If you've read any of my posts, you will know I struggle with brew water temperature control on my 18 month old Profitec Pro 500 PID (HX design). By taste, it seemed like my espresso was bitter (Edit: nooo.. SOUR.. I have learned that at least) after the first shot. I suspected cool brew water and attempted to measure with various thermometers and thermocouples, but never got what I thought was a consistent answer. I finally broke down and bought/installed Erics thermometer, which increased the cost of my machine 10%. However, I never really knew how it related to the SCACE. Some people thought that after 20 seconds, the SCACE and Erics seem to be identical. From snippets I've read in other places, they claim Erics is about 3-5 degree higher. I believed the former, since I seldom got any readings on Erics higher than 205 and thought "surely the brew water isn't that cool".

There is much information on HX pertaining to older HX designs with no restrictors in the TS and flushing routines. But there is little hard data on the newer designs with PID and restrictors in the TS loop. I have no idea whether my machine behaves similar to all other Pro 500 PIDs; I can only assume it does.

I plan to do more SCACE measurements.. I just received the unit today.

The SCACE measured 211.9 in my steam test so my meter is pretty accurate (I can spit in the sea). It gives 57g of water in 30 second pull. Within specs I believe.

My workflow. (NO flushing!)
Machine on for 60 minutes..
SCACE/PF installed only in last 15 minutes (I know a weak point of the measurements.. will fix on next round)

Pulled shot #1
Steamed 5 oz water to 140F (simulate the cappo/latte)
2 minute wait
Pulled shot #2
Wait for rebound for 12 minutes
Pull shot #3

My PID is set for 253F (Profitec only recommends 248)

Shot #1.. the brew water averaged about 197. Not too bad. I was shooting for 200F. Eric thermometer about 4 degrees higher than SCACE.

Shot #2 .. before this shot, I steamed a small pitcher of milk to 140F. Then pulled shot #2 at about the 3 minute mark. Here you really see the effect of the restrictor in the TS. My brew water never got above 193. (Here's where I was hoping that Erics reading during the middle of shot was the same as the brew temp... it wasn't)

Shot #3 .. I waited until the group thermometer to read at least 196 and then pulled the last shot. This took an additional 12 minutes.. This shot was about 196.. maybe just a tad cool.. I likely could have waited longer.

Very preliminary conclusions..

1) Tables in HX manuals that attempt to relate boiler PID settings to brew water are not useful. By Profitec's manual, my brew water should be about 204F.. I am at almost 10F below that.

2) Eric's thermometer is an excellent tool for relating the actual brew water temp to its reading. (about 4 deg high with my workflow)

3) The rebound in these newer designs with restrictors is unforgiveably slow (maybe the milk steaming had much to do with this). I do not believe anyone should pay $2k5 (Can) and only be able to make 2 lattes in 20 minutes.

How on earth could you ever guess at what the brew water temperature was without at least a group thermometer? It is an essential part of using the newer HX designs.. Because of manufacturing tolerances, I suspect that each instance of this machine might behave thermally slightly different.

However, the SCACE does confirm my 'gut suspicions' ever since I've owned this machine. If I am shooting for 200F in my brew water, I've been using this machine for 18 mo and still have not found a workflow that will produce that. I'm confident I now will; but HX designs with restrictors might not be a very user-friendly beast.

4) Manufacturers could do way, way more to make HX designs more user friendly. Specifically, (especially on E61 groups), there needs to be a temperature monitoring point near the screen of the group. Then this machine would be much easier to use. I see the MaraX has attempted to address some of these issues.. not sure how their new design works.

Interestingly, the Profitec Pro 500 now can come with flow control by replacing the mushroom, but that eliminates the possibility of installing a group temperature thermometer (unforgivable! and shows a lack of awareness on Profitec's part of the necessity of a group thermometer).

Feel free to criticize the experiment.. I'm very new at this.. I will bump up the PID and add some flushing before shots to see the effects and will publish over time, if there is interest.

This is in part why I haven't been convinced of the value of a PID in an hx machine. It hold the boiler temp constant, but the relation to brew temp . . .

Seems like money is better spent on Eric's thermometer than the PID.

Great information Dave!
I'm currently looking into upgrade from my small lever to bigger pump machine, and Pro 500 PID is my number 1 so far. But this long rebound times sounds not good
One thing i've noticed when reading your posts, that size of restrictor differs. You've mentioned about 2,25mm, but currently on Pro 500 PID parts diagram they list 2,5mm. Perhaps this has been changed recently to improve this issue?

https://wiki.wholelattelove.com/images/ ... iagram.pdf

I'm not sure why I said 2.25.. maybe I got that from some early research, or I mistyped.. my parts list that I got online says 2.5mm as well.. I have not actually measured the restrictor. So I might have made a mistake in saying it was 2.25. I actually wrote 2 communications and sent to Profitec in Germany complaining about slow rebound and they did not mention anything about different size restrictors. So I'm assuming mine is 2.5mm.

Edit: After taking apart.. mine was indeed 2.5mm.

Do you plan to remove the restrictor to compare at some point?


Thanks

Randy

Don't think that is possible from what people have told me.. It is 'hard attached" to the group or the copper tubing.. likely soldered in place.. So won't do that.. if it was easy.. I'd have done it long ago.

Later edit: I did eventually take apart.. Wasn't hard.. just had to go get a 21mm wrench. Good experience. See later posts in thread

lukehk wrote:I found something similar with my Rocket Cellini with PID. Using a homemade version of a SCACE device it took about 15 minutes to get back upto temp after each shot. If i made back to back shots they became progressively sour. I ended up setting the PID so the idle temp on the group head thermometer was 94c. I then use a flush and wait technique. After this I could pull back to back shots every 2-3 minutes.

Yours and my machine seem to behave quite similarly.. I wonder if steaming milk also delays the rebound significantly. This TS mod that all manufacturers have made really sucks the performance out of these machines. Like running large virus protection on Windows .

For those interested in how Lelit are addressing this problem in HX designs, look at Jim's description in this link. It is excellent in detail and likely a much better engineering solution than what some manufacturers are currently using (restrictor in TS).

One test I hope to conduct is whether creating a milk-based drink (ie: steaming 5-6oz) has a significant negative effect on rebound.

Lelit MaraX

Bluenoser wrote: .. I will bump up the PID and add some flushing before shots to see the effects and will publish over time, if there is interest.

It would be really interesting to see how flush between shots affects rebound times, without changing boiler temp.

I understand it as follows:
After pulling a shot, temp in HX tube is lower than on group head. Thermosyphon loop starts circulating in back-direction, sucking heat from group head to HX tube. At the same time HX tube is getting hotter from steam boiler. Then at some point, thermosyphon loop comes back to normal direction, since HX tube will get higher temperature than group head.
How long will it take for HX tube to closer temp to steam boiler level? I guess it depends on size of tube. But when it's hot enough, by making flush, we could speed up group head heating. Then wait a minute for stabilization and pull another shot.

So the question is when the earliest is possible to make this flush. Or maybe I'm missing something?