Scace Graphs

[Endo]

Does anyone have a Scace and can post some graphs of shot temperature versus time?

I'd love to see overrlapping graphs of successive shots to see the shot-to-shot temperature consistancy. I'm also curious to see how a shot with a short 10 minute warm-up time compares to one that is fully stabilized for one hour. What does it look like compared to a HX E-61?

I've been looking for these graphs for over a year, and while people have hinted they exist, I've not been able to find them.

[JohnB]

That would be because most of us that Scaced our machines just ran the tests, scribbled down some notes & when we finished set the offset & tossed the notes. Chas has some temp records on here somewhere but if you really need graphs then rent someones Scace & make some.

[Richard]

I'm also curious to see how a shot with a short 10 minute warm-up time compares to one that is fully stabilized for one hour.

The Scace device itself requires far longer than 10 minutes to stabilize (more like 30-60 depending on environmental temperatures) even when starting with a machine that has been on for hours, so that's a nonstarter.

[Endo]

I'm also curious to see how a shot with a short 10 minute warm-up time compares to one that is fully stabilized for one hour.

The Scace device itself requires far longer than 10 minutes to stabilize (more like 30-60 depending on environmental temperatures) even when starting with a machine that has been on for hours, so that's a nonstarter.

I figure if I'll keep the Scace at room temp (or in boiling water) just before pulling a 0 min warm-up shot versus a 5, 10, 15 and 30 min warm-up shots, it will at least tell me "realtive-wise" when things begin stabilizing. Not perfect, but still might be helpful.

Once I get the Scace from Gregor, I plan on making some HD video and transferring the times to make a intershot stability graph at 1 sec intervals. No ideal, but I don't have a datalogger so it will have to do.

[Endo]

Here's a graph from the 2007 Scace YouTube video by someone (S1Vivaldi2...I think it was Niko?)

I'm not sure what the warming flush volume was, but as you can see, the temperature fluctuates by about 5 deg C depending on time and volume between 1st and 3rd flush. Are you listening all you 1 deg F guys?

Point here is, you need to be very consistent with your warming flushes if you want to stay within 1 deg C. You also need to get your offset right depending on which cooling flush you use.

Kinda reminds me a HX a bit......scary! Well, at least the profile is flat (unless you fush 3 times it seems).

I'll be doing some more tests once I get the Scace.

[Endo]

OK. Here is my latest graph:

Details:

Mini Vivaldi with pre-infustion
94C setting
+3C offset
Warmed up for 45 min.
30 sec PF removal at start
30 sec PF removal between shots
25 sec shots (about 2 oz per shot)

The graph confirms the 2 shot warmup. Although perhaps a 3rd warm-up shot might be best since the Scace flow restrictor doubles the time on warming shots.

Some funky stuff happening during mechanical pre-infusion (< 6 sec). Not a big concern though.

Biggest surprise? All shots showed a steady ramp-up of temp of about 2 deg C during the shot. This was very disappointing to see.

6 shots with 30s PF removal between shots

Scace 6x25_30s_wait.jpg (167.2 KiB) Viewed 52135 times

[oton]

It would be interesting to see what happens with the 7th, 8th, 9th and 10th shoot. Temp can not rise forever, so I think the it will stabilize at some point.

I do 5-6 flushes when I start the machine before the first shot...

The 6th shoots ends at +95.5 ºC ... are you sure you need a +3ºC offset?

[Endo]

Look at the graphs again. The 5th and 6th shots are identical so any extra warm-up would do nothing.

The temperature ramp-up of 2C happenns on EVERY shot, so it is a characteric of the machine and not the warm-up.

The real question here is: What is causing this?

I understand the regular Vivaldi has a very flat temperature profile across the shot. Is this a Mini Vivaldi phenomena? The vibe pump perhaps? I can't see how.

Is this caused by my Scace repair? Is the response somehow slower?

I guess the only way to answer this is to wait for another Mini owner with another Scace.

[chas]

I doubt there's much that you can do about it. Perhaps more important is, where in that temperature rise do you pick to be "the" shot temp? Once the initial ramp up is complete, if the temp rises from 93-95C over the remainder of the shot, what do you want to call "the" shot temp? You probably don't want to consider it to be 93C, you probably want to consider the shot temp as either 94 or 95C for the purpose of setting the offset.

