Scace Thermofilter Temperature Device

[BobY]

What is this thing called? The Greg Scace Device? The Scace Portafilter Thermometry Device? The Thing that Barry Carried Around the Conference Ticking Off Espresso Equipment Manufacturers? Whatever you call it, please welcome guest reviewer Bob Yellin! He will lead the discussion and I will chime in Bob's commentary with my own observations comparing it to my "dinky TC" (Abe's words, who will also join the review). Thanks to Terry Z for supplying "the device" to HB.

Update: Added thermofilter instructions and WBC procedures.

Part 1

This will be a review of Greg Scace's Espresso Machine Thermofilter Temperature Device (known hereafter in this review as "TTD"). This device is a "snap-in" basket for an espresso machine "chopped portafilter" and measures the temperature of water directly out of the brew head. I will post the review in installments over a period of about two weeks, but if there are questions or comments along the way, please feel free to post them.

I used an Extech 421508 digital thermometer to measure and data-log the temperatures. The Extech has a resolution of 0.1 deg F and an accuracy of ±(0.05% reading + 0.6 deg F). The QuickMill Andreja Premium E-61 HX espresso machine was the "test bed".

To begin the process, I photographed the device and measured vigorously boiling water. Water at my altitude (about 500 feet) boils at 211.1 deg F. The TTD measurement showed the boiling water to be 211.2 deg F. I also checked temperatures against a NIST-calibrated standard digital thermometer, down to about 180 deg F (the range of interest) and there was never more than a 0.5 deg F difference between the TTD and the standard. Very impressive!!

In the next installment I will discuss the construction of the device and my initial impression of it "as received".

Photos of the device are available in other places but there are some differences between them and this, the "production" version. So here are some shots of the TTD:

[BobY]

[BobY]

[espressoDOM]

How much is the actual temperature display.... (not included in the price)

just curious... and how IMPORTANT is this upgrade....

Please give opinions... (or is this not the place for this)

[another_jim]

How much is the actual temperature display.... (not included in the price)

just curious... and how IMPORTANT is this upgrade....

Please give opinions... (or is this not the place for this)

It depends. With a $10 thermocouple and $12 basket you can make a "schomer device," drill a hole in the basket, epoxy in the TC, fill it with coffee, and measure while you brew. So the real question is what is gained by Greg's set up.

First and foremost -- repeatability between shots and people measuring. Every Schomer basket is unique, every fill and tamp is too. So using the same basket will let you know if the water is getting into the coffee at the right 1C or so range ballpark, but no more. Using different baskets on different machines will tell you very little -- the position of the TC is critical when it comes to measuring how quick the machine gets up to temp, etc etc.

With interest now in 0.1C accuracies in shots, and in characterizing how fast a shot settles, or what the temperature profile during the shot is, the required accuracy and repeatability are ratcheted up, and the Schomer basket is in beyond its depth. Greg's device is the first attempt to meet that need for more accurate measurement. Greg is a NIST scientist and espresso machine expert, so no better person is even imaginable to design this "next step" in thermometry.

There is a steep price rise when going to measuring gear that is one magnitude more accurate. This is true of voltmeters (not readout accuracy, but real accuracy), gauges, or anything else. To get that extra decimal point in repeatable espresso measures requires a big step up from drilling a hole and epoxying in a TC. Whether that extra effort is worth it in your practice is for you to decide.

[another_jim]

What is this thing called? The Greg Scace Device? The Scace Portafilter Thermometry Device? The Thing that Barry Carried Around the SCAA Conference Ticking Off Espresso Equipment Manufacturers?

Pity Greg doesn't work for the German TUV. Then it would be the
Druckkaffeetemperaturmessungseinlage.

Of course, if Greg were German then he wouldn't sell it, he'd just write the spec:
Druckkaffeetemperaturmessungseinlagebestimmung

[gscace]

There is a steep price rise when going to measuring gear that is one magnitude more accurate. This is true of voltmeters (not readout accuracy, but real accuracy), gauges, or anything else. To get that extra decimal point in repeatable espresso measures requires a big step up from drilling a hole and epoxying in a TC. Whether that extra effort is worth it in your practice is for you to decide.

