Sorry about the latent reply, but I have been very busy -

cafeIKE wrote:No one would doubt this. :roll:

The topic is how changing HX boiler temp affects the brew temp. Brewtus is not an HX. Amica and Alexia are not HX.

Umm, OK ...

I absolutely understand the OP question was based on the use of an HX machine.

cafeIKE wrote: Incoming water has minimal effect as the water is injected into the HX and is completely isolated from the boiler probe. Regardless of where an HX boiler probe is located, changing the set point will change the brew temp once the machine has restablizied, assuming the barista does not change the build process.

It's not that simple, and I really don't care to convey my thoughts on why if it's going to be a "prove who's right" conversation.


.

Matthew Brinski wrote:Sorry about the latent reply, but I have been very busy -

Latent reply? Your reply is not at all hidden Or are you giving off heat Taking the Mickey

Matthew Brinski wrote:It's not that simple, and I really don't care to convey my thoughts on why if it's going to be a "prove who's right" conversation..

Perhaps I was a bit testy in my post... perhaps you'll agree your's is completely off topic. Eric's is spot on.

"I believe that the position of the probe can have profound effects on the machine's behavior." is as obvious as "An electric espresso machine must be connected to an electrical outlet." If I said that, I'd expect a chorus of "DUH!" As an engineer, I often state the contrary and see if it makes sense "The position of the probe can have no effect." Following on, "I can place the probe on the counter and successfully control the boiler." As my physics prof said, "it's intuitively obvious" this is false and therefore the first is equally obvious.

Let's establish some givens:
- The PID is tuned and the boiler temp is stable at the set point.
- The boiler temp is adjusted so the user can walk up and pull a shot without the need to flush to reduce the group temperature.
- Shot interval is > 10 minutes.

And the obvious:
- Cold water injected into the HX causes a change in the boiler environment.
- There are no currents caused by physical water flow, only thermal eddies.
- These thermal eddies are not grossly different from idle boiler operation.

When I placed the probe in the boiler, these were the criteria :
- NOT above the heating element to reduce the likelyhood of over / under shoot
- CLOSE to the HX pipe for decent response to shot changes
- AWAY from the wall to minimize false change detection.

Some time in the future, I may see what happens with the probe in the steam area, not in the water at all.
Ken Fox has posted on water level and HX PID performance

Matthew Brinski wrote:I would imagine that the variances with an HX would have the potential to be more profound.

Have at it. I'm truly interested.

Here is data on a Vibiemme Super Domobar whereby you can get the general trend of shot temperature as boiler pressure varies. A machine that has a PID controller would not behave in the same manner as regards boiler pressure and group/shot temperature BUT the correlation between group temp and shot temp would still be there.

Hi Guys,

Thanks for all of the input thus far; I'd love people to contribute more data or link to more relevant threads if possible.

By combining Eric's latest graph and Dave's steam saturation table, it looks like Ian's paraphrase of Ken's results is more or less right for the VBM machine ... ie. change the boiler temp 2C (or ca. 0.1 bar) and you change the brew temp ca. 1C.

Eric, it is somewhat surprising to me that the group temp measured by your probes basically matches the brew temp. (Where do the probes typically end up sitting when installed? Are they surrounded by air?) Seems to speak a lot for the influence of the group on the final brew temperature. When comparing e61 group variants, I wonder if it is better to have a lighter group that holds more water in the thermosyphon or a heavier group that holds proportionally less water in the thermosyphon? Do you happen to have any data that shows what relationship holds after a cooling flush?

The probe placement thing strikes me as an interesting problem. I would have thought that most people would be placing their probe in the port left open at the top of the boiler when the pstat is removed. If so, do the probes that people have used end up with the tips in water, or in steam? What sort of a difference would this make? Would we be better off using a pressure transducer, as Andy, Greg, or perhaps both, have suggested in the past?

Cheers,

Luca

Luca - probes are positioned as per the drawing in this thread:

Monitoring Brew Temperature - E61 Groups

Whether to put a PID probe in the steam or water or whether to use a pressure transducer ($) would be a subject for a few glasses of wine. Personally I would opt to measure the steam temp because that is what I am trying to control.

As I mentioned, a PID'ed machine will definitely exhibit different group temps for the same corresponding boiler pressure (temperature). I can explain it and will do so later on tonight with, of course another graph.

luca wrote:Eric, it is somewhat surprising to me that the group temp measured by your probes basically matches the brew temp. (Where do the probes typically end up sitting when installed? Are they surrounded by air?) Seems to speak a lot for the influence of the group on the final brew temperature. When comparing e61 group variants, I wonder if it is better to have a lighter group that holds more water in the thermosyphon or a heavier group that holds proportionally less water in the thermosyphon? Do you happen to have any data that shows what relationship holds after a cooling flush?

The probe placement thing strikes me as an interesting problem. I would have thought that most people would be placing their probe in the port left open at the top of the boiler when the pstat is removed. If so, do the probes that people have used end up with the tips in water, or in steam? What sort of a difference would this make? Would we be better off using a pressure transducer, as Andy, Greg, or perhaps both, have suggested in the past?

On my last go round with the meter, I futzed with the placement so the displayed temperature is the brew temp after a couple of seconds. Reads about 6F low at idle.

One of my criteria in choosing the Vibiemme was the large volume of the thermosyphon and mass of the group. Other machines with smaller volume and lighter groups are more difficult to control. The larger volume changes less with cold inlet water, does not recover as fast after a flush and the larger group mass levels the water temp changes.

A pressure transducer could be better for steam response, but it would need to be very sensitive to respond to a shot in time to make a difference. Cost is also a factor. As is scale build up.

I put the probe tip in the water because the response to a sheathed probe in free air / steam is somewhat lagardly. Here's a link to Omega TC Response.