How tinkerers brought PID temp control to espresso machines
home.lamarzoccousa.com
home.lamarzoccousa.com
I have a PID'd Silvia (https://www.seattlecoffeegear.com/pid-retrofit-kit-for-ranci...) and it does make a difference. But man, the black hole that is making repeatable excellent espresso at home doesn't end there. You start wondering about your tamp pressure, need a $700 grinder to have consistent adjustable grinds and start roasting your own coffee.
It is an expensive hobby, but a tasty one at least.
If you pass through Seattle, make sure to drop by Vivace, they define what proper espresso is.
When I was in Seattle this spring I made a tour of Vivace, Victrola, et al. Really fantastic stuff going on there. Here in the bay area everyone seems to like sour notes. I feel I beat out all the local shops with my home setup. Schomer definitely beats me out, though.
PS: Most people fail blind taste tests with this stuff. But, just like the high end audio market spending money on junk makes some people feel good.
The espresso process is EXTREMELY finicky and highly sensitive to initial conditions. Recipes vary from bean to bean, machine to machine, and based on personal preference, but they all generally specify the same 3 things: the mass of coffee going in ("dose"), the mass of the resulting shot, and the time for the extraction to complete.
I typically am pulling shots with Madcap's Third Coast blend. In their shop they target an 18.5g dose yielding a 31g shot completed in 25 seconds (with no pre-infusion). The dependent variable here is the shot time. Anything +/- 4 or more seconds is dumped in the sink as completely undrinkable. The difference is not at all subtle - the shots are awful even to a novice palette.
The reason we have PIDs (and rotary pumps, heated group heads, calibrated tampers, doses measured to the 1/10th gram, etc) is to control every last variable that will impact this process. Every variable here can and will create wild swings in the resulting shot time, so we fix everything we can as accurately as we can to avoid dumping $1 worth of espresso down the drain. With everything else fixed, what we then control as the independent variable is the size of the grind.
This then is how espresso is tuned to land within the time given by your recipe. Grind finer (smaller particles), the process will take longer. Grind looser (larger particles), the process will run shorter. Anything more than 3 seconds one way or the other is another dollar bill down the drain. Very small changes in grind size can yield very large changes in the resulting extraction time. Every time you miss the mark, another dollar bill hits the sink.
As a result, grinder performance is probably the single most important piece of the espresso puzzle, and it's easy to quantify the difference.
It isn't an ego thing, it is a consistency thing. People go mad about getting higher end grinders, machines, etc... because they are trying to control for all possible variables. Espresso is incredibly finicky about those variables, but you can control for them with the right equipment.
It is pretty easy to demonstrate that the higher end burr grinders have greater adjustability and consistency than the lower end ones. It is also totally obvious that they are needed on any higher end machine.
I used to think it was all bullshit too, then I got a Silvia, started with a blade grinder, then a low end burr grinder, then a rocky before I could finally get the grind both fine and consistent enough to pull a proper shot.
It isn't a subtle difference between a good shot or a bad one, anybody can tell the difference.
First off only ~1/3 of coffee drinkers take it black which masks a lot of taste variation.
Second, taste sensitivity varies widely. https://en.wikipedia.org/wiki/Supertaster
Put together well under half of coffee drinkers notice subtle differences in grinding methods. Assuming the devices are setup to similar grinding levels and use similar source beans.
Note: You can find examples on both sides of this issue. But, the blind taste tests that repeat over time fail to provide the expected consistency.
PS: There are a few people that both notice and care about these issues, but they really are the minority of coffee drinkers.
Espresso is brewed under pressure, if you don't have a consistently fine grind then the water will pass straight through without extracting any of the flavor. If your grind is too fine, then it will block the machine and you'll get a bitter extraction.
Again, this isn't a subtle thing. If you've had a real espresso machine (one that has a pump that applies the proper pressure) then it immediately becomes obvious that you need a consistent grind that you can adjust.
The variable you track is generally the length of the extraction, ie, how long it takes to get an ounce or an ounce and a half of espresso for a given amount of time. (aiming for 16-22 seconds or so) Too fine a grind and you'll block it up and you'll never get the oz, too coarse and it will all come out too quickly. You often need to adjust the grind minute amounts every day based on local temperature, humidity and the age of the beans.
This isn't audiophile stuff, it is obvious to anybody who makes espresso and actually wants it to taste good. (ie, isn't drowning it in milk, sugar and syrups)
I couldn't match the quality of my local Dunn Bros coffeeshop until I started asking them to grind my beans when I bought them. Switching from my cheap blade grinder to their in-shop one made the largest difference in taste.
Also, time sitting around post grinding is a large factor.
Yes, time is a factor, but a week after the beans were ground in store, they still tasted better than mine, so...
why didn't companies in the field just implement this as part of their product development(maybe aiming for the higher end) ?
For the home market, lots of machines are not heat-exchangers and therefore need to run a lot cooler ( 100c instead of around 120c ). A pressostat does not work well in this range and they just mount cheap thermostats.
Disclosure: mine.
The meCoffee can be installed in practically every espresso machine which has no electronics of it's own. It is Faston(tm) connectors all the way down...
But boy was it a lot of work, your home-roasted beans have to be the right freshness, the milk has to be super fresh and your tamping technique has to be spot on.
Now that we have good coffee in Austin I cant see going back to doing it myself.
Don't get me wrong, Austin has some great shops. However, I feel like there was such a strong move away from the "burnt Starbucks norm" that all of the good shops feature extremely light roasted coffee that, IMO, doesn't stand up in milk drinks well and is usually way too acidic for me. If you know of a place that serves a great medium-roasted, chocolatey/nutty cappuccino, would like to find it!
Roasters became accustomed to the sourness of the coffee, to make each new coffee taste "right", they had to make it more, and more sour. For those of us who didn't consume coffee in quite the volumes of the professionals, it didn't go over well.
Maybe not as accurate, but a simpler explanation: http://plchowto.com/plc-pid/
For steam boilers, pressure stats are actually quite effective, especially as they get larger.
* Denver is at 5,130–5,690 ft (1,564–1,731 m) [0]
* The boiling point of water at this altitude is between 201 and 203 °F; at sea level it is 212 °F [1]
* "Your brewer should maintain a water temperature between 195 to 205 degrees Fahrenheit for optimal extraction. Colder water will result in flat, under-extracted coffee, while water that is too hot will also cause a loss of quality in the taste of the coffee." [2]
[0]: https://en.wikipedia.org/wiki/Denver
[1]: https://en.wikipedia.org/wiki/High-altitude_cooking#Boiling_...
http://www.freshcup.com/high-altitude-brewing/
http://sprudge.com/in-colorado-delicious-meticulously-brewed...
You can use the PID to maintain the temperature of the wort at ~65C by passing the wort through a heating element and a temperature probe (this is called Recirculating Infusion Mash System).
http://www.keesvanderwesten.com/mirage-special-features.html
Awesome use of a PID loop.
That's a graph from La Marzocco's internal documents and represents the total heat stability architecture of their machines, including boiler and grouphead design, preheating, tuned PIDs, etc.