For comparison if you haven't already checked it out, my GS/3 temp profile graphs can be found here:
http://gs3cafe.com/viewtopic.php?f=21&t=16

You can see that the GS/3 rises throughout the shot like the S1 just MUCH less so. With my offset set to the mid position the temp remains +-0.2C across the shot. If you use the same mid-setting philosophy on the S1 you'd end up at +-1C across the shot.

Unfortunately, I didn't buy the logging Fluke thermometer until after I sold the 53mm Scace and bought the 58mm Scace so I don't have graphs for the S1.

[JohnB]

It really sounds like your offset is too high. Also you should be running closer to 65-70ml through the scace in each test. After the warming flush my S1 would hit the set temp fairly early in the shot, climb maybe 1*F higher or less & then stay there even if I ran the shot 40 seconds. If I bumped the offset up 1* I'd see a higher overshoot. Either your readings are way off or your offset is just too high.

[Endo]

... you should be running closer to 65-70ml through the scace in each test.

An extra 5ml or so during the warm-up will not make a significant difference. It will only reduce warm-up time very slightly.

After the warming flush my S1 would hit the set temp fairly early in the shot, climb maybe 1*F higher or less & then stay there even if I ran the shot 40 seconds.

This is what I expected too!But it's not what I got.

If I bumped the offset up 1* I'd see a higher overshoot. Either your readings are way off or your offset is just too high.

No. This does not make sense. Offset is simply a bias on temperature and will have no effect on the shape of the curve. For example, my 94C setting with +3C offset is the same as 97C with no offset.

If you had overshoot after an offset change, you most probably did not wait long enough after the change for temperature of the metal to stabilize.

[oton]

Could it be the pre-infusion chamber? The chamber it's outside from the group, so perhaps it drops the water temp initially, hence the slow ramp and temp delta.

[Endo]

Could it be the pre-infusion chamber? The chamber it's outside from the group, so perhaps it drops the water temp initially, hence the slow ramp and temp delta.

I doubt it. It fills up in 6 seconds and after that all water simply goes out the group head.

[Endo]

Here's another graph:

This time I recorded back-to-back shots with NO PF removal between shots. (Sorry, the second shot got cut-off early).

Back-to-back shots

back2back.JPG (139.9 KiB) Viewed 52088 times

Wow. It's a miracle! Flat profile. (Except for the normal sine wave type variation caused by the heater going on and off).

This gives a good idea of why running back to back shots (like in the blue thermometer video) is for marketing purposes only.

This Scace has raised a lot of questions. Most important are:

How realistic is the ramp-up and what effect does it have in the cup? Should I set the offset for the beginning, middle or end of the shot?

[JohnB]

No. This does not make sense. Offset is simply a bias on temperature and will have no effect on the shape of the curve. For example, my 94C setting with +3C offset is the same as 97C with no offset.

If you had overshoot after an offset change, you most probably did not wait long enough after the change for temperature of the metal to stabilize.

BS, you have no idea what you are talking about.

[Endo]

BS, you have no idea what you are talking about.

I thought producing this data and sharing it was the purpose of this Scace rental program.

I have lots more stuff to share, but I don't like being insulted, so I guess I'll just keep the data for myself. God knows I do enough of this stuff in my day job, so I'm not looking for extra work if nobody wants it.

Sorry, just one last graph (a graph is worth a thousand words):

Offset effect

no_offset.jpg (107.6 KiB) Viewed 52085 times

[JohnB]

No. This does not make sense. Offset is simply a bias on temperature and will have no effect on the shape of the curve. For example, my 94C setting with +3C offset is the same as 97C with no offset.

If you had overshoot after an offset change, you most probably did not wait long enough after the change for temperature of the metal to stabilize.

A +offset is going to raise the boiler temp over your set brew temp (set 94*/+3=97*), a - offset is going to lower the boiler temp below your set temp & a 0* offset will keep it the same as your set temp. Are you seriously telling us that you believe that the actual boiler temp will have no effect on the ramp up curve & your possible overshoot??