I'd also like to add that it turns what used to be a real pain in the yass measurement into a completely trivial exercise. No more packing coffee around fragile thermocouple wire. No more snaking wire up through baskets or over basket lips. Just pop it in and go to war. The amount of real good data you can get in a very short time is pretty mind blowing. With Schomer's method, or other methods of snaking thermocouple wire inside the pf, you couldn't make continuous duty cycle measurements because of the time required to set up each run. Now you can map out machine performance from intermittent duty to complete full-on, line out the door and into the street, panic mode continuous shot pulling. Not that I have any interest in this of course.

-Greg

[gscace]

Oh, yeah. to clutter this up further, another cool thing that is now possible is to use the portafilter as a training tool. It's really easy to get feedback on flushing and rinsing techniques. If you are learning a new machine you can sort it out very quickly. It's also now possible for a shop or a coffeegeek to actually run some sort of quality control program in which the brew temp got measured once a day, which means that a shop can stay on top of changes in machines over time. Makes diagnosis of temperature related problems real simple. Also makes machine replacement easy because you can quickly dial in the brew temp of the replacement.

I hope these points resonate with machine techs and shop owners who I think will really benefit from this.

-Greg

[Abe Carmeli]

Greg,

Since you already have a flow adjusting valve on it, adding a pressure gauge would have been nice. Perhaps for your next upgrade?

[BobY]

Sorry not to have posted the entire review in one shot but I have house guests and full days with them until late next week so I'll be posting this piece-meal. Most of the raw data is collected. It just has to be written up. It will include charts of data, comparisons with other measurement methods and lots more.

As far as flow rate is concerned, I believe it is fixed, or at least it is recommended that it stay as received. But Greg can correct me on this if necessary. Here is a quote from an email to me from Greg (with his permission, I hope). I noticed that the flowrate gave me between 2.25 and 2.5 oz in 25 seconds. Greg said:

"I suppose I need to clarify this for everyone, but the reason that the flowrate is a little higher than you think it ought to be is because the coffee cake absorbs water that you don't see in the shot, but which passes through the group. We measured the water uptake of a bunch of shots (well actually Barry did the measurements), and the value is about equal to the amount of coffee in the pf. So the 2.5 ounce number is actually just a shade shy at 25 seconds. The Italian standard is for 30 secs extraction and the WBC standard is as well, so the volume collected that you see will be very close to right at 30 secs. Should be around 76 ml at 30 secs."

I hope to have the next installment tomorrow if I can lose my grandchildren for more than 10 minutes!

BobY

[gscace]

Greg,

Since you already have a flow adjusting valve on it, adding a pressure gauge would have been nice. Perhaps for your next upgrade?

It's a good idea at first blush and one that I've tried out. The problem is that the internal volume of the plumbing gets so large that the espresso machine dumps a bunch of water out into the thermofilter on startup, which doesn't reflect what's going on when brewing. So until I can get the internal volume down it's gonna be a temperature only device. It's worth my looking at because the measured pressure reflects the pressure when pulling shots, rather than the pressure against a blind filter. On my Linea, the difference is around 1 bar, which surprised me.

-Greg

[another_jim]

It's worth my looking at because the measured pressure reflects the pressure when pulling shots, rather than the pressure against a blind filter. On my Linea, the difference is around 1 bar, which surprised me.

-Greg

Interesting. After reading our various "how to adjust vibe pump posts," some people complained when the found their brew pressures were 1/2 to 1 bar below their blind pressures. I guess this is a characteristic of OPVs rather than the pumps they regulate.

If the drop from blind to flow would be constant across all machines, shot flow rates, and OPVs, it would be no sweat. Just set it by the blind filter and everyone will get the same result. However, if that creates different actual brewing pressure for different machines and shots, then setting pressure with metered flow from the group also become very desirable.

[BobY]

Part 2

The TTD was received with a chopped portafilter (happily) but not the one shown on the Espressoparts.com website. There, one sees a rubber-handled, original La Marzocco portafilter, but the one shipped with the TTD is an after-market portafilter with a plastic handle. The shipped pf is a very tight fit on an Andreja Premium due to the 8mm group head gasket that comes standard with the machine. This pf would be a better fit on a machine with a 6 or 7mm head gasket. While these are minor points and have nothing to do with bench testing the TTD, I feel they are appropriate in a review of a complete product, as received.

Also, there was no documentation of any kind with the TTD. I would recommend that an information sheet with some basics specs and a guide for less experienced users, be included.

