In the endless search for better coffee I have been experimenting with brew temperatures again. My starting point was Mr Schomer's 202.5 degrees F. What I didn't know was the temperature offset from inside the boiler to the grouphead, so all temperatures quoted are from the boiler. (My machine is a PIDed Linea)

(My palate is not particularly refined, so please pardon the imprecise taste descriptions.) Taste resuts were:

202: not good, probably sour.
205: quite good.
206: not as good.

I do not have access to a Scace device, so I have made one from a thermocouple set into a portafilter. I then tried to measure the temperature in the portafilter and get close to 202 degrees there. This seems to require an extreme offset of about 13 degrees. This may in fact be correct, because the thermocouple in the brew boiler is set in the old La Marzocco thermowell, so I could be losing a few degrees there with the various junctions and mediums. I tried this much higher temperature:

215: Definitely not sour or bitter, quite smooth, but not strong, probably weaker than the lower temperatures.

I suppose my questions are along the lines of:

The coffee I got at the much lower temps was not good, but drinkable. I would have thought that if it was 10 degrees too low, it would be pretty nasty; why isn't it?

On the very sketchy test results above, does it sound like the highest temperature is most likely to be close to an acceptable one? (I realise people are going to say "It depends on your beans, blend, etc, etc")

Does the fact that the coffee tastes smoother and not as stong at the higher temp sound right? If I were way too hot, would it definitely taste bitter and unpleasant?

Any help would be appreciated.

I'd say that your makeshift Scace is probably not reading all too accurately, and I am not sure how much meaning you should expect to derive from the numbers you see there. Question: at '215,' do you see flash boiling? If that's even slightly accurate you should be getting lots of steam coming out of the group when you brew. It should also be burning the crap out of your coffee and making it overextracted, smoky and revolting.

That said, you nailed it-it's tough to make meaningful generalizations when the only details you've got are probably-inaccurate temperature readings. Might help to say what coffee you're using, and assuming it's something few of us have tried, you might mention origins and roast level if you know it. Other important details would be what kind of dose and volume (or brew ratio) you are using when you pull.

Off the cuff it sounds really weird to me that you'd get a "smooth" shot at "215F", though considering that you got a sour shot at "202F," I would guess think that your makeshift thermometer is reading much higher than it should. Have you calibrated it in a steambath?

One of my theories is that the brew PID is underreading, so that when it says 215 it is more like say 206, and with a 4 degree temp offset from boiler to brewhead, "215" in the boiler is 202 at the grouphead.

No, I'm not getting flash boiling at the group, which leads me to be more confidant that the 202 at the brewhead with the makeshift Scace is more believable.

The beans are Allpress Rangitoto roast (from their site: Brasil, Colombia, Guatemala and Sumatra origins roasted in the classic Milanese style. Full bodied with balanced, complex flavours and a long finish. Allpress are the La Marzocco agents in New Zealand.) I am using a La Mrzocco single basket, and aiming for 30 ml, and pur time is about 30 seconds.

I haven't calibrated the thermocouple in a steambath; I will google and try that.

NickA wrote:I haven't calibrated the thermocouple in a steambath; I will google and try that.

Oops, I meant to add this link when I asked you in the first place. Remember you're not looking for 212 per se, looking for whatever the boiling point is at your altitude/atmospheric pressure.

Hi Nicholas, yeah, I wasn't having much luck googling steam bath; most of the links are for people bathing in steam.

I had seen that picture before but I'm not too sure what the setup is. I am guessing that it is a pot with some water in it, with a mesh/sieve to raise the thermocouple out of the water so it is positioned in the steam; is this roughly right? Do I then just get the water to boil, and then see what the thermocouple reads? Does the pot lid need to be off so there's no additional vapour pressure? I'm trying desperately to remember my highschool physics.

Yep, you've got it. I think you want the lid on so that it holds in most of the steam but you're right, I believe if it has trouble soft-sealing as water condenses around the lip, you will need to seat it crooked on the pot, as you're right that you don't want any pressure accumulation. I'd defer to Eric S. to say if that's a realistic concern, though--I don't have the experience!

Well, I've tried the steam bath test. The lid appears to be fairly critical to the results, so I'm not sure how much of a seal is required. My thermocouple/meter setup reads about 209.5/210 degrees F, which is pretty much spot on as we are about 330m high.

In the link that Nicholas provided, I had the cover off the pot in order to make the pic useful. The cover is on during testing but has two 1" vent holes plus the typical loose fit lid. The pressure in the pot is negligible. The temperature that you should be reading depends on both your altitude (which you acknowledged) AND your current barometric pressure.

As regards your temperature and taste buds, I offer this:



Naturally, the "X" axis can be shifted horizontally and the curve can be made "fatter" or "slimmer".

In general: Higher temperatures, above 93C as measured at the top of the puck, favors bitter flavors, while below 93 favors sour ones. Very low temperatures, below 90C, flattens and sweetens the taste, like toddy coffee. This is only useful for making palatable shots from roasts that are too light or too dark for espresso.

Now the pesky details: once your temperature is roughly right, the barista variables, grind and dose, shot time and flow, trump temperature. An improperly packed puck that channels, kills all shots regardless of flow, temperature, pressure etc. I'd lay 10:1 odds that if you perceive a big difference between settings of, say, 202 and 203, the reason is that your dosing or packing has changed between shots, rather than such a small temperature change producing a large taste difference.

To put it more bluntly: I mostly disregarding people's diagnosis of temperature woes on new equipment. If they haven't yet learned how to dose and pull consistently on it, they will wrongly attribute the changes caused by their unsteady skills to the machine. So before you pull out the gauges and wrenches, make sure you're aces with grinder and tamper.

I suppose the 2 basic questions I'm asking are:

If the coffee tastes a bit sour at temeprature xF, and tastes smoother, weaker, but definitely not bitter or sour at a temperature x+10F, then is temperature x+10F more likley to be around 202F?

Would coffee at brewed at 212F be poor?

I realise there are a hundred caveats and variables; I'm talking in very broad generalisations here.