it’ll be awhile before it’s in the budget. Besides, I’m still learning how to use this current setup, having had it for a year and change.
I love the e61 group head though.
I think my current dream machine is an Elektra MCAL. I am drawn to the idea of a machine with no machinery. Just a boiler and pressurestat and the spring lever to make coffee. The sweet sound of silence. Only the rush of water and the pouring of coffee to be heard.
I nearly got a Linea Mini, then discover the amount of electrics. The plastic, the fake lever. It got me thinking about what would last. I want manual. Prior to engaging the pump (pre infusion) you’re getting line pressure and it’s good.
I need to try a lever, and also a real Linea.
The Electra Belle is a ludicrous machine that looks amazing and is even less practical - saw one in the flesh in the middle of nowhere in New Zealand a while back. Old and tarnished and going strong.
My use case is a bit different from most as I don’t often have electricity, but honestly it’s better than most but the highest end coffee shops I’ve been to (across Seattle, Portland, LA).
Yeah the Robot is definitely very stylish but the Flair seems a bit more rugged and practical looking, in a camping way. Excellent to hear that they designed it to pack small and assemble quickly!
That machine requires some real TLC to keep in shape, as the clear coat over the brass will get damaged over time if water/coffee splashes are left on it. Once the clear coat is gone, the copper and brass will tarnish and then the machine needs a professional to restore it.
It’s the same story for their home models but on a much smaller scale, so a bit more reasonable for a loving coffee enthusiast to take care of. I’ve heard you can take the machine to some auto paint shops to get a fresh layer of clear coat sprayed on. People have also said you can repair chips/scratches with some touch up tools at home.
[0]: https://www.kickstarter.com/projects/meticulous/meticulous-e... [1]: https://www.youtube.com/watch?v=N-dxaLZTfGc
Both of these things are surprisingly difficult to do properly, with enough consistency and control. I believe temperature profile can noticeably affect the roast, and being able to consistently hit a profile (and thus experiment and improve) is really difficult with something like a heat gun.
I did the heat gun thing for a while, and the air popper thing for a while, and even modified the air popper to be able to control fan speed and heating element separately. I got good results but hit a wall and needed more control. I started building an Arduino-controlled air popper to be able to get more consistent results, then decided brewing top-quality coffee coffee was challenging enough when roast quality was delegated to experts. Since then, I've been buying from local roasters.
Sounds like you need a popocorn maker :)
My go to is Yirgachefe.
Every so often I buy beans to see what I’m missing and I’m usually very underwhelmed. It’s time I did this again.
New Zealand is lucky as the importers don’t bring in much crap coffee and so the average is good. A large bulk is brought in by just a couple of companies and all seem to have high standards.
The key to home roasting for me is freshness. I like coffee a couple of days after it’s roasted. I can control this and struggled when I was buying beans.
I have such machine and for that price combined with not having to wait for a Faema to heat up, take a rather large amount of space and whatnot, and given that I drink one max two espressos a day, and that those are so close to what a professional automatic machine produces: totally worth it and no regrets yet. I used to dream of an espresso machine producing those super tasty beverages which in my country are really hard to find (compared to Italy of course), now I just have one, and it's maintainance and hassle-free as well and is going to last a very long time (basically only one rubber ring which is replacable). Which cannot exactly be said of automatic machines.
And whatnot. I started turning it off at night (not recommended by manufacturer) and at the same point in time I started charging an electric car that is used daily.
The power bill has gone down.
The electric car uses less power than the Faema E61. Not good.
it's definitely good, but needs love to become "really good"
Pffft... Jackpot! Great machine, especially by the looks!
All the best for your espresso adventure!
I make add flour and water to the starter at night. Add the rest of the ingredients in the morning. Knead in the evening and put it in a basket then bake.
I’d like a bread-maker that I can use for the daytime bit where it warms a little and stirs at intervals I can choose. It needs a pan that won’t have the coating come off after a few months. Every one I’ve tried and have been careful with has done this. I’ve never baked with any break-maker I’ve owned, just used various dough cycles and only my hands go into it, nothing sharp has scratched them.
They hate the acidity or alcohol or something from long ferments.
I don’t even want to bake it in the bread-maker, just control the knead and heat while keeping it relatively environmentally controlled. What I want is more like a complicated stand mixer with a warming element.