A few graphs back you complained of running 2-3*C over your set temp (i.e. overshooting). If you were running your tests correctly & your test equipment was functioning correctly your offset would have to be set too high. By the way I would consider the ramp up to be the part of the curve before you hit your set temp & everything after that to be your overshoot.

Then you tell us that you eliminate the 2-3*C overshoot by leaving the p/f locked in which should in fact give you hotter readings.

Did you check the Scace calibration after your repair?

[oton]

BS, you have no idea what you are talking about.

lol. Do you had a bad day or something?

so I guess I'll just keep the data for myself. God knows I do enough of this stuff in my day job, so I'm not looking for extra work if nobody wants it.

Come on, you really know that there are lot of people interesed in you graphs... Me. . J/K. I apreciate your work, endo. Thank you.

So it seems that delta temp is due to the p/f thermal mass loss? But I'dont understand why the temp peaks at +95.5ºC?

[JohnB]

I do 5-6 flushes when I start the machine before the first shot...

The 6th shoots ends at +95.5 ºC ... are you sure you need a +3ºC offset?

Are you saying that you do 5-6 flushes to warm up a cold machine right after turning it on or that this is your standard procedure before pulling a shot?

[oton]

I do 5-6 flushes when I start the machine before the first shot...

The 6th shoots ends at +95.5 ºC ... are you sure you need a +3ºC offset?

Are you saying that you do 5-6 flushes to warm up a cold machine right after turning it on or that this is your standard procedure before pulling a shot?

Only to warm up just after turn it on and I'm in hurry. I checked that I need at least 6 flushes (-I mean, wait for light stop blinking > flush > wait > flush...) to get a decent stable temp shot, and only in 11 amazing minutes. (at 21ºC temp. ambient)

[Endo]

So it seems that delta temp is due to the p/f thermal mass loss?

Simply put, when you remove the P/F for 30 seconds, it takes roughly a 30 second shot to bring it back to it's original temperature when it was first removed and then up to the set brew temp. The rapid rise you see in the first 5 seconds is simply the P/F refilling with hot water. The slower ramp is the PF and inner grouphead area warming back up.

But I'dont understand why the temp peaks at +95.5ºC?

It peaks only slightly higher than the back-to-back shots. Keep in mind, besides the natural conduction in the system, there is a heat source (heater) as well as 2 heat sinks (cold water and surrounding air) at play. All these elements are also continuously changing (heater on/off, water on/off, PF in/out). So the whole system is quite complex and dynamic. You can simplify things greatly by making two of these variables constant (i.e. water always on, PF stays in). I did this first by pulling extra long one minute shots, and I basically got a constant 95C with only a +/- 0.2C variation caused by the heater cycling. But this is obviously an over-simplification.

I now pick my offset based on a "time weighted" average from 5 to 30 seconds, following a representative number of warm-ups and PF removal time. Im not interested in the peak temp.

[Endo]

Are you seriously telling us that you believe that the actual boiler temp will have no effect on the ramp up curve & your possible overshoot??

That's right. As my chart below clearly shows (and as Chas said as well), the shape of the curve is a characterictic of the machine and is only moved up or down by the offset (hence the name offset).

A few graphs back you complained of running 2-3*C over your set temp (i.e. overshooting).

This is something I noticed from someone elses's video (before I received the Scace). I've seen no overshoot what so ever in my tests.

Did you check the Scace calibration after your repair?

I had a lab calibration done on the instrumentation prior to installation of the TC. I'll do another post installation check after I'm finished testing, just to be sure.

[slo]

Go on Endo. I am interested.

No judgement yet. I do not think that I fully understand what is happening. And it is not behaving the way I would have intuitively thought it would...

Good Data gathering (and sharing) work.

[JohnB]

[quote="Endo
This is something I noticed from someone elses's video (before I received the Scace). I've seen no overshoot what so ever in my tests.[/quote]

Really? Then what were you complaining about in this post? When you exceed your set temp by 2*C or more I'd call that overshooting.

