Engine Pressurization

[VincentG]

Matt, thanks that was your best explanation yet of the regen pressure issues your always mention. Thats one reason I advocate ditching the regen all together. IMO there should be hot space, cold space, and anything else should be an insulated surface with minimal thermal mass to try and maintain gas temp until displacer exposes heat exchangers. You can't have a regenerator with out forcing a volume of air with average temperature at a time you don't want it.

I bet you have some fancy math to prove me wrong. But maybe real world testing is needed to validate this stuff.

[Bumpkin]

I’ve called the ratios different things at different times. There are times it matters and times it doesn’t. I favor Tom’s definition of compression ratio as the total system volume change; accepting that pressure is entirely different, with exchanger/regenerator effects on compression heating. Many years ago on the forum there was a debate on where to place the port in/out of the displacer chamber in a Gamma. It was generally agreed that it didn’t matter: If the hot air had to pass through to the cold end it would lose some expansion — if the cold air had to pass through to the hot end it would lose some contraction — compromise at the middle and lose a bit of both. So put it at the cold end and at least the piston can run cool. OK — I know you’re saying “duh.” But the thing is that the lower the displacement ratio is, the less that matters, since only a small portion of the air suffers that loss. So I think that for low compression designs, it’s just not that important.

On the thread Topic and for whatever it’s worth: With even the slightest blow-by, the closed crankcase pressure will equal the average cylinder pressure within seconds. A snifter valve can only raise the system pressure by whatever the operating pressure difference is, so it should have the most effect on a high-temp engine.

Bumpkin

[stephenz]

You can't have a regenerator with out forcing a volume of air with average temperature at a time you don't want it.

That's totally true, but what's the alternative?
Without the pre-cooling (or pre-heating) that the regenerator provides your heated expansion will start at a lower temperature (no pre-heating) and your cooled compression will start at a higher temperature (no pre-cooling).

The issue isn't the regenerator IMO, it's a genius device. The issue is there is no (simple) real world mechanism capable of separating the 4 steps of the cycle in a clean way. We're using alternators to convert to electrical power and converting transnational motion into circular motion which means sinusoides i.e. compromise. I think if we were able to compress gas solely in the cooling volume (i.e. without pushing it through the regenerator), and expand the gas solely in the heating volume (i.e. without pulling it through the regenerator), then the real PV diagram would already look much more like the theoratical diagram. Greater power would be extracted and greater efficiency would be achieved.

[VincentG]

Stephen, I don't know if you have seen the latest version of my LTD engine here, but I think it simply and almost completely solves these two issues. The spring dwell mechanism very simply and reliably solves both displacer dwell and timing issues, better I think than any rhombic drive could. This system would work great in a beta type engine as well. Then, the standard LTD layout with the flow port cut into the top of the displacer almost entirely solves the issue of the working gas being in contact with both exchangers at the same time.

I agree the gas then has no pre warming or cooling, but I think the gas will overall spend less time at median temperatures this way. It also seems to me that much effort should be focused on fluid flow to encourage good tumbling of the gas as the displacer moves. This way the gas will gain and lose temperature much faster.

Bumpkin, I guess there's many ways to skin the cat here. For simplicity's sake I'll revert to using the terminology that's more common here. I think at the end of the day it's the least of our concerns with these engines.

[stephenz]

Can you share a link? I'll look for it in the meantime.


edit: found it I think. Yes I came across the paper Matt Brown shared in that thread.
edit 2: I think that was a follow up thread

[VincentG]

This thread shows the spring, viewtopic.php?f=1&t=5516 but I guess I never posted a picture of the latest displacer made of xps foam but here is a picture that shows an earlier version. The displacer makes full contact with the cold plate on top and the gas flows through the channel that is cut out to expose the power cylinder. It seems to work very well.

displacer modification.jpg (392.14 KiB) Viewed 3743 times

[stephenz]

ingenious!

If I understand correctly, you make the displacer top/bottom out, so the position of the displacer looks something like this?

trimmed-displacer-stroke.png

have you given any thought to what is looks like in terms of energy loss/conservation?

[VincentG]

Similar to that graph but the transit time is much faster than power piston speed at any given rpm.

Sure, while some energy is lost in compressing the spring a part of it is recovered when expanding. How much I'm not sure but the spring should be as light as possible while still maintaining control of the displacer. But I think there is also a partial power stroke to be had with the displacer at bdc on the hot exchanger as Matt and I have discussed.

[stephenz]

That's really clever. As the spring compresses it gains potential energy which once released will result higher displacer velocity.
Getting the stiffness of the spring right isn't trivial but conceptually this has potential.

With that prolonged (time wise) "contact" at top and bottom, 2 interesting consequences come to mind:
- maybe the phase angle could have a new better sweet spot (different than 90 degrees)
- on a designer with a static regenerator (i.e. the displacer is not the generator), there might be a way to use the "time off" to make the displacer "block off" channels in and out of the regenerator and keep the heated/cooled gas solely in their respective volumes and almost no where else

[VincentG]

Somewhere i talked about how the timing is adjusted very effectively by the height of the displacer without actually changing the crank angle. If you watch the slow motion video you can see how far off the phasing is from stock.

I'd be interested in what you come up with for a "piston port" style of regenerator. I think it could lend itself well to dual one way regenerators. As a denouncer of regenerators I'll leave that up to you haha.

Edit: here's a link to said video. https://photos.google.com/share/AF1QipM ... pKY01faXl3

[stephenz]

Somewhere i talked about how the timing is adjusted very effectively by the height of the displacer without actually changing the crank angle. If you watch the slow motion video you can see how far off the phasing is from stock.

