Engine Pressurization

[stephenz]

The 5-mesh simulation is done:
https://file.io/2YFvPwq4JDuH

same comparable parameters, just 5 meshes in series.
steps are very obvious.

the thermal mass of each is much too high, as even the first mesh doesn't get heated much.
the total of thermal mass for the 5 meshes is pretty good as the temperature of the gas exiting the last mesh is close to Tcold.

the next step is to make those meshes with significantly less thermal mass.

My next simulation is 5 meshes, but the mesh size has changed: thickness is 0.1mm vs 0.2, holes are now 0.45mm instead of 0.4mm, and holes are staggered (the center of 3 adjacent holes form an equilateral triangle). The mass of the mesh went from 0.2g to 0.07g.

hopefully we should see that the temperature of the mesh is higher, which should result in a temperature of the gas at the outlet higher than before (since I kept the quantity of meshes to 5).

[stephenz]

so when considering a regenerator based on N stacks of mesh-style regenerator, the thermal mass of the total amount of gas must far exceed the thermal mass of each individual mesh, otherwise (in the context of gas flowing from heater to cooler) the first mesh will never rise in temperature high enough to be efficient on the way back.

as it stands, the thermal mass of gas is really low and it's difficult to fabricate a mesh that would work, particularly on a non-pressurized system. if you pressurize the He, things change as the mass of helium increase dramatically (I haven't checked yet how the specific heat increases with pressure and temperature yet).

But this would explain why thin stainless steel wool is a pretty good medium. Although there is mechanical contact (hence conduction) between "fibers" they're so thin that conduction through individual fiber should be minimal. I can't really model those as it stands though.


I started receiving parts for the Alpha engine that I will use to test a wide range of regenerator materials. But considering how I built the housing hosting the regenerator material I wonder if it's going to work. The housing itself will probably have some significant losses.

[Tom Booth]

I started receiving parts for the Alpha engine that I will use to test a wide range of regenerator materials

So how did these tests turn out?

[stephenz]

So how did these tests turn out?

I wasn't able to get the Alpha to work, I found a lot of small issues left and right.
I keep a lot of the parts the parts and redesigned some to convert it into a Gamma.
It still wouldn't start, found the problem to be poor ratio between cooler/heater/regenerator and there the easiest way to try to get it to run was to actually redo the power piston and its cylinder and head. Reduced the diameter from 70 to 54 mm (basically half the swept volume) which allowed a much less challenging 1.5 ratio between the Displacer and Piston swept volumes.

I was able to get the engine to run, and have been tweaking since then.

The main issues now are vibration - I haven't even tried to balance it, although I have a vibration meter and the means to improve balance significantly. But I suspect I have a lot of losses due to that. Other than that, I have been working on regenerators obviously, pressurization and various heater designs. The best I was able to achieve is about 30 W electric going from the generator/MPPT to the battery.

From I've learned, assuming one gets the right distribution of volumes right, the keys to power and efficiency are engine pressurization and reduced friction.


Oh and one you might like. Literally two days ago, as I was preparing with my last few tests with the Mk-I design (as I finished making a lot of new parts for what I hope to be significant improvements) it had not run for a week or so, busy making parts, I completely forgot to turn on the liquid cooling unit (both pump and fans were off). The engine didn't seem to work as well as the last time it ran so I look at the GUI and see the cooler temperature at 100 C - whoops that can't be good - I immediately turned on pump and fans, the temperature started dropping immediately and the RPM started to rise. Nothing surprising here.

Anyway I am hoping to get an order of magnitude better output with the Mk-II.

[Tom Booth]

...
was able to achieve ... about 30 W electric going from the generator/MPPT to the battery.

...

Expensive to build? Looks like you've been busy. Way more work than I have time to put into my little toys.

30 watts isn't too shabby IMO considering these 15 watt engines have already long been in commercial production:

Resize_20231231_140824_4105.jpg (276.1 KiB) Viewed 2418 times

It would cost me about $1,400 for two of those, to be shipped to the US for an equivalent 30 watts and I have to assume they are selling.

[stephenz]

It would have probably costed a lot more if I had ordered from a machine shop.
I designed it to be as modular possible which obviously tends to make things more complex and costly.