OK. Here is my latest graph:

Details:
Mini Vivaldi with pre-infustion
94C setting
+3C offset
Warmed up for 45 min.
30 sec PF removal at start
30 sec PF removal between shots
25 sec shots (about 2 oz per shot)

The graph confirms the 2 shot warmup. Although perhaps a 3rd warm-up shot might be best since the Scace flow restrictor doubles the time on warming shots.

Some funky stuff happening during mechanical pre-infusion (< 6 sec). Not a big concern though.

""Biggest surprise? All shots showed a steady ramp-up of temp of about 2 deg C during the shot. This was very disappointing to see."" :cry

If your tests are accurate for your Mini (big IF) I'd seriously consider moving on to another machine. No offense to Chas but I did not see these issues with my S1V2.

[Endo]

OK. Here's another one. A long one this time to see when the ramp stops. I went for a little less then 4 oz of water. This about as long as I want to go without risking overheating my vibe pump. It's good enough though to answer one of the questions many have probably been thinking. "When does this ramp stop?"

The curve is fairly smooth and shows the typical limited parabolic growth I'd expect when heating the PF and grouphead back up to a set temp after losing some heat from the 30 sec PF removal.

You can see that after the first 8 seconds we're back to within 1C of the set temp. After that, the shot climbs another 2C over the next 30 seconds as the grouphead and PF regain the heat and stabilize. No too bad, but it does show one of the weaknesses of the Vivaldi (no bragging you ex-S1-ers with your big saturated heads).

Some might think I should drop my offset from +3C to +2C so that but as you can see from the graph, this would make the first half of the shot too cool. Instead, I aim to hit the set temp about mid-shot (about 15 seconds in).

Extra Long Shot

long_shot.jpg (105.77 KiB) Viewed 52029 times

[JohnB]

Some might think I should drop my offset from +3C to +2C so that but as you can see from the graph, this would make the first half of the shot too cool. Instead, I aim to hit the set temp about mid-shot (about 15 seconds in).

Why don't you try it & post some graphs. I was hitting my set temp right about where you are & I wasn't seeing the large overshoot that you are so don't blame all Vivaldis.

[JmanEspresso]

You have a Version 1 mini right? With the older temperature sensors, that were not only NOT directly in the boilers water, but inside a thermowell, AND they needed to sense a full degree(C) change to react. Thinking along those lines, it would make sense that if a 1degree DROP is tolerable before it heats back up, a RISE(overshoot) of 1degree is not all that hard to swallow, and Endo, YOU, Im sure, know the Vivaldi is NOT a PID controlled machine Plus, its not sitting immersed in the boilers actual water, which, IMO, is not an unimportant fact either.

I would not be surprised one bit if the version 1 machines exhibited such behavior, and then prompted Chris to source a more accurate temperature sensor which kept the hysteresis under better control... Better in this case meaning more stable. This would also easily explain why John saw much better results on his/my machine.

[Endo]

You have a Version 1 mini right? With the older temperature sensors, that were not only NOT directly in the boilers water, but inside a thermowell, AND they needed to sense a full degree(C) change to react.

That's not a bad guess, but I believe it's not the case. For a couple of reasons:

First, my Mini Vivaldi is a one year old non-US version sold in Canada which came directlly from LaSpaziale. As far as I've been able to tell, the only difference between my machine and the current CC version sold today is the single round steam gauge in the front versus the square dual gauge (and the sticker that is missing a II). Originally it came with 5C increments on the LEDs. But this can be changed easily by holding down one of the buttons for 10 seconds, and then it switches to 1C increments. As far as the thermo sensor and logic is concerned, since the Mini only came out a little over 2 years ago, I believe the Mini never had the V1 setup. (This was also observed by Dave from Bella Barista in his review of the non-US Mini where he says the sensor looks identical to the V2).

The second reason is because even with the slower responding 1C sensor, the lag should only be in the order of a couple of seconds at most. The 2C intrashot ramp I show lasts for over 30 seconds.

[JmanEspresso]

If it had the V1 control board(5C increments), why wouldnt it have everything else the V1 machines had? This is something that, personally, Id ask Chris about, to be sure.(I could ask, or u could?)

The second reason is because even with the slower responding 1C sensor, the lag should only be in the order of a couple of seconds at most. The 2C intrashot ramp I show lasts for over 30 seconds.

