Stirling Engine Design (LDT & high power)

Discussion on Stirling or "hot air" engines (all types)
Post Reply
zytrahus
Posts: 1
Joined: Sat Oct 27, 2007 9:10 pm
Location: location

Stirling Engine Design (LDT & high power)

Post by zytrahus »

Hello,
I am working torward a Stirling engine prototype for a particular application. Let me start with a small description of what I am going to do (try) anyways:

Because of the application, I will experiment this with a small electrical heater:
- heat input: 50-100 W (heater thermall insulated on all sides but the die)
- die size: 20x20 mm
- die material: Copper
- thermocouple in the geometrical center of the die (top surface)

I will not control the hot side temperature as I will control the heat input (50-100 W). Depending on the overall thermal resistance of the device, I expect a temperature of 75-80 C on the hot side of the engine. Basically I would like to keep my heater temperature (thermocouple on top of the die) below 80 C.

I will keep water temperature between 20 and 30 C. So I expect a maximum temperature delta of 45-60 C.

I have equipment to measure/control heat input as well as cold temperature (chilled water). I can also measure output RPM.


---

Now my problem:
I know that stirling engine might not be very good for thermal dissipation as it uses a gas that is pumped. But let's imagine a stirling engine that is thermally good, i.e. with a low overall thermal resistance... You can give it a lot of heat input and the hot side of the engine will not get too hot, which means a low delta of temperature between the two sides, hence a bad efficiency (cf Carnot). Anyways, even if the efficiency is like 5%, 200 W heat input, you could get 10 W output power.
Now since it's a LDT stirling engine, 'it is said'* that RPM are low which will a 'high' torque on the shaft. (*I have read about it but never actually seen equations - actually I think it's more of a design choice = choice of volume size to reduce RPM...).
But here is my problem/understanding: low RPM = big thermal resistance: with low RPM I don't see how we can keep a low delta of temperature between the two sides because the gas used will not move enough energy per second to leep a low temperature. So the hot side temperature will increase, increasing the RPM, decreasing the thermal resistance. The efficiency is also increasing with the hot side temperature (cold side kept constant).
To me it ends up, since I don't know a typical stirling engine thermal resistance, I don't know what will be my hot side temperature. Without this, I cannot determine RPM neither volume sizes.


any comment/thought on all this, please go ahead :D
jesterthought
Posts: 21
Joined: Sun May 20, 2007 5:22 am

Post by jesterthought »

Hi,

What you describe is similar to tests I was doing last Winter with a "coffee cup" LTD engine. I was interested in slow speed and no power, so the heat input was much less than yours.

A resistor, epoxied into a block, was pressed up under the aluminium base plate, which rested on a scooped-out foam block. There was little gap for air escape. The cold plate was of copper and convection cooled by soldered-on fins. I did not measure temperatures, but noted the free-running rpm at various watts as development continued. The slowest I got, with a light flywheel, was about 30rpm. That required about 1/4W, IIRC.

The temperature was low. The engine would not only rum on my palm but on two fingers, with the rest of the base exposed. It even ran on one finger of a nice warm hand in a cold room.

Thermocouples are excellent for measuring temperature, especially our principle parameter, DELTA T; just loop the hot junction and the cold junction between the hot and cold plates and you get the differential read out directly. Isolate them electrically, of course, but get a good thermal bond. I plan to use at least one enamelled wire and to paint the junctions, so the twisted pair can then be shoved deep into a driled holes full of grease. Then thermal conduction along the wires will be shorted to the plates, leaving the junctions accurate.

I found that for each throw (swept volume) there was an optimum heat input for peak speed. More heat than that and the engine slowed again. An adjustable throw proved invaluable.

For sizing the power cylinder initially, I aimed for a displacement of half the expansion at constant pressure from the expected DELTA T.

I'm not sure that most LTD engines have much regeneration. I call mine a hot air engine, not a Stirling.

Have fun! Build first, ask questions afterwards.

Jetser.
Post Reply