I have a Kenwood Cooking Chef (the older Version, has a 6.7L bowl). It kneads really well and has a heating function for up to 180°C (so you can also use it for cooking or continiously heating malt for bread baking). There is a functions where it stirs once every 15s and once every two minutes iirc. So its to frequent for the ripening but im sure there is a way to hack it. Or maybe check the newer versions they have a touch screen and might be configurable from the get go. The newer version also has a built in scale. There is also a "Patissier" version that is a little cheaper and can only heat up to 80°C.
And the bowl is made from stainless steel, so no coating that can come of.
1. add 500g strong white,2 tbl spoons starter, 300ml water, 10g salt to bread machine
2. start dough mix programme (go for longest duration one available)
3.leave for 24hrs or until has doubled in size
4. bake for 70 mins
Not the best sour dough bread. But I think still quite delicious. And really very low effort.
I hadn’t really thought about bread machine with sourdough, because it’s “supposed” to be rustic and old world. But having a box that does most of the work and I bake it the next day would probably increase my sourdough production which has averaged zero loaves for the last year or so.
To summarize, the ZPM was sort of a DIY espresso machine that turned into a kickstarter project to deliver a product. It ended up failing, but the effort informed a lot of stuff that came after.
That metaphor popped up for me in another context recently, too, as I found the usual host of the FortNine motorcycle channel, Ryan, to be the Alton Brown of motorcycles. Lots of people watch his videos that don't even ride because of his nerdy enthusiasm and excellent explanations about aspects of motorcycles and motorcycling.
Hopefully tomorrow, someone inspired by this can get more out of a cheaper grinder, then after that, who know what?
certainly there's a point of diminishing returns above, say, $1500 on the machine, but there's a host of fantastic grinders at $500 to $700 that do a GREAT and consistent job with espresso. If you're handy and don't mind fiddling, you can get by at even lower levels (say, $200).
I ask because I want to know if DIYing kitchen gadgets like this is somehow unsafe or is it possible to build food grade gadgets yourself?
https://www.bpf.co.uk/packaging/plastic-packaging-safety.asp...
To the best of my knowledge it is less about leaching chemicals out of plastic as safety precautions. There's a bit in there about BPAs (endocrine disruptors) though.
For instance right now we are largely replacing baking plastics and nonstick coatings with silicon because it is known not to leech.. but now we are finding that silicon breaks down after 200-1000 baking cycles. There isn't any conclusive data to say when it becomes unsafe, if it does at all..
All that to say, if a plastic was designed to melt immediately upon contact with a heating element and bind to the first thing it touches... I'm not using it for tea.
https://www.fda.gov/food/food-additives-petitions/questions-...
> Yes. Based on FDA's ongoing safety review of scientific evidence, the available information continues to support the safety of BPA for the currently approved uses in food containers and packaging. People are exposed to low levels of BPA because, like many packaging components, very small amounts of BPA may migrate from the food packaging into foods or beverages. Studies pursued by FDA's National Center for Toxicological Research (NCTR) have shown no effects of BPA from low-dose exposure.
-------
FDA isn't "industry", they're the government regulators in charge of food safety of the USA.
So we swapped from one plastic (BPA) with a century of research to other plastics with no research. They're all going to "leach" into our foods because when food comes in contact with molecules, it will absorb those molecules.
If you cook food in an iron pan, the food will have more iron content. If you cook food in a copper pan, the food will have more copper content. Its just how things work. The question of safety is about studying the effects of iron, copper, or... BPA. And as far as I can tell, only low quality alternative-medicine journals seem to ever have anything to say about BPA.
BPA exerts endocrine disruptor action due to its weak binding affinity for the estrogen receptors ERα and ERβ. BPA exerts other effects by activating the membrane receptor GPER (GPR30) and/or other receptors such as the estrogen-related receptors (ERRs).[1]
In the presence of BPA and DES, the ER-binding region of the HOTAIR promoter was enriched by trimethylation on lysine 4 of histone 3 (H3K4) and by H3K4-specific methyltransferases, which are known to activate transcription73 (Table 2). Moreover, developmental BPA exposure enhances H3K4 trimethylation on genes, which are subsequently more sensitive to regulation by testosterone [2]
[0] https://www.mdpi.com/1422-0067/21/16/5761#:~:text=Furthermor....
[1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6864600/#:~:tex....