Even if the lag was, only a couple seconds.. It would be in the boiler. . . the back of the boiler. Not the group.


Im not saying your wrong, or Im right.. Im just saying Im not ready to write it off that easily yet.

[Endo]

The think all Minis have the V2 control board (US and non-US). I know my machine has a "secret mode" that allows me to switch from 5C to 1C increments by holding down the hot water button for 10 seconds. In order to do both modes, I suspect the the temperature sensor in the brew boiler is no longer a thermowell and is actually this one on all Minis:

http://www.chriscoffee.com/products/hom ... tempsensor

The reason for this is simple.....only one production line for US and non-US models. Swap back and forth with the press of a button.

I haven't verfied this yet by having a close look at the back of my brew boiler. But next time I have it open I'll try and confirm. If anyone has a photo of the back of a older V1 brew boiler with a thermowell sensor, that would help.

[JohnB]

The think all Minis have the V2 control board (US and non-US). I know my machine has a "secret mode" that allows me to switch from 5C to 1C increments by holding down the hot water button for 10 seconds. In order to do both modes, I suspect the the temperature sensor in the brew boiler is no longer a thermowell and is actually this one on all Minis:
.

All S1 Vivaldis had the ability to be switched to the "Fine" (1C) setting from 5C increments from the beginning in 2004 until the V2 came out.

[Endo]

Ok....a couple more graphs for those interested. (If not.....READ NO FURTHER).

Here's a graph after a short 15 minute warmup with the Scace in place. Again, ignore the wonky stuff during the 6 sec preinfusion. This time I'm seeing the same ramp I've seen before, but everything is much colder. Actually, even after 6 warming shots, I'm not yet at my target 94C setting at the 15 sec point (mid-shot).

My conclusion: If I use the same 2 waming shots and only a 15 min warmup, I'll be roughly 3C to 4C under temp my midshot set temp target.

Lesson Learned: Wait 30 mins for warmup. Looks like I'll be getting up earlier (or buying the $$$ timer).

Oh...by the way...I did a Scace calibration check following Greg's instructions (just to be doubley sure), and it was A-OK.

15 min Warmup

15min_warm.jpg (126.45 KiB) Viewed 51980 times

[Endo]

Perhaps this graph is a bit more telling of the warmup:

It shows the shot temperature at the 15 second point (mid shot) for both the 45 minute warm-up and the 15 minute warmup.

Note how the 45 minute warmup stabilizes at around the setpoint of 94C after 2 warming shots. On the other hand, the 15 minute warmp shots never get to the setpoint and no amount of flushing (only waiting) will get you there.

15 min warm and 45 min warm

15 sec_temp.jpg (103.57 KiB) Viewed 51979 times

[JohnB]

Interesting but no big surprise. When are we going to see some graphs with a +1* & +2* offset??

[BobUSN]

Note how the 45 minute warmup stabilizes at around the setpoint of 94C after 2 warming shots. On the other hand, the 15 minute warmp shots never get to the setpoint and no amount of flushing (only waiting) will get you there.

Endo, thanks for posting these. The mid-shot graph is the best one yet, I think.

I'm considering the timer, too. Buying it is cheaper than leaving the Vivaldi on 24 hours a day (in my case it pays for itself in the third year), but if you're shutting it off each night it's hard to use cost as justification.

On the other hand, waking up earlier would just mean wanting coffee earlier for me...

[michael]

i finally hooked up my timer and i think its great. while i have not checked out the machine with the scace for a while, i found that the coffee tastes better when the machine has been on for 45 minutes to an hour as compared to the shot just after the lights stop blinking after around 10 - 15 minutes

[JohnB]

I'm considering the timer, too. Buying it is cheaper than leaving the Vivaldi on 24 hours a day (in my case it pays for itself in the third year), but if you're shutting it off each night it's hard to use cost as justification. ...

I'd start by insulating the steam boiler if you want to save electricity. I saw a 42% reduction in heating cycles after insulating mine. With that kind of reduction you could leave it on 24/7 & save another $200 on the timer!

[oton]

I'd start by insulating the steam boiler if you want to save electricity. I saw a 42% reduction in heating cycles after insulating mine.

