Equations and boiler pressure - Page 4

Want to talk espresso but not sure which forum? If so, this is the right one.
jedovaty (original poster)
Posts: 537
Joined: 13 years ago

#31: Post by jedovaty (original poster) »

Greg: Exactly, that's what I discovered when the display had 212F at room temp haha. So I've changed the code to "heating" until the vac breaker locks in and pressure begins to build.

John: Yup, would be nice to know, but not critical since I will have a few reference points to work off of. Remember from the first post I wasn't too concerned about precision. That said, curiosity does drive this so I will probably install a thermocouple when it comes time to replace the group gasket (might be a few years?) and then we can have a nice comparison unless someone beats me to it.

Jeff: agree, and why not :)

I ran the machine to 32psi today, and it made espresso. Sadly, I had a couple typos in the serial output so the data is screwed up.

jpender
Posts: 3905
Joined: 12 years ago

#32: Post by jpender »

I guess I glossed over the part where you said that accuracy wasn't important and focused on your interest in a better formula than the one you'd found.

If the actual water temperature isn't important to know then why even calculate it from the pressure? Just figure out what boiler pressure range makes good espresso and use that as your metric.

But my question would still apply if what is true for heating moka pots is also true for espresso boilers. That is, there may not be a one-to-one correspondence between boiler pressure and boiler water temperature because of the time lag for equilibrium. Or maybe not. But that was what my question was about.

jedovaty (original poster)
Posts: 537
Joined: 13 years ago

#33: Post by jedovaty (original poster) »

jpender wrote:If the actual water temperature isn't important to know then why even calculate it from the pressure? Just figure out what boiler pressure range makes good espresso and use that as your metric.
Lots of reasons! A few quick ones: easier to relate to, more widely accepted terminology, helps in communication and troubleshootin, fun, challenge, why not it's just math, etc.

Installing the transducer was ridiculously easy. Applying the equation wasn't difficult, either. In the end, it is kind of silly and anti-climactic considering it takes 40-60 minutes to heat this machine up for 1-2 drinks that take 15-30s each. But that's all half the fun :)

I do want to do the temperature recording as well, but again that will have to wait until the group gasket needs replacing. It'll be fun to compare the equation then to actual measured data! That said, I may measure water at the portafilter and compare that to the boiler data as well, goal there would be to hone in on Dr G's offset for this machine, and see for myself what happens with a flush!

Meantime, I realized I had some new decaf from Sey and decided to heat up the machine again - it was still warm 6 hours later haha how ridiculous. I didn't wait for the group to fully heat up.

Collected the serial input, and published here (I start recording when the vac breaker engages and pressure begins to rise):

https://docs.google.com/spreadsheets/d/ ... 0i/pubhtml

Not sure what to think of this data. You can see the pressure drop when I ran the steam, and the recovery time. PID units are from the library's sample sous-vide control program, intended to be tweaked if necessary (I have no clue haha).

jpender
Posts: 3905
Joined: 12 years ago

#34: Post by jpender »

jedovaty wrote:...it takes 40-60 minutes to heat this machine up for 1-2 drinks that take 15-30s each.
I had no idea. If that's the time scale then my concern is moot. And that's dedication. I can barely tolerate the 5-6 minutes it takes my electric kettle to boil when I forget to start it in advance.

Carry on!

gscace
Posts: 757
Joined: 19 years ago

#35: Post by gscace »

It's very fast. Can't tell you in seconds but it ain't many if any.

-Greg
jpender wrote:How much of a time lag is there between the temperature of the water and the temperature calculated from the pressure? I ask because in the case of moka pots there was a study done and the investigators found that they could not use the steam tables to calculate the pressure contribution from the water vapor using the water temperature. Instead a better fit came from using the temperature of the vapor. Apparently liquid-vapor equilibrium isn't ever really achieved in that system. Maybe an espresso boiler is at quasi-equlilibrium very quickly but I wouldn't know how to tell other than by measuring.

If it were me I'd want to know the actual temperature while I was designing a system to measure the temperature using pressure, as a way to verify and calibrate. But of course if you could do that you wouldn't need to bother with the pressure.

jpender
Posts: 3905
Joined: 12 years ago

#36: Post by jpender »

Thank you for the response.

I figured it was relatively quick but seconds could matter if you're aiming for some set point. I am a human PID when I want my electric kettle to stop at a temperature below the boiling point. I stick a thermometer in and turn it off in advance because I know the thermal inertia will carry the temperature higher. So in the case of using pressure, if there were a lag between the pressure-temperature relationship then that would potentially add another factor.

But if the machine takes two hours to heat up that's clearly not an issue.

gscace
Posts: 757
Joined: 19 years ago

#37: Post by gscace »

In theory it's instantaneous, but there is probably some mixing in the boiler that adds time but I think any added time is negligible and certainly no reason to discard temperature as a means of controlling pressure in this application.

-Greg

jpender
Posts: 3905
Joined: 12 years ago

#38: Post by jpender »

You say "in theory" but of course that's the simplified theory that generally works well enough. I imagine it's a question of heat transfer rate. I was thinking of the moka pot paper I read where the vapor temperature lagged the water temperature for most of the brew cycle.


jedovaty (original poster)
Posts: 537
Joined: 13 years ago

#39: Post by jedovaty (original poster) »

Hmmmm... John and Greg, your discussion makes me hope the group gasket starts leaking soon so I can get that TC installed :lol:

I reviewed last week's data with my brother who said the machine is taking a long time to actually reach the target temp so to start tuning he suggested doubling the P and I each time I run the machine and observe the results. I went ahead and did that once for today, and made two cappuccino drinks.

Today the machine came to temp much faster and coincidentally introduced yet another variable: boiler volume. I'll keep track of that in my log but it will be eyeballed. The right way would be to have some sort of float sensor or something.

Other thing I noticed is my shot-clock tracking idea with the lever and a reed switch isn't working quite as planned. More details later, short version: when the lever comes to rest, the coffee still flows.

The spreadsheet has been updated with today's pair of milk drinks.
https://docs.google.com/spreadsheets/d/ ... 0i/pubhtml

I'm struggling to make a nice looking graph so feel free to play with it. Based on this latest run, it appears doubling the P and I brought a more stable temp. I'm sure having less water in the boiler helped. It's 1/4 full now, so when I next run the machine I'll fill up to about 3/4 and we'll see.

jedovaty (original poster)
Posts: 537
Joined: 13 years ago

#40: Post by jedovaty (original poster) »

Couple more shots this morning, with 3/4 full boiler. I doubled P, but left I and D alone. Interesting result.. it didn't get to 32psi until I pulled the first shot. Then it overshot and stayed up there.
https://docs.google.com/spreadsheets/d/ ... 0i/pubhtml
Figured out how to graph, too :mrgreen:
For next weekend, I am considering putting P back, and doubling I.