I'd be interested in what you come up with for a "piston port" style of regenerator. I think it could lend itself well to dual one way regenerators. As a denouncer of regenerators I'll leave that up to you haha.

Edit: here's a link to said video. https://photos.google.com/share/AF1QipM ... pKY01faXl3

I had found the slow motion video. It demonstrates the concept really well. As far as timing goes, some questions:
- your crankshaft still has 90 degree phase, right?
- you obtained the -time-off contact- between displacer and heater plate by making your displacer connecting rod longer than stock, right?
- you obtained the -time-off contact- between displacer and cooler plate by making your displacer taller than stock, right?
Mechanically speaking, playing with these 2 lengths allow you to create time off, but really the 90degree phase is still there, the location of the displacer between these 2 hard stops is still the same as it would without the flexure and the oversized rods/displacers - I mean, aside from the "bounce" created by the spring expanding.

So funny, I literally made a sketch of a dual 1 way generators concept this morning. Ports getting opened/closed by the piston and displacer in a beta configuration. This results in ports naturally running in a single direction. And being an advocate for regenerators, these 2 regenerators to transfer heat to each other. And obviously, steel doesn't make for a good solid to solid heat exchanger, so this 2-sided regenerator would need to be made of something more thermally conductive.

[stephenz]

If I understand correctly:
Obviously the increases are vastly exaggerated to illustrate it.

Fl;exure-Displacer.png

[VincentG]

You can view the height of the displacer above and below its centerline and its height in relation to the the displacer cylinder as the determining factors in timing control.

Yes the crank is still 90. But I would say the phasing is completely different than that and acts as if it was driven by a cam, like my first design used.

Could you explain what you mean by the 90 degree phase still being present?

Edit: here is my description of the height adjustments. I'm definitely not in the right head space to describe this again as my work on this is a winter time project.

"The spring is mounted mid height in the displacer. The height of the displacer is set to a certain percentage more than the height of the displacer cylinder minus the stroke of the displacer crank. The percentage over that number is what determines the dwell and timing of the displacer. The taller the displacer, the longer the dwell time, and the more retarded the timing is. The shorter the displacer, the shorter the dwell time and the more advanced the timing.

The timing can be swayed toward the hot or cold side by adjusting the center position of the spring. The only real disadvantage from the cam system is you cannot change one timing event without effecting the other. I think that is ok, considering the friction reduction."

[VincentG]

If T1 and T2 are "timing control points 1 and 2" then yes I believe you have it.

[matt brown]

The issue isn't the regenerator IMO, it's a genius device.... I think if we were able to compress gas solely in the cooling volume (i.e. without pushing it through the regenerator), and expand the gas solely in the heating volume (i.e. without pulling it through the regenerator), then the real PV diagram would already look much more like the theoratical diagram. Greater power would be extracted and greater efficiency would be achieved.

sq cycle.jpg (58.59 KiB) Viewed 3717 times

I really like this pic, except that the legend atop the event diagram needs adjustment towards the left. This diagram is widely used to explain basic Stirling cycle events, but I've never seen anyone call out some details. Assuming there was a mech that could achieve this, notice that the cold gas is compressing into regen prior high pressure blow. In an ideal regenerator, the heat exchanging would only take place during flow, so compression of gas into regen would be self limiting by the temperature of the gas wherever it's located (regen vs cold space). In this manner, the regen space is not a dead volume as commonly assumed. Furthermore, if we consider this represents a thermal ratio of 2x (300-600k) with ideal everything, then the gas density in regen is always 2x greater at hot end than cold end. During compression, this 2x density will continual and 'some' cold gas will be forced into regen. However, it's not as if this small amount of gas entering regen is causing massive backwork added to compressing gas. The key difference is that this diagram represents 'external' heating & cooling (the cylinders) that are isolated from the regenerator, and assuming ideal everything, there's no heat short. If this diagram represented internal heating & cooling (inline heater & cooler) than the dT from heater would always be shorting thru the regenerator, and adding to backwork during compression. In the real world, this heat short from inline heater is why most 90 deg V-2 compressor mods fail to perform.

This mass distribution of gas must be carefully considered. Continuing with this event diagram, when the high pressure blow occurs, the compression gas does not magically jump into the hot space. Nope, while the pressure might be the same in both cylinders (assuming ideal everything) the density is still 2x in cold space vs in hot space (for this 300-600k ex). Now, referring back to this event diagram, a sharp eye will notice that the 'high pressure' blow starts from a mid pressure value after compression. So, the pressure will increase in both spaces during the high pressure blow, but the only way for the cold gas to increase pressure is by further increasing density (further compression). The full impact of this 'constant volume' regen is best seen when carefully compared to constant pressure regen (like Ericsson eng).

Now, if we consider this event diagram to include valves isolating the regenerator, we find that the pressure after compression is greater than the pressure after expansion, thus compression work loss will occur when valves connect hot & cold spaces. Granted, no work loss when ideal regen with 0 volume, but in reality this would entail noticeable work loss.

I've been sitting on a mech that can run a clean cycle for 20 yrs, but didn't solve various regen issues until last year. This mech can run any cycle, but there's better ways to run Ericsson and Otto. I think most guys have bought into Stirling since it's nearly the only ECE discussed. I just don't see how even an optimized Stirling can produce realistic power when limited by 'isothermal' input and regen issues. In the end, I think the only ECE contenders are small Otto, mid-sized Ericsson, and large Brayton which is very similar to current ICE.