Solar monotube boiler design 1

[altalab]

I am constructing a hybrid solar/ solid fuel generator system. An insulated water tank will evaporate R134a, in a coil(s) configuration,always below the boiling point of water open to the atmosphere for safetys sake. The vapor will drive one or more rotary vane a/c compressors modified as expanders then be condensed and returned to the tank. My questions are:
1) How do I control the feedpump to maintain the proper amount of fluid in the monotube coil(s) according to variable demands for power?
2) Is there a particular design feature of the coil needed to prevent slugging and facilitate evacuation of accumulated refrigerant oil in the coil(s)
Thanks

 

[davefitz]

There are several alternate ways, but you have not specified how the power into the system is controlled or modulated.

In steam power plant applications, a once thru steam generator is controlled to maintain a fixed outlet steam temperature, and the ratio of ( feedwater flow/ fuel firing rate) is perturbed to maintain a fixed outlet steam temperature. As a change in power output signal is recieved, both the firing rate FR and feedwater flow FW are adjusted proportionally. A +10% increase in power is addressed by a + 10% change in both firing rate and feedwater flow, and if the steam temperature departs from setpoint then the ratio of ( FW/FR) is perturbed to return the setpoint.

In the refrigerant case, it is not likely that final temperature is useful, so it is likely that one will need to use some other measurement to monitor the final property setpoint ( enthalpy). Perhaps a measurement of outlet conductivity, or indication of outlet density via gamma ray densitometer, or ratio of outlet flow element DP vs inlet feed liquid FE DP can yield a reliable indication of outlet dryness .

The indication of power input would likely be the temperature difference or LMTD between the bulk water bath minus the temperature of teh refrigerant at the midpoint of the coil. Thus FR= LMTD in this application. One would adjust the feedwater flow in direct proportion to the current LMTD, and adjust the ratio of ( FW/FR) if the outlet conductivity departed from setpoint. Also note the outlet setpoint property ( conductivy or density) will vary with outlet pressure.

 

[altalab]

Unfortunately, the heat source is quite variable (solar/wood) The only modulation is the fact that the water storage mass is large. I commited to a coil design that has a large pipe that will behave like a downcomer. I might be able to place some sort of level sensor in it's wall and maintain the level that way. However Im not sure if the R134a will work with a conventional resistance or capacitance pair of electrodes. Have any thoughts? Your response was quite valuable. Thanks.

 

[davefitz]

If you have room for a large downcomer, then there may also be room for a refrigerant level tank ( drum) within the water tank . If so, the unit can be controlled like a traditional drum type boiler- simply feed liquid refrigerant at a rate which will maintain a fixed water level in the level tank. The vapor leaving the tank would pass thru a "superheater" that will further dry and haat the vapor prior to its output to the downstream process.

If the unit is designed as a "once thru " unit using a coil and no downcomer or drum , then the prior message would hold true. The design of the coil would normally use a smaller diameter for the inlet half of the coil ( liquid pre-evaporator section&quot and a progressively larger diameter as the fluid boils to vapor. This use of a progressive diameter will reduce the geysering and chug flow instabilities.

 

[altalab]

Indeed, I have 3 parallel (1/4" Cu x 60 ft) coils connected from the bottom to the middle of a 1-1/4 ss pipe with caps. At the top of the pipe it goes to a 3/8 x 15ft Cu superheater coil (through the hottest layer of water), then to the outside world. The feed"water" enters at the top center of the pipe with a short interior drop tube to delver fluid to the bottom of the pipe. I also have an exterior column for sight glasses. It would be easiest to use that column for level control since it is outside, as long as the liquid level is true with the inside pipe. I dont know what sensor(s) would be best for simple level detection. Any thoughts?
Bob
PS: The boiler assembly interior is now together and was hydrostatic tested ysterday to 900psi. No leaks, yea.

 

[davefitz]

if you have 3 x 1/4" tubes in parrellel, then that opens up the possibility that you can have flow instability problems. Also, to reduce the tendency for steaming in the downcomer, the downcomer should be insulated . It is a good idea for you to calculate the circulation of the system, by balancing the pressure drop for fluid flowing down the downcomer with the pressure drop of the fluid rising up the 3 heated risers.

The prior remarks about progressively increasing the diameter primarily applies to the 3 heated risers operating in parrellel ( parrellel channel instability)

 

[altalab]

I can insulate the downcomer, I guess I'll have to wait and see about the instability. Exactly what happens with parallel instability? symptoms? I think the heat transfer surface far exceeds my demand (over 15 sq feet in the 3 coils alone) Perhaps this lower demand allows slower flow and increased stability?
As far as FW level, I am considering a cylindrical float in a tube (w/ opposing sight glasses) with light source/photodetector switch. That will minimize vias to the outside world and decrease leakage opportunities.
Thanks for your guidance

 

[davefitz]

For your case, there is no risk of overheat, but when a parellel channel instability occurs in boilers or reactors then there is a high risk of tube overheat and tube failure. In your case, an instability might only result in chugging and difficulty in getting a good reading of level in the downcomer.

If the pressure increase for the fluid flowing down the downcomer ( at a CR of 1:1) does not exceed the pressure loss of the fluid flowing up the heated risers, then the system might dryout and not generate the vapor at teh expected rate. For your system, a good CR ratio might be 2:1 . If the system pressure drop calculations do not indicate a 2:1 CR , then you may need to modify the heated risers by increasing the outlet secttion diameter ( upper half of tubes).

 

[altalab]

I'll let you know what happens. I still have to build the motor, feedwater preheat & condenser circuit before I can run the R134a. Thanks!