Adjusting PID offset temperature to avoid flash boiling

#1: Post by **threesips** » April 10th, 2020, 3:22 pm

Seeking some advice, please.

The other day I received my first bag of Caffe Lusso Lionshare. They recommend brewing at about 203F at the puck, so I adjusted the brew boiler temp on my Profitec Pro 700. That's the hottest water I've used in the year I've owned the machine. Depending on the coffee, I usually set the temp at 198-201F.

When I turned on the pump at 203, the water was flash boiling, which made me suspect the PID offset was not correct. (It also made me wonder if I've been inadvertently brewing coffee way too hot for a year, but that's another question.)

Thanks to the magic of SEARCH, I found another_jim's very helpful post Getting accurate shot temperature displays on PIDed double boilers without a thermometer, and DaveC's post on the same topic, http://coffeetime.wikidot.com/izzo-alex ... t-settings.

Following their instructions, I changed the PID temperature reading from Fahrenheit to Celsius, found the lowest water temperature that caused flash-boiling, and lowered the PID offset (setting E1 on the Pro 700) by the corresponding amount. In this case, the lowest temperature that caused flash boiling was 95C. The default offset setting on my machine was 14. So I lowered it by 5C to 9. That eliminated the flash boiling.

I then changed my temperature setting back to Fahrenheit. The default offset of 26 (as set by Clive, where I purchased the machine) now reads 16, which I believe reflects the conversion from C to F.

Having done all that, here is my question for those far more knowledgeable than I am: Is it common to have to lower the offset by such a large amount? It just seems like an awful lot, especially since I've never seen anyone mention doing that. This is one of the newer machines with the updated PID and higher steam power, so it was manufactured after the initial glitch with the original Pro 700 that resulted in conversion errors between Celsius and Fahrenheit.

If this is normal, then I will go upon my merry way. Just want to make sure it's not a sign that something else might be amiss.

Many thanks for reading this too-long post and for any insight you might have.

#2: Post by **Kaylee** » April 10th, 2020, 8:01 pm

Your conversion is fairly correct a delta of 5°C = 9°F.

I don't have a PID, but depending where in the system the thermal couple (temp probe) is and also it's calibrated accuracy WRT typical test points (0°C and 100°C) you get what you get. Knowing the offset for your machine is more important then number on the temp display. I treat it like we do grinder settings, most equipment is not consistent between products or in many cases the same model based on manufacturing drift and inherent tolerances.

I guess what I am saying is, as long as it is consistent then it works and roll with that. You could recheck the steam test in some time and see if it is still occurring at the same indicated temp on the PID controller. A more accurate method would use a calibrated thermometer on the free stream water as it left the group head.

#3: Post by **another_jim** » April 10th, 2020, 10:07 pm

I was surprised too when I first ran into it testing the Bianca (my first DB E61). After asking around, I found out that E61 thermosyphons are so finicky, that the flow varies even with the same model machine. This means the relation between boiler and group temperature varies individually, not just by model.The factory sets the offset by model, but the same model may have offsets varying by as much as 10 degrees C. So it's up to the owner of these machines to calibrate them properly. Even LM double boilers have to be individually calibrated, although for them the spread is about 4C.

My guess is that the convection currents in a boiler/group combo are so chaotic that even the smallest variance, even those well withon manufacturing tolerances, creates these differences. But I have no way of proving this.

In any case, I'm glad the flash boiling method works, since it's cheaper than buying measuring gear.

#4: Post by **jrham12** » April 11th, 2020, 8:25 am

After seeing Jim's video, I was curious about this as well. I reached out to Clive (where I purchased my Synchronika) and specifically asked them if they adjusted the PID offset when they bench tested the machine. Their response was "yes" and that when they bench test each machine they adjust the PID offset by using a scace device.
Knowing that, and since I am enjoying the flavors I am getting, I am still planning on doing the flash boil test but have left it alone for now... Just need to carve out a couple hours when we are not going to be using the machine to do the testing! ( I want to make sure I give it sufficient time to stabilize after each temperature change...)

Josh

April 13th, 2020, 8:40 am

Thanks very much for the responses, everyone. Since adjusting the offset all is well. I was concerned that lowering the temp by such a large leap would introduce sourness, but that hasn't been the case.

#6: Post by **Me2** » April 14th, 2020, 3:08 pm

My question for the more scientifically oriented and elevationally challenged is...

If someone is at higher elevation and flash boiling occurs at 198F or ~92C, is the relationship between boiling and brew temp linear?

Let's say in the case of Lusso GMC, recommended brew temp is 203F or 9F below boiling, can/should one extract by that same difference vs relative boiling point or 198F-9F=189F?

Believe I've read all the elevation related posts and threads. Can't remember seeing anything related to this.

#7: Post by **another_jim** » April 14th, 2020, 3:14 pm

People living at high elevation use absolute temperature, not offset from boiling. For instance, if a blend recommends brewing at 202F, that's what they will use. They just use the changed boiling points to properly calibrate their pstats and other controls.

However, nobody has actually tested this. Coffee brewing does not require phase changes like baking, so it doesn't seem likely that the changed boiling point will change brewing recipes like it changes baking recipes. But stranger things have happened, so who knows?

#8: Post by **Me2** » April 14th, 2020, 3:36 pm

Thanks.

Boiling is 201-202F on avg here and I've found if I go above flash boiling brew temp of 198F on my e61 db, my extractions suck. I've found 193-197F, depending on bean, yields good results as does pulling shot a second or two short. As soon as the water leaves the group it is subjected to whatever ambient environmental factors exist, so if you're above boIling it burns the sugars in the shots and negatively impacts flavor. When big storms roll in off the continental divide, the pressure drop can change extraction temps by 2-4F and totally mess with steaming milk.

#9: Post by **another_jim** » April 14th, 2020, 8:00 pm

Yeah, that's nosebleed altitude. If the coffee boils after exiting the puck, you're screwed. It can be above boiling point at the top of the puck, if the pressure is high. But with the E61 soft preinfusions, you might be getting boiling water above the puck in the period before the pressure ramps up. That probably won't do the shot much good; and could be the reason why you've had to drop the brew temperatures down that low.