[2] https://www.nature.com/articles/s41574-019-0273-8
As for your point about FDA not being "industry"
https://www.npr.org/sections/health-shots/2016/09/28/4956945...
(The same FDA also banned bpa in products for children under 3)
EDIT: I know the stupid "Dihydrogen Monoxide" trick. Just because scary chemical words are used in a paper doesn't mean anything. You're gonna have to make an actual point and argument if you want to convince me of anything.
Hint: What does BPA do, and why are you scared of it? In your own words, supported by the quotes you chose above.
The summary of those excerpts are "endocrine disruption and hormone modulation"
If that is still jargon to you, the effects of those processes which have been substantially linked to bpa are reproductive problems, developmental delays, and increased risk of certain cancers.
So you should be able to link a study demonstrating these reproductive problems. IE: X people have less exposure to BPA, Y people have more exposure to BPA, and Y people have said problems.
--------
Traditionally, that's the kind of study that's done. For example, X people are non-smokers, Y people are smokers. Y people have Z% more cancer than X people.
A quickie look through the papers you copy/pasted from are not of this traditional style of research. The links are rather... weak.
BPA is in every person, and fetal and childhood exposure is the most critical period, and cancers don't develop until later in life.
I severely doubt that. Different people have different habits. I go to the grocery store. I can look at my neighbors food choices and see that some people buy cans, other people buy frozen. Any hypothetical BPA-issue will affect the canned people more, and the frozen-buyers less.
A quote from the article you posted. This suggests that older people born before 1957 should have different properties compared to people born after that year, especially if you are alleging that babies are especially sensitive to this plastic.
I know my Grandma is older than 1957. There's probably plenty of people who can participate in such a study of the anti-BPA crowd cared.
https://www.researchgate.net/publication/362167164/figure/fi...
I think that is actually easy to do. Ceramics are food grade, and people have been making those at home for centuries.
For espresso gadgets I think the biggest risk is the heating element.
Like, cmon man, just admit defeat.
Rant over. Yes, I’m bitter.
Edit: to be clear, it’s not the product I want anymore. I just need closure.
Clockwise (1986) - https://www.imdb.com/title/tt0090852/quotes/qt0177777
Beyond the initial charge of the boiler, swapping jars and cleanup, everything was automated. I even had a load cell to pause the boiler when either the jar was removed or reached a certain weight so that was ready for a robot arm or belt.
It's an extremely rewarding process to automate a physical plant like that because so many things can go wrong and there are so many edge cases, weird feedback loops and idiosyncracies of equipment that isn't obvious until you try it with another thing. I used nodered on a raspberry pi for all of the brains, mqtt for coordination and a swarm of nodeedu running tasmota for all of the I/O.
The main challenge is not getting too comfortable and starting to take shortcuts with safety. Pressure vessels, high voltage switching, fumes and flammable gases/liquids can ruin your day.
Obviously, then you're still missing temp control. I guess I just like more simple stuff
Espresso was largely solved without any computers a long time ago.
There is a small community working on various software apps around it. The machine is controlled through open BLE comms. All the control software AFAIK is OSS and tinkerable. The machine itself is top notch, and in my espresso-nerd opinion, beats anything on the market except in appearance and "analogeness".
So if you like espresso and hacking on software, check it out.
Some more info on the protocol here: https://3.basecamp.com/3671212/buckets/7351439/documents/499...
I don't think there are any better "formal" docs.
And it is not very clear why disassemble the coffee machine to assemble another coffee machine that is no different from other coffee machines.
It's all definitely funny. But it's not entirely clear why.
> Or you could do what Norm Sohl did and build a highly configurable machine out of open source hardware plans and the thermal guts of an Espresso Gaggia.
(My emphasis)
You can buy almost all parts for the small Gaggia machines online. A complete group costs around 50 Euro, I think, and you can do with that whatever you want.
Would you prefer CNC milling it by yourself, based on some plans?
Difference is really that the open design could be repeated way before design based on existing part would go out of produciton.
Seriously now, I appreciate the effort. We need this!!
Original post, 120 comments, here ... https://news.ycombinator.com/item?id=35320729 Re-post, 4 comments, here ... https://news.ycombinator.com/item?id=35347638
Ars has fallen a long, long way since they were bought by Conde Nast. Almost every article they put out is this garbage these days. Most of their original content is them sending someone to auto manufacturers junkets to drive cars and then regurgitate the manufacturer's marketing material.