The Espressoparts.com web page (http://www.espressoparts.com/product/EPTF) has an adequate description of the device so there's no need to dwell on that here, other than to mention that the TTD (sans pf) consists of a standard 58mm basket containing three major components, as seen in the photos: 1) a thermocouple probe placed across the outlet: 2) a grey, epoxy-like insert or cake: 3) a 1/4" pipe and an end cap with a tiny orifice and brass-colored metal "filter material" inserted in the pipe, all of which acts as a flow restrictor/controller. These components are designed to simulate the temperature and pressure found in actual brewing conditions. All this was based on observation. If I've misspoken about any of this, I would hope that Greg will jump in and correct me.

The TTD gives the impression of a very solid, well-designed device utilizing high-quality components, including a Teflon-jacketed, strain relieved, type T thermocouple cable connected to a standard mini-plug. All that's needed in addition to the supplied device, is a digital thermometer that can handle type T thermocouples (and an espresso machine, of course!).

In the next installment I will describe how I measured temperatures and used the measurements to set up my espresso machine to produce a consistent temperature profile shot after shot. I will include charts.

BobY

[Abe Carmeli]

I started with my testing today. First problem I see on my unit is flow rate. It is exactly 2 oz, or 60 ml. Lower than what I understand it should be. (around 75 ml in 25 sec. ?). Documentation could help there.

Flow rate is important in brew temp measurements. It affects not only the measured average brew temp, but also the variation of temperature within a shot. A 55 ml flow rate will have a slightly lower brew temp, but it will also cut the brew temp variation within the shot by half. Oh, the glory of ristrettos.

Bob, what exactly is your flow rate in 25 sec?

[HB]

Isn't the flow rate dependent on the pressure with a restrictor? I am guessing that Bob's brew pressure setting is higher than yours and that explains the higher flow rate.

[Abe Carmeli]

Isn't the flow rate dependent on the pressure with a restrictor? I am guessing that Bob's brew pressure setting is higher than yours and that explains the higher flow rate.

Sure thing Dan. My point is that fixing flow rate, as Greg has done here, may not be a good idea for that very same reason. Different shops use different flow rates for their extraction, and different pressure. Without an adjustable flow rate valve, this is becoming a problem. That is of course if you want to be precise within a tenth of a degree.

[gscace]

I started with my testing today. First problem I see on my unit is flow rate. It is exactly 2 oz, or 60 ml. Lower than what I understand it should be. (around 75 ml in 25 sec. ?). Documentation could help there.

Flow rate is important in brew temp measurements. It affects not only the measured average brew temp, but also the variation of temperature within a shot. A 55 ml flow rate will have a slightly lower brew temp, but it will also cut the brew temp variation within the shot by half. Oh, the glory of ristrettos.

Bob, what exactly is your flow rate in 25 sec?

That's interesting. I've been checking all of the flow rates before sending them out. They are all being run at 9 bars and should produce around 75 mls in 30 secs. The WBC standard and Italian standard is for 30 secs extraction and the tolerance is +- 5 I believe. Could be that your brewing pressure is too low.

-Greg

[BobY]

I started with my testing today. First problem I see on my unit is flow rate. It is exactly 2 oz, or 60 ml. Lower than what I understand it should be. (around 75 ml in 25 sec. ?). Documentation could help there.

Flow rate is important in brew temp measurements. It affects not only the measured average brew temp, but also the variation of temperature within a shot. A 55 ml flow rate will have a slightly lower brew temp, but it will also cut the brew temp variation within the shot by half. Oh, the glory of ristrettos.

Bob, what exactly is your flow rate in 25 sec?

I'll note this in my review, but the flow rate begins at about 2.25 to 2.5 oz in 25 seconds. But strange...as the sessions progress the flow rate goes down. For example, yesterday I think it was at about 1.25 oz after a few hours of pulling shots. That could be a function of my machine rather than the TTD but it needs more investigation.

[BobY]

Isn't the flow rate dependent on the pressure with a restrictor? I am guessing that Bob's brew pressure setting is higher than yours and that explains the higher flow rate.

This will appear in the next installment when I begin to discuss the actual testing, but for the record, my machine is set so that while pulling a shot with the TTD in place, the pressure reads exactly 9.0 bar with a brewstat setting at 1.0 ± 0.05 bar.

[Abe Carmeli]

I'll note this in my review, but the flow rate begins at about 2.25 to 2.5 oz in 25 seconds. But strange...as the sessions progress the flow rate goes down. For example, yesterday I think it was at about 1.25 oz after a few hours of pulling shots. That could be a function of my machine rather than the TTD but it needs more investigation.

It could also be coffee clogging the exit hole. Greg, how do you clean it and release a clog?