Why LaSpaz didn't that? I don't understand it.

[Endo]

La Spaziale designed it so the unheated boiler would keep the cups hot. Remember, they designed this for the small commercial market where a few pennies saved are not nearly as important as having warm cups for the customers.

That said, in the home environment, it's a simple modification if you prefer. It may add some life as well since the heater is not cycled as often.

I didn't bother since I live in Canada where half the year I'm heating my house anyway, so no heat is wasted.

[Endo]

OK.....moving away from insulated boilers and back on topic.............

This one is for you Johnny!

The result was to be expected (a 3C offset....no surprise), but I figured I better do it before I return the Scace or John would never let it go!

94C with OC and +3C Offsets

94_with_offsets.jpg (106.32 KiB) Viewed 51950 times

[Endo]

So....now that this is over. I have more questions than ever!!!

Most importantly....What is causing this +0.1C per second ramp I see on every shot?

It may simply be the result of the cooling and reheating following the 30 sec PF removal but......

I'm curious to see a couple of follow-up tests in order to eliminate some other possibilities:

1. A back-to-back "repaired Scace" versus another Scace.
2. A "repaired Scace" comparison of a Regular Vivaldi and a Mini Vivaldi.

Perhaps I'll try posting on CoffeeGeek to see if the the wider audience or perhaps Greg Scace might provide a clue.

[BobUSN]

Just one more OT...(sorry all, thanks Endo!)

I saw a 42% reduction in heating cycles after insulating mine. With that kind of reduction you could leave it on 24/7 & save another $200 on the timer!

Woo-hoo!

Did you post your procedure here? Anything special I should know?

Agree, Endo, it helps keep the kitchen warm & toasty. But at almost 17 cents per kWh, I need to rely on my natural gas furnace for heat instead.

(Otherwise I won't have money for my wife's beer. )

Thanks again!

[JohnB]

OK.....moving away from insulated boilers and back on topic.............

This one is for you Johnny!

The result was to be expected (a 3C offset....no surprise), but I figured I better do it before I return the Scace or John would never let it go!

Not exactly what I asked for, I'd still try a +2C. Might be time to start thinking about that nice 58mm p/f on the Duetto 2.

[JohnB]

Woo-hoo!
Did you post your procedure here? Anything special I should know?

viewtopic.php?f=57&t=860#p12484 Best advice I can offer is to use an insulating material designed for the job. In this case that is foil backed ceramic wool (KO Wool). Fold it around the bottom just leaving the drain ex posed & over the top peeling the foil back away from the electrical connections. Seal it up with high temp metal tape.

[JohnB]

Most importantly....What is causing this +0.1C per second ramp I see on every shot?

It may simply be the result of the cooling and reheating following the 30 sec PF removal but......

I'm curious to see a couple of follow-up tests in order to eliminate some other possibilities:

1. A back-to-back "repaired Scace" versus another Scace.
2. A "repaired Scace" comparison of a Regular Vivaldi and a Mini Vivaldi.

Perhaps I'll try posting on CoffeeGeek to see if the the wider audience or perhaps Greg Scace might provide a clue.

Are you the first Mini owner to rent that Scace? If not has anyone else had the same problem dialing in their machine?

[Endo]

OK. Latest update.

Checked with the lab again. They couldn't find anything wrong, but still they suggested to swap out the thermocouple and try again. So that's what I did.

Not much difference.

Perhaps there is something in the Fluke setup I'm missing? I looked through the setup but couldn't find anything. Still, I'm no Fluke expert and I don't have the manuals so that is always a possibility.

Just for fun I tried setting the fluke to K-type instead of T-type (just in case there was some screw up there). That offset the temperature by +2 deg C but the curve stayed the same. Interesting, but no closer to solving the mystery.

Anyway. I've run out of ideas. So I'll be sending the Scace back to Gregor soon.

This is starting to really bug me so I think I may drive my machine down to Albany this year and have Chris try to calibrate it. In the mean time, and until this mystery is solved, take all these graphs with a grain of salt!

Good luck next renter. Hope you have more luck than me!

