I am in the process of putting a thermocouple in the grouphead of my Faema Compact, and I was planning on using an Auber PID as an inexpensive temperature readout. How do people typically power the PID controller? Erics's instructions mention using an extension cord with the end stripped. Could I just wire it into the power switch on my machine so that the controller turns on when the machine does? Thanks

Unless you have the PID already, it would be a lot cheaper to go with a multimeter or dedicated TC reader. More importantly, there is less risk of electrocution with the battery operated device. Stripped electrical cords will work, but just telling someone to use a chopped cord over the internet seems negligent. If you're really looking for a flush mount display, there are still cheaper options than a PID that are panel mounted. They are simply temperature indicators and can be conveniently wired into the main switch.....if you know what your doing.

I was under the impression that a PID was the cheapest way to go. The Auber ones are about $35. However, I'd rather use another option if it's no more expensive. Do you have any recommendations for readers?

If you're planning to run the wires inside the machine and/or mounting the PID inside the machine, then heat will be a consideration. Check with Auber, but I suspect the temperatures inside the enclosure exceed the design limits of the device. My Watlow PID for example, is rated for 0 to 65°C (32 to 149°F). I have measured air temperatures inside my Isomac case and they get to about 152°F at idle (although that's at the top of the case; if you mount it at the bottom it might not be as warm.

Heat is also a consideration for the wiring; cord with low heat ratings will become brittle and crack over time under high heat conditions, and if it ever gets loose and touches a part with high heat the insulation will melt and short. Bad news in either case. Standard lamp cord for extension cords will be rated most likely for 60°C or possibly 75°C, both of which are too low for use in that kind of environment. I would stick with cord rated at 105°C - the temperature rating is printed on the cord. My LED installation has wire rated at 105, and it's doing fine inside the case: http://coffeegeek.com/forums/espresso/machines/442638

PS: The power for the Auber can probably be run from a tap of the power coming into the Faema. On my Isomac, there were extra terminal block lugs that I could tap into. See the bottom right corner of the photo in this post: http://coffeegeek.com/forums/espresso/machinemods/440427#440427

$35 is not bad, I was thinking about the packages where you get the SSR, TC, Project box, wire harness etc for well over $100. I dont know much about auber units but they should have a mode for just being a temperature indicator. In theory, you're paying for all those nice PID algorithms but not putting them to use, but for 35 bucks, you'd be hard pressed to do much better on a plain old indicator. You can try Love Controls or Omega or Eurotherm to see if you can save a few bucks. Sweet Marias has a real cheap single TC reader (handheld) for 29.99 if you want to keep it simple. Any hardwired devices will tell you the specifics of how they need to be wired and you should refer to them over any individual's advice. All should be able to draw from the main switch, given you have the ability not to do anything dangerous.

PS. I have a Seaco that I wired with an Omega PID and was able to splice into the main switch by creating pig tails with wire nuts running to shielded connectors.

Erics's instructions mention using an extension cord with the end stripped.

Like so:


or



Using an inexpensive PID as a temperature meter has several advantages - cost, sampling rate, ability to input a correction factor, ability to handle numerous types of inputs and the best of them all - the ability to use as a true PID at any time in the future.

edit - The ability to use as a true PID at any time in the future for essentially any espresso machine.

The power cord is stamped 60 C and is extremely flexible 18 gage stranded wire.

I understand the disclaimers about trusting the advice of individuals over the internet. I've worked on the electrical parts of my machine enough not to do anything dangerous. Also, I was planning on mounting the display outside of the machine, so temp shouldn't be an issue. Of course, I'll get wires rated to the appropriate temperature.

Eric: Thanks for the pictures. That's what I was looking for.

Re: The Sweet Marias meter: It's an attractive option if it works as well as the PID. Any idea how the sampling rate or accuracy compares to a PID?

I cant comment on the actual sample rate of any device but it should be in the devices technical notes. The sample rate of a indicator/controller can be any arbitrarily fast rate but wont help if you have a big fat metal probe that takes a longer to heatup than the sample rate. Its safe to assume most TC readers like the SM one would sample faster than what looks like the 1/8" or 1/16" sheathed probe in the picture. If you had an exposed junction TC you might notice the difference between a faster and slower sample rate. if your up to the project box method and want the convenience of power up with the machine, the PID is a fantastic option because it give you room to expand the project by adding a cheap SSR.

The Sweet Marias meter: It's an attractive option if it works as well as the PID. Any idea how the sampling rate or accuracy compares to a PID?

Frank, you've had too many side trips to Lynchburg The meter that SM offers is EXCELLENT for roasting functions but at $10 differential between it and a PID, the contest is not even close. A typical Chinese PID sampling rate is 4/s, the Fuji PXR3 does 2.5/sec, inexpensive hand-held meters do 1/s, Fluke 54 series do 2.5/sec. But, this is just one attribute - all needs to be considered - as you know.