"Espresso! My Espresso!"
An Ongoing Internet Novelette
by Randy Glass - Copyright 2009 - All rights reserved
E-mail me at email@example.com
Diagnosing and Solving Bad Extractions
One More Example
Sunday, April 5, 2009
Over the last two weeks I have been in the middle trying to solve a problem the cause of which I had not been able to identify, but this morning I believe that I have found the solution. I offer this to you as one more example of troubleshooting problems in making espresso and why you cannot take anything for granted.
A common question or problem is, "I have done everything exactly the same- used the same dose, the same tamping and distribution, the same temperature, pressure, etc., and all I changed was a different coffee and I am having problems.
THE BACK STORYFor the last 6 or 7 years I had been getting my green coffee from the same local supplier- a small commercial roaster who sold to me at a very fair price, and sometimes even gave me some green. While I had gotten better coffee in the past, the price was so good and the quality quite acceptable so I continued on with this supplier.
About one or two years ago the company was sold and the original owner semi-retired. I established a rapport with the new owner and continued purchasing there until he bailed out of the business and it was permanently closed and the equipment sold off.
Now what? I was personal friends with the original owner, and his widow gave me the name of another local roaster whom I called and he told me that he was glad to support hobbiest home roasters such as myself, and after talking he agreed to sell me coffee at his cost since the amounts I buy were so relatively low. I, of course, talked coffee with him for a bit and gave him my Espresso URL.
THE COFFEEI made an appointment to meet him at the shop where the roastery resides and I bought a selection of 4 varieties and we had a very nice talk abut coffee, his business, roasting, and all the sorts of things two coffee guys talk about. I even brought one of my Hottops with me to show him, and we roasted a small sample batch in it... so small a batch that it was too small. After I extinguished the flaming chaff-tray handle caused by the chaff-tray fire (totally my fault for roasting such a small batch) we talked some more.
THE BLEND AND ROASTI got home and put together a blend similar to what I had been using with the other supplier's beans. This "new" blend was, iirc, 50% Brazil, 25% Bolivian, and about 25% Kenyan. Pretty standard fare. I roasted it as I had been doing with all my batches recently. Allow roaster to hit about 225, adding beans, bringing it up to abut 300 and holding there until beans turned tan in color (lost all green color) then ramping up to beginning of 1st, lowering the heating element power and allowing the beans to cruise through to a fast second and ejecting manually.
THE PULLSI had done two batches of the new coffee with the above roast profile but had been having some of the poorest pulls I had experienced in years. While the espresso was improved from the first pull with the new beans, the extraction was odd. It started with multiple streams in a doughnut pattern on the bottom of the basket (using a bottomless PF), then joining together in one stream after a good 10 seconds of extraction, then after about five more seconds the flow looked like the "Mushroom" part of the E-61 group, looking like an inverted bell hanging off the bottom of the basket, and very thin in viscosity. Sort of like the portafilter was trying to blow a bubble that would not release into the air. While I had been getting good pulls producing over 2 ounces of crema with the old coffee, I could not get this to happen with the new coffee.
Mind you, even with this poor extraction performance the espresso tasted better from the very first pull than it had with my old supplier. But what was the cause of these abnormal extractions?
My first thought was the new roast followed closely by something about the coffee, but the overwhelming thought centered on the solution. I was using the same grind range, and the same dose and distribution; the only change had been the coffee. But again, what was the solution? I tried a fairly wide range of grind settings, far more than I had previously used with the Kony. This did not solve the problem. The coffee tasted better so there was no going back. I even removed the shower screen and cleaned it to be sure it as flowing properly; it was.
THE SOLUTIONThe solution was.... change the dose. This morning I "down-dosed." I do not want to call it under-dosed because that would infer I used less coffee than required. I cannot quantify the amount of coffee because I do not have a 0.1g. scale, but using less coffee I could judge the amount by eye and how far the tamper went into the basket. The result was two wonderful, back-to-back pulls that tasted about as good as anything I have produced.
CONCLUSIONIf something is wrong, change what you are doing (The "DUH" Factor). I changed coffee and a change in the extraction for the worse was the result. What to do? I changed the grind setting- that didn't work. I changed the dose- that worked. Taking any portion of our preparation for granted can lead to some long, frustrating mornings.
THEORYThe new coffee is of a higher quality and, more importantly, fresher. My new bean supplier takes his coffee very seriously, actually going to South and Central America, visiting farms and farmers. He also cups the coffee before purchase and is particular about what he buys. The old supplier got quality coffee, but was more interested in the price than getting top-notch coffee. The result was that much of his coffee was older than one would like.
So I think that this new coffee expands more in the extraction process, and if I use the same dose as I was before it causes fractures and channeling. It may even have been that the expanding coffee hit the screen in the center with more pressure and caused a decrease in the flow of water in the center of the puck; a clogging in that area which led to the doughnut extraction. I had examined the pucks postmortem, but it revealed nothing, putting one more data point towards puckology being a fairly worthless science.