[JmanEspresso]

Woo-hoo!
Did you post your procedure here? Anything special I should know?

viewtopic.php?f=57&t=860#p12484 Best advice I can offer is to use an insulating material designed for the job. In this case that is foil backed ceramic wool (KO Wool). Fold it around the bottom just leaving the drain ex posed & over the top peeling the foil back away from the electrical connections. Seal it up with high temp metal tape.

And depending on what your definition of a warmed cup is.. even with the boiler insulated, the cups get nice and warm. But, thats with a cup rack about completely full, and a machine on all the time.. If you like a SUPER hot cup, like using the hot water tap would get you.. then you will still find them not warm enough.

Only thing I can think of, worth mentioning, is that there is definite hot spots/cooler spots. Basically, what I do, is only keep my espresso cups on the cup warmer. The back row, the cups are upside down, so the "catch" the heat. The front row, the cups are sitting normally, because upside down, they get really hot(too hot for me). In between the two rows, is my other portafilter(either the bottomless or single spout..whichever one Im not using at the time). This keeps all my espresso cups, at the ideal temperature for removing from the cup rack, and putting right under the spout.

For milk drinks, I just use the hot water. Using hot water to warm the cup, makes it too hot for me, espresso wise.. but for milk drinks, it doesn't matter.

With the boiler insulated, running on 15amp is fine. Obviously, when I move and have my espresso bar setup, Ill have her on 20amp, because, well.. why not. But, if for some reason you had to run on 15amp, you wouldn't worry about it.. I never think to myself, "man, I really need 20amp!"

[richardcoffee]

Wow! There's a lot of information here but I have a simple question that I'm not sure is answered here. When I do a double shot, and I have the temperature set at say 94C, I almost never get the lights to stay on for the duration of a 35 second shot. I have an S1VII - bought new in January, 2011. I have tried varying ways of flushing - from no flushes to 2 flushes - always a minimum of 1 hour warm up - occasionally, even with no flushes, the lights will stay on, other times even with 2 flushes, waiting till the lights all stay on and then doing the shot, the lights go out. Is there a correlation between the lights going out and a flat temperature profile? Or is it necessary to use the Scace device to get a true temperature profile?

[peter]

I'll venture a guess that the blinking lights only signify that the boiler element is on, and that's only because the group boiler is small enough that a flush and a shot draws the water down and it's refilling and reheating. But I don't think that will affect the temps of the water going through the puck.

[chas]

The last light that is on steady indicates the temperature that the temperature sensor is currently reading which can drop quite a bit during a shot as the cold water is mixing in with the boiler water toward the back of the boiler. The offset is set so that the temperature of the water at the top of the coffee puck is as close as possible to the same as your set temperature. This normally means that the water in the area of the boiler heater element and the temperature sensor must be a few degrees higher than the set temp so that as the water cools while flowing toward and out of the group it stays at your set point.

When I did my initial tests of the original S1 back in 2004 the Scace was not yet on the market. I created my own version by 1) drilling a small hole in the middle of the single basket which I never use anyway 2) cutting a circular piece that just fit into the single basket out of a kitchen sponge 3) running the lead of a K thermocouple probe up through the PF spout, through the hole in the bottom of the single basket, then through a tiny hole I cut in the middle of the sponge so that the thermocouple tip was flush with the top of the sponge.

One key difference between this and the Scace is that the Scace device partially blocks the the bottom of the PF spout to that at 9bar about 2oz come out in 25 seconds to much more closely emulate the presence of a coffee puck. A device like this is probably good enough to use to get a handle on the offset needed for your machine. However, to most accurately profile your machine and/or compare those readings with others, the Scace device is pretty much the only game in town since Bob Scace* is considered an expert in the field of lab testing, equipment and certification.

* Robert "Bob" Scace has been an active leader in the SEMI Standards Program since 1975, as a founding member of the original Standards Committee and its successor the ISC until the present, as the chair of the ISC Regulations Subcommittee from its inception through 2006, and as the "corporate memory" of the SEMI Standards Program. Now retired from NIST, he represents SEMI as a US expert in ISO TC 229 and IEC TC 113 on nanotechnologies.

[richardcoffee]

Thanks..