EspressoForge wrote:Just had a great shot with a dark roasted Yemen, lower group temp, higher TB temp (240F). To me when I tried a shot of this at 230F but warmer group temp it didn't seem as good. I'm thinking once I get some Scace data, I may try to calculate a shot temp, so you could pull one with a lower group temp, or you could adjust TB temp once the machine is more warmed up. Obviously the temp curve would be more falling or more stable depending on how far apart the temps are. But I think this would get me to the result I'm looking for.

With thermocouples reading and displaying the group head temperature. I'm able to change the shot temperature a couple of degrees by altering the preinfusion time and the time between shots/flushes. Longer preinfusion gives higher shot temperatures. Longer time between shots/flush gives lower shot temperatures.

Have you tried varying the time between shots and the preinfusion time? What results have you seen?

FJ60Cam wrote:If you had the cartridge heaters (and since you aren't using the TB for steam), couldn't you run the group and TB at the same temp? Is a 200 degree group temp possible?

Nope not possible. It's not possible on any machine IMO. But some dual boilers they hide the "real" temps, and apply an offset, hopefully constant and hopefully calibrated with a Scace. Sam and Curtis have already mentioned, but basically without crazy insulation this wouldn't be possible. Insulation downside would be that if you ever want to adjust temp cooler, it would take a long time!

samuellaw178 wrote:Oh, I just came across some pictures of Faema Mercurio yesterday. With a metal sheet or two, it seems you can easily pretty up the OS lever. Probably you have seen this anyway...

I could...but honestly I'm not an artist, and I probably would end up with something that looks the same, or likely worse. Hand hammered copper like on the Victoria Arduino looks sweet! But if I attempted it, it would probably end up looking like the beginning of an episode of This Old House

https://upload.wikimedia.org/wikipedia/ ... a-edit.jpg

pizzaman383 wrote:With thermocouples reading and displaying the group head temperature. I'm able to change the shot temperature a couple of degrees by altering the preinfusion time and the time between shots/flushes. Longer preinfusion gives higher shot temperatures. Longer time between shots/flush gives lower shot temperatures.

Have you tried varying the time between shots and the preinfusion time? What results have you seen?

I haven't, mostly because I'm not certain what the temp actually is inside? I've been preinfusing to a pressure and playing with that...with my needle valve I can control how long it takes to get up to 2 bar, which seems to work better than I expected for light roasts. I thought some initial flow would be a problem, but I reduce this hammer effect by just pulling the lever slowly.

I also believe the time you leave the lever down won't change the max temp much (to be Scace confirmed). But rather the group temp and TB temp are what drives that max temp. I'm thinking now of a control loop setup like the this:


The formula might need to be more complex, I plan to gather a bunch of data and then determine an accurate formula. But from this example, lets guess that 240F TB and 180F G1 temp make a 205F shot temp. If you want 200F, lets say you lower the "virtual" shot setpoint, this then calculates what TB temp is needed (at the given group temp). And if the group temp changes slightly over time, the TB setpoint will adjust to compensate.

In theory this should let you pull the same shot temp (which is actually the max temp of the shot on a Scace curve), at different group/TB temps. The closer they are to each other, the more flat that curve will be. Essentially giving some amount of control over temp profile. Of course this would only work within certain range, say when group temp is at least 140F or something, as I'll have to put a soft cap on the TB max temp so that it doesn't drive too high.

This cascading style of loop is a lot more work to tune, but from my past experience in control systems it can produce really good results if you constrain it. If you don't, the loops can run away from you and potentially produce some bad results, so I'll start with tuning each of them somewhat separately and using pretty conservative PID values for the 2nd loop.

Couple updates.

Here's a shot of the pressure gauge during warm up at about the max pressure I've seen (about 7 bar). My line pressure is regulated to 2 bar, so during heating, this is what increases. The "adjustable" OPV doesn't kick in, so I'm thinking maybe I need to change out the spring. Ideally I'd be able to adjust the pressure from about 2-8 bar relief, but this one I'm guessing kicks in at around 8 bar minimum.


I also added a quick calculation of shot temp (out in the top right):


Though the temp probably isn't accurate, it gives me a quick idea of a comparison number. Right now I've been using it in "HX" mode. Run at a bit of a higher temp (to get group warmed up fast etc), flush to the temp I want, then lock in and pull. This seems to give me repeatable results so far, but in the future I'm not certain if I'll keep it this way, or get it to a no-flush type of system. Though most DB systems rely on a warm-up flush, and most HX need the cooling flush anyway.

Planning to take a video, maybe today I can as I'll have another HB member visiting and wouldn't have to set up the tripod and everything.

My visitor came bearing gifts, along with the handles I've gotten my bottomless. That means Scace data!

Those are gorgeous handles!!

I should have mentioned they are olive wood and ebony. The lever handle has a custom stainless cap. All is very solid and feels as amazing as it looks.

Planning to clip the zip ties unless there's a reason not to? Seems like hot water will get all over the cable otherwise.



Also planning to do the "calibration" offset determining procedure outlined here:
Scace Thermofilter Temperature Device

and some other thoughts here by Greg Scace:
Scace Thermofilter Temperature Device

Overall looks that although calibration isn't needed for the type T, I feel if others want to follow along it may be helpful.

EspressoForge wrote:I should have mentioned they are olive wood and ebony. The lever handle has a custom stainless cap. All is very solid and feels as amazing as it looks.

They were really fun to make! Hope you enjoy them!

Also thanks for the shot and tour of the machine; I really like how discrete it is!

Do not use the published corrected barometric pressure from local weather station, use your actual elevation above sea level and then correct the saturation pressure/temperature for elevation. That will get you very close to the correct values. If you want absolutely correct you will have to come up with the uncorrected barometric pressure at your physical location. The easy way to get local pressure is with a mercury column barometer that is only compensated for temperature.