(This is a continuation of Playa Espresso Cart 1 - Art Cart Concept discussing the design and construction of a propane powered, manually pumped portable espresso machine)

Playa Espresso Cart 2 - Thermoregulation

What I imagine my brother-in-law Ben, the Steam Wizard of Worth, pictures about to occur in Tucson...


Allon, Spiffdude, and Eric S brought this story to my attention, and discussed it, in /espresso-m ... t20780.htm. I certainly hope that by posting my calcs and designs on the web, peer review may help prevent me from injuring anyone, including myself.

Playa Espresso Cart: Thermoregulation Requirements

1) Propane Power, no batteries or wall electricity for control
2) Commercially sourced components
3) Feedback control: Heat turns off when preset exceeded, and on when second preset is undershot; hysteresis ("dead zone") controller
4) Failsafe; if wind blows out flame, propane flow stops
5) Reasonable operating time on 5 gallon BBQ propane cylinder

Heating Capacity Calculations:

My calculations regarding heat generation capacity follow:



The gist of it is that to heat the 12 lbs of water in the boiler from 60 to 260 degrees, about 2400 BTU would be required. To check, given the machine is supplied with a 3kW electrical heater, then it should take the stock machine about 15 minutes to heat up to brew temperature. That sounds about right. So, my best guess is that I need a burner capacity of around 10K BTU/hr. This would provide about 36 hours of burner on-time for a 5 gallon BBQ propane burner.

CONCLUSION: Burner needs to be able to produce 10,000 BTU / hr.

Design Proposal 1: Control using a two-burner system

There is a fair amount on line about small, propane powered hot water heaters ("Flash" heaters) used in RV's and cabins. I thought perhaps one could be adapted. Maybe; but never was I certain of what components were going to be inside the package that arrived. There is a big marked for home brew bbq equipment, and lots of suppliers of propane bits for people who make their own smokers. Reading about their systems, as well as home-brewers (silly me- I thought the term only applied to electronic fabrication... it refers to people who make their own beer first and probably foremost).

Ultimately I came up with the following design. It uses two sets of controllers; a conventional pressure stat for the boiler; and burners and venturis from a home fireplace gas log set.



Feedback control is provided by electricity provided by thermocouples and thermopiles. These gizmos are hot (from being in the propane flame) on one end, and cool (from being outside the flame) on the other. When hot from propane, they generate low voltage (millvolts) at a low amperage (milliwatts) but that is enough to energize an electromagnet that keeps the gas flowing. If the flame goes out, the thermocouple cools down, the magnet stops being a magnet, and the valve closes and stops the flow of gas. Here is a picture of such a device- a safety pilot light.



One such device would be used to control the overall flow of gas: as long as the flame is on, the gas would be able to flow to the system.

The design I came up with is more complicated, because it needs both a safety pilot, and a second valve that is controlled by boiler pressure:



In my design, there are two burners. Each burner has a separate needle valve that regulates the flame size when the flame is on.

The first burner, the "Basal" burner, provides enough heat to mostly keep the boiler up to temperature when it is mostly empty. It should not produce enough heat to exceed the heat loss capacity of the boiler itself- when the basal burner is on, it should always slowly lose heat.

The second burner, the "AfterBurner" - is a real flame thrower. It kicks on when the boiler pressure is low (as sensed by the stock pstat). It kicks off when the pstat flips state. The pstat has a dead zone that is adjusted identically to the method used if it was switching 220. In order for the pstat to turn on and off the gas to the afterburner, a second control valve (the Honewell VS8420) is in the system. This is a small controller for a fireplace gas log set, and has a capacity for 60K BTU/Hr so it appears to be appropriate. It uses a "thermopile"- a bunch of thermocouples - that produce enough voltage to actively pull in and out a solenoid valve (unlike the pilot assembly that requires a manual reset).

Not shown, because I have not yes sourced, is a failsafe thermostat that is in series with the thermocouple from the pilot assembly. I don't know how hot the boiler would get if it went runaway - say 300 degree F? - but if it hit that temp, the connection from the pilot thermocouple to the pilot valve would be interrupted by a thermostat breaking the circuit, thus cutting the flow of gas to both burners. I think this would most likely happen if the boiler were allowed to run dry- a very real possibility.

ADVANTAGES OF DESIGN:

1) Commercially available parts
2) Readily available
3) Reliable- uses pstat that is well understood

DISADVANTAGE:
1) Expensive - If I buy this stuff new, I am looking at several hundred dollars.
2) Complex - I am not sure if all the stuff would fit in the alloted space.

Design Proposal 2: Servo-controlled Regulator

Somewhere in H-B there is a posting where someone describes the gas system on their Astoria. It is just what I would have liked to have found: Direct feedback regulation of gas flow to a single burner. As the boiler pressure goes up, the flow of the gas goes down. As pressure goes down (when fresh cold water enters) the gas flow goes up and it gets hotter. A single burner that operates smoothly and automatically adjusts for ambient temperature and water level variations.

I read somewhere that gas conversion kits might be available.



Surprise! Conti still had in stock a kit to put an Astoria 2-Group into propane service. I placed an order, and very shortly exchanged emails with Edith, who asked "Are you SURE you need this part?". I felt like James Bond, going to my bank and "arranging a wire transfer for funds to a bank in Monaco" !!! The entire assembly came in less than a week. Thanks Conti!

I have not wired it up yet, and want to put the thermostat in the pilot system as a failsafe. Overall, I felt very lucky to learn that the conversion kit for the Astoria was still available. The boilers must be very close in size (LSM and Astoria), for the burner appears to be perfectly sized.

ADVANTAGES:
1) Experienced engineering team designed the burner and venturi
2) Servo, rather than on-off, regulation of gas flow
3) Expense. The entire kit was less than the Honeywell millivolt regulator discussed above.

DISADVANTAGES:
1) No idea, other than "tricky", how difficult it will be to set the loop gain for the regulator. For all I know, it could start oscillating.
2) No online documentation for the regulator. The bits looks like they have been sitting on a shelf since 1970.

Conclusion: Use the Astoria Conversion Kit and hope that I can adjust it properly

The next post will cover the pumps (boiler and group).

Thanks for your review-
Joe

Excellent write up. Looks like you've been doing your homework.

Those of us considering gas systems will be following along closely and looking for applicable tidbits and experience.

Continue to next in thread:

Playa Espresso Cart 3- Lever Pumps