Scace Portafilter Thermometry Device

[HB]

Barry mentioned Greg Scace's Portafilter Thermometry Device on the BGA forum (link). You can read about The Scace Device on his website.

If I'm not mistaken, this is the same "highly accurate modified portafilter, geared with both pressure gauge, adjustable flow rate valve, and temperature sensors" Abe uses. In addition to extolling its wonders, Abe derides others with lesser equipment; my setup is "a dinky TC" in his mind (hmm-m, I may have TC envy!?!). I don't know if Greg will be selling them directly only through vendors. As I recall, TerryZ will be selling it. Anyone know for certain the price and availability?

Updated: See HB's review of the thermofilter.

[malachi]

As I understand it...
expensive
soon

[bobroseman]

I think Greg's data looks very interesting. Look at the five shots on the Linea. They are all over the place both in shape and absolute temperature.

Bob

[HB]

Pointer to a recent related discussion in coffeed.com, including an exchange on the "Cannonfodder Device" (page 2).

PS: Discussion of the CD split to a separate thread.

[terryz]

The device is now available!

[HB]

Terry, I assume you're referring to the thermofilter. Any plans to offer a "Cannonfodder Device"? It should be a lot less expensive and delivers considerable practical benefit.

[malachi]

While such a device could be less expensive, if done right I think it would probably cost about the same...
The big issue for me with the 'cannonfodder' device lies in the display unit which is both slow and inaccurate (imho). I'd want something with the fastest possible response time and with 1/10F resolution.

[cannonfodder]

While such a device could be less expensive, if done right I think it would probably cost about the same...
The big issue for me with the 'cannonfodder' device lies in the display unit which is both slow and inaccurate (imho). I'd want something with the fastest possible response time and with 1/10F resolution.

I agree completely as far as the latency/accuracy issue with my little rig. I have never disputed that, it was only intended to help develop a baseline in my flushing routine. But, you are comparing apples to oranges, or pro to novice.

Having used my little rig for a month now I do wish I had something a bit more precise, but it has served it purpose. I want to make the same outfit but with a TC instead of the thermometer. If I ever get the fittings I ordered (delivery is slated for 8-30), and they actually fit, I am going to make another using a type K TC in a rigid housing. The cost should be very reasonable but not as covenant to use as Gregs PF, you just lock and go with it. I do have to say it stirred up some conversation and got a lot of folks thinking and tinkering, which is a good thing.

[HB]

I'd want something with the fastest possible response time and with 1/10F resolution.

I think there's room for market segmentation. Judging from the line of reasoning many home enthusiasts express in their decision making process, a major motivation for buying a dual boiler machine is simplification. The merits of one temperature profile versus another aside, some rightly view HX temperature management as an unnecessary hassle; that's what motivated my question in the Bench forum (Expected differences between dual boiler and HX?).

If the key is indeed simplification, then the pertinent questions aren't around response time and resolution, it's about consistency. A newbie HX owner may hope for 4F shot-to-shot consistency and an experienced one can expect closer to 1.5F. Now if the "Cannonfodder Device" moved the newbie along the curve to begin at the point near their experienced peers, we've made some genuine progress for that group (especially in the agonizingly difficult HX versus dual boiler debate).

The value of this admittedly imprecise feedback is of questionable value to a seasoned professional or advanced home barista. But as Dave points out, that wasn't the intended audience in the first place.

[barry]

I think Greg's data looks very interesting. Look at the five shots on the Linea. They are all over the place both in shape and absolute temperature.

one of the first things i learned after probing my linea was, "stability is relative".

i also purposely used a group which had been sitting idle for at least a half hour. i wanted instability in the shots to better gauge the response time and how the data from greg's device compared to the data from my banjo bolt probes.

fwiw, last night i did some further testing on my home linea, and after the first few shots, the temperatures stabilized pretty well. as soon as i format the data and generate some graphs, i'll put it up on the page at angry-bunny (probably not until friday or saturday late; gotta do some graphics work for the wife first).

the raw data file is at http://www.angry-bunny.com/stuff/scace/test2.txt. "shots" were 30 seconds each at 30 second intervals, with the device removed and wiped between each "shot" (hence the temp drop).

i also did some response testing of an inexpensive thermocouple panel meter, but i need to do more before making any comments/decisions.


--barry

[gscace]

The difficulty with this stuff is that once you make the measurements you have to then interpret the data. Interpretation of the data also means that you have to understand the effects of the boundary conditions on what you got. It's a mistake for people to draw conclusions about general machine stability in actual usage if you purposely measure an idle group as a demonstration of agreement between two probes, which is a valid test of agreement, but not a valid test of machine brewing performance.

The temperature measurement standard that Barry, Bill Crossland, John Sanders and I wrote for the WBC addresses specifically the measurement method to be used in evaluating machinery for the WBC. I suppose the standard ought to be made very available so that people use the same methods once people are armed with good measuring gear.

-Greg

[barry]

It's a mistake for people to draw conclusions about general machine stability in actual usage if you purposely measure an idle group as a demonstration of agreement between two probes, which is a valid test of agreement, but not a valid test of machine brewing performance.

in fact, i found stability to be a hindrance when testing the panel meter. i worked on it after working with the thermofilter, and the machine was up to temp, so i couldn't tell if the lack of temp fluctuation on the display unit was due to brew temp stability or poor meter response time. i need to redo the tests from idle to better gauge meter response.