Just curious, what datalogger are you using ? Do you have a thermocouple to take the place of the thermometer ? If not, do you want one ?

Some interesting "experiments" to try:

With a relatively constant ambient temperature, the grouphead temperature will vary with boiler water temperature because of the normal pstat cycling as shown below:



With a relatively constant ambient temperature, note the grouphead temperature with the portafilter and basket in place or just the portafilter with no basket. The machine should be fully warmed up in this condition for 1 hour. Now remove the portafilter and read the grouphead temperature after another 30 minutes. Grouphead temperature will decrease with the portafilter in place.

Naturally, grouphead (GH) temperature will also vary proportionally with ambient temperature. But, to get back to your original quest, I would install the thermofilter and have it in the machine for an hour. Remove same and flush the machine to a GH temp of 205 F, reinstall thermofilter and immediately pull a data logged simulated shot. Do a 1-2 second screen flush and reinstall thermofilter. Observe the GH thermometer and you should see it dip to AROUND 195 F. Pull another data logged shot with the thermofilter when the thermometer reads 197 F. Note the time it took to recover to 197 and compare the readings of the two shots.

erics wrote:Just curious, what datalogger are you using ?

I am using an Beha-Amprobe TMD-56 (plus Scace thermofilter).

Since 2003, Beha was a Fluke company, until it was sold to Amprobe (USA) in 2006. Being a layman, a Fluke 54 Series II plus hardware and software was way too pricy for me, so I have chosen one with a better price-performance ratio, trading some accuracy/precision for it.

erics wrote:Do you have a thermocouple to take the place of the thermometer ?

Now, I am not sure anymore what you mean. On my grouphead, I have installed the Digital Thermometer & Adapter for Exposed E61 Groupheads.

To make myself super clear: I have the "erics" thermometer installed in the GH and additionally a Scace 1 thermofilter in the portafilter with a data logger thermometer attached.

Thanks for the "experiments to try". I will do it tomorrow and post the results.

erics wrote:Grouphead temperature will decrease with the portafilter in place.

Yes, I have watched and experienced that already. But I haven't data-logged it .

Now, I am not sure anymore what you mean.

I know what you do have . . . the Amprobe is probably the best value data logger that exists. All of the grouphead temperature graphs I have ever posted on HB were performed with a thermocouple in the grouphead vice the thermometer. It is a 30 second change from one to the other.

Gentlemen, here are the results.

Please read the graph's headings as they contain additional information and answers as well.

erics wrote:...install the thermofilter and have it in the machine for an hour. Remove same and flush the machine to a GH temp of 205 F, reinstall thermofilter and immediately pull a data logged simulated shot.

Note: I wasn't able to exactly flush to 205 °F but to 206 °F.

erics wrote:Do a 1-2 second screen flush and reinstall thermofilter. Observe the GH thermometer and you should see it dip to AROUND 195 F. Pull another data logged shot with the thermofilter when the thermometer reads 197 F. Note the time it took to recover to 197...

Note: After the 2s flush the GH was after 10s at 197 °F (= 91.6 °C). But as it was dropping too fast I waited until the temperature was rising again, being more stable (like with flush-n-wait).

It is difficult to flush exactly to a specific temperature because the thermometer is reacting quite rapidly at that point in time.

The important item to gleen from this is that if you wanted to extend that 2-1/2 minute wait between the 1st shot and the 2nd shot, you should do a 4-5 second flush after the first shot to reach your 5 minute mark.

erics wrote:if you wanted to extend that 2-1/2 minute wait between the 1st shot and the 2nd shot, you should do a 4-5 second flush after the first shot to reach your 5 minute mark.

Thank you so much Eric for your time, patience, and consultancy!

I do want to share a last thing with you in this thread. Hopefully it doesn't drive you crazy or bore you to death but rather helps some being in a similar situation as I am.

I pulled totally random shots (not all being shown below) - without being systematic at all. I was doing what you shouldn't do - changing more than one variable at a time. Why? Because I just wanted to get a feel on how my Rocket Evo behaves under different conditions. I also wanted to see how flush-n-go changes the temperature profile.

So I varied everything - the time between shots (interval), the amount of flushed water, the rebound time, flush-n-go/flush-n-wait, preinfusion/no preinfusion.

Here's how it looks. The grouphead (GH) temperatures were read with Eric's thermometer. The graphs were logged using the Scace thermofilter.

To conclude and answer my initial question on how to systematically develop HX flush routines for specific (average) target brew temperatures I would say one needs to experiment by changing one variable at a time:

• time between shots (interval time)
• amount of flushed water
• rebound time
• preinfusion or no preinfusion

If you are crazy like me, a lot of time would be needed to be spent with testing!

If you want to continue reading about this topic, here are two additional threads I found (too late):

• HX flushing with Giotto: consistent temperature possible?, already from 2008 but very very good and interesting!
• Can I brew at exact temperatures with HX espresso machines? from 2010

Some additional things to hold constant:

• flow rate profile
• boiler temp (see above graph for how it's not constant with a pressure-stat)
• room temperature and air flow (eg, an overhead ceiling fan)
• incoming water temperature