Thanks. Overshooting is a possibility but may not be needed with the Prestina on a timer, once I get it back there. To fully heat it to pull a shot, even when it's up to temperature Jim Schulman suggested I do a heating flush and just check the temperature on the group (briefly!) with my hand.

I haven't found it necessary to PID a La Pavoni with a PSTAT. Instead I prefer using group thermometers with these and having the boiler pressure idle slightly low to keep the group cool enough but ready.

I wasn't sure what you meant by crimping > soldering. Crimping and then soldering creates a good electrical connection. Crimping alone can let go.

Sorry to revive this, but 6 months in, how is your SSR / element holding up?
Any issues?

*edit looked again definitely Zero Crossing

It's all holding up well. My installation mounts the SSR far away from excessive heat and on the frame. Plus it's shielded by insulation covering the boiler. No problems with the heating element. As we speak the Prestina is running and pulling beautiful shots that I can dial in very easily.

I'm about to change out the PSTAT because the Jaeger one that's in place limits reaching full temperature for very light roasts. Instead I'll install a Ceme PSTAT that allows a higher pressure setting. The PSTAT is there as an additional safety device at this point.

Eventually I will wire in a resettable temperature switch. That will be a better safety device because when it's tripped it shuts the machine down. Then it easily resets. The PSTAT would allow the machine to power on after a small amount of cooling.

drgary wrote:
Eventually I will wire in a resettable temperature switch. That will be a better safety device because when it's tripped it shuts the machine down. Then it easily resets.

Gary, I like that idea. Do you have a particular one in mind? Here's one. But I don't like the idea of tapping my boiler endplate.

I've got one for a home machine. It has a flat bottom and I would figure out a way to fasten it under the insulation or something.

Gary, i just ordered most of the parts to mimic your setup (really like your thermocouple location in particular.) With the added exception of replacing the PID with an arduino YUN and TC4 ( I want to be able to turn my machine on via the web) my final part to buy is the thermocouple itself. I'm looking at either

http://www.amazon.com/gp/aw/d/B00I2XI9U ... 69&sr=8-31

http://www.amazon.com/gp/aw/d/B00I2XIA7 ... p_s_a_1_30

In short.
K vs T
K has glass insulation
T has Teflon insulation
Further both of those are beaded, I've been finding a lot on beaded vs wire for roasters, not so much for espresso machines.

Thoughts?

It looks like you know what you're doing. I don't have anything to add. Please let us know how it works.

Update

I finally programmed the offset between my sensor's reading of the boiler temperature water and desired brew temperature. This was after measuring with a thermocouple inserted through a filter basket with a hole drilled in the side and tuning the Prestina to a coffee that I know brews ideally at 202F. It was interesting to find out that the temperature at the top of the coffee cake initially measures above boiling but settles down quickly into the desired brew range and the coffee doesn't taste burned. I found my temperature offset to be about 56F.

Programming the PID was a bit tricky. When I entered the offset difference in the PID it did not compensate for the altered reading by automatically adjusting the target temperature I'd programmed. For just such a possibility I watched the manometer start to climb after inputting the offset. Seeing this, I toggled off, released some steam to get temperature back in range, and re-set the target temperature in the PID from 258F to 202F. Now I can set the PID to desired brew temperature.

I will wait until I'm serving multiple shots in a row to see if the brew group heats so the offset reading is too low. Also the offset may be different for much lighter or darker roasts. For now, though, it's pulling delicious shots at proper temperature.

Another advantage of attaching a PID is that I was able to swap the PSTAT for a vacuum breaker without changing the steam pipe configuration. The Prestina releases its air pocket during warm-up automatically. I've tested the machine twice while watching it and feel comfortable that I can now start it on a timer and have it ready to pull shots when I wake up. Warm-up time is somewhere between 30 and 40 minutes. If it hasn't been sitting idle for a long time, I flush some water through the group to heat that to brew temperature. Feeling the outside bottom edge of the group near where the portafilter locks in tells me whether it's hot enough.

As an extra safety measure I'll connect a thermal safety switch to the boiler before putting it on a timer and will attach the leads that formerly went to the PSTAT. Although this PID fails open and wouldn't need such protection, should the machine spring a sudden and large leak the thermal switch would avoid damage to the heating element from the boiler running dry.

Update on the Update

Two days later I find that the offset programmed temperature setting works really well. I just dialed in the expected temperature difference for a lighter roasted coffee and the flavors were as expected. The group may run a touch hotter when water level in the boiler gets about a third lower than full as seen in the sightglass. As with any manual espresso machine there's some learning curve involved.

Update 5/24/18: A few months ago I woke up to my Prestina overheating when it started with a timer. It was not subject to heavy use, but I've had electricals in other machines overheat when connections loosen just a little bit. The element was not damaged but the solid state relay (SSR) had partly melted. So I decided to shore up the connections and outboard the PID controller and SSR to preserve their longevity and make them easily swappable. I also installed a resettable heat safety switch for a La Pavoni lever at the top of the boiler, similar to what Paul Pratt did in a restoration he documented. His safety switch was a little fancier, but this one works and actually shut off the other day when I was tuning the PID at a high setting. The safety switch is held in place by a string around the boiler and kept flat onto the boiler by the boiler insulation. I also installed non-toxic heat conducting paste underneath.

Where I had connected two inbound wires to join a single one, I discarded spade connectors, cut a small piece of copper tubing and crushed the wires together inside the tubing using a large bench vise, then surrounded the crimped connection with heat shrink tubing. As a further safety device I re-installed a Jaeger pressurestat. The pressurestat is at maximum setting, so it doesn't cycle unless the PID is set too high or stops working.

I acquired a new 15 amp InkBird IPB-16 PID unit for under $90. The back of the aluminum InkBird case sinks heat, so it's offset by bent metal flanges with screw holes. This isn't friendly to a countertop, but I found that rubber power cord gaskets that work for my home levers fit in those screw holes and provide a soft rubber contact to the counter. The InkBird came equipped with a non K-type sensor, so I adapted a K thermocouple from a female mini-connector to an audio port that fits the PID unit, which is settable for many kinds of sensors. I waded my way through the version 1.0 Chinese to English instruction manual, and fortunately found YouTube instructions posted by home brewers that helped me achieve the needed settings. The thin-wire Teflon thermocouple had creased the group-to-boiler gasket, which had become baked on. Rather than remove it I cut a gasket from soft silicone sheet and threaded the thermocouple wire between that sheet and the group flange. With the group fastened tightly on the boiler the silicone fills the gaps and solved the leak. I used a General Tools gasket cutter that worked wonderfully well after I sharpened the blade with a metal file and used a leather punch for the group stud holes.

My Prestina is retrofitted with a vacuum breaker that empties to the drip tray and has an overpressure valve.

It is now working again on a timer and is a thorough pleasure to use, a steampunk dinosaur with precise, computerized temperature control.







To reiterate why I am using a PID instead of relying on the pressurestat, it allows temperature setting to the 10th of a degree when I change coffees and operates within a small deadband, so when I change temperatures I'm not fiddling with the imprecise screw control on the Jaeger that also has the potential to jiggle and loosen spade connections. Using a PID also minimizes duty cycles to the Jaeger.