Stirlingengineforum.com

I've been looking at this TMG alongside TK motor's, (Tristan's) engines for years. The internal design and structure of each are actually remarkably similar.

Both are stubby, relatively short and wide with a large internal "displacer" that barely moves, or in actuality, does not move at all.

Consider how tight fitting and heavy the TMG displacer is and the high frequency. The inertia is too great for such a tight fitting heavy displacer to move at such great frequency. IMO, it is likely simply suspended, not moving at all, acting only as a regenerator. TK motor's engine actually has the "displacer"/regenerator fixed in place.

The only real difference is TK motor's uses a crank, whereas the TMG has a linear generator.

Both use large wide diaphragm pistons that cover the entire top of the wide body engines.

Both run remarkably well, at high frequency. T k motors engine would likely run faster without the crank mechanism, and vice versa. The TMG would, or could, that is, would be forced to run in step with a crank rotation if it had that instead of a "free piston".

Remove the crank and a Stirling will run at a stable "natural" steady frequency. Add it back and the speed is more variable, controllable.

Other than all the minor, mostly inconsequential slight variations, the TMG with a copper diaphragm and this engine of Tristan's are, overall, functionally, pretty much, or nearly identical


https://youtu.be/r9lYsW0Df08


Another common feature is the simple, very inexpensive construction. The TMG is at a disadvantage in that respect. I believe the difficult elements, metal diaphragm and linear generator are unnecessary, or optional.

The copper could just as easily be a rubber membrane and the electrical generator a crank, take your pick, but the displacer/regenerator, likely does not need to move at all, and I think, probably doesn't in either.

So, potentially, both are just one moving part, the diaphragm piston, and similar also to mower of doom's flywheel free engine, with no displacer, though elongated.

The TMG appears to not have a regenerator, but the walls of the close fitting stainless steel pots serve the purpose

Tristan's engine is not so close fitting, the small gap is taken up by steel wool.

So, to that end, I found this set of SS nested containers online, along with the large sheets of silicone rubber, the idea is to make a kind of hybrid TK / TMG engine.

But I cannot continue to use the baby grand piano as a workbench. So the priority at the moment is finishing the workshop, which is why I was over there mixing concrete yesterday, and will be back there again today, pouring footings for some badly needed support posts down in the basement.

Some more info:-

Also, one other thing. Comparing the TK engine with Ted Warbrooke's Stirling 1, both have the regenerator chamber doubling back on itself, to one degree or another. This mostly just saves space over the elongated so called thermal Lag, laminar flow, acoustic engines. But also likely helps retain heat.

Having spent an hour or two reading about the Harwell TMG, I have formed one or two early impressions.

It seems to me that the devil is likely to be very much in the detail when it comes to making it and then making it work. The real Harwell engine has (had) a pressurised helium working fluid. Although it will work with just air at atmospheric pressure (at least I assume barumman's isn't pumped up), it is designed to operate at pressure, which is one reason it has such big 'o' ring flanges.

Also, curiously, the 'top hat' casing shown (and shown in barumman's video as a red case that fits over the generating coils) is actually part of the engine design in that it acts as the "bounce space" to utilise the otherwise wasted acoustic energy produced by the beryllium-copper diaphragm. The "bounce space" is a gas spring. How clever to utilise the space around the generator to improve the behaviour of the engine.

An enthusiast has built a Harwell TMG engine and reported that he struggled to get more than a few Watts out of it, but that is actually pretty good considering the professionally built originals barely produced more than 20 Watts running pressurised, helium filled and flat out!

Tom,

The worrying thing for me would be the beryllium-copper diaphragm. I thing that drilling it and forming it would require great skill and knowledge, and the inventor (Cooke-Yarborough) himself has published the fact that the TMG diaphragm operates at 65% of its theoretical fatigue limit even with such a small (1 millimetre) stroke. That, to me, indicates a level of tolerance that I couldn't achieve in my workshop. He acknowledged that this proximity to the fatigue limit was the main limit to increasing the performance of the design (which was actually pretty good - it would be fuelled with half a ton of propane at a time, once every two years and output a reliable 25 Watts continuously between refuelling every 24 months).

Significantly, barumman tells us on his video that he met Cooke-Yarborough and was able to discuss the finer points of the TMG's inner workings. Another member of this forum ('Luka") says that barumman based his design (the one you see working in the Youtube videos) on those discussions with Cooke-Yarborough. Given that fact, and given the fact that barumman is very clearly an extremely talented and capable builder of fine engines, I think the chances of success in trying to build a TMG myself would be essentially zero! I could get one built for me by engineering contractors (something I have a great deal of experience in doing) but the cost would be utterly prohibitive and the risk would be entirely mine!

So..... if you press ahead with your version of the TMG then I wish you all the luck in the world - you may need it! Having said that, if you succeed then that would be a truly impressive achievement indeed and a very real feather in your cap.

Back to the Ted Warbrooke ("Striling-1") type of Thermal Lag Engine (TLE), and the question of whether or not it can be scaled up (Tom thinks it can, and he may well be right. I have some reservations).


The thing is..... the TLE is a very simple piece of machinery that anyone can make in a shed in a couple of hours. It is a very forgiving design.

The Harwell TMG, in contrast, is not a simple machine at all and requires precision engineering that would take a much higher level of skill and a much higher standard of machining.

It is that mechanical simplicity - and whether it can be scaled up - that I am interested in.


(I'm using the word "simplicity" in this context to indicate it's amenability to making with basic tools by low-skilled enthusiasts. The underlying physics, however, is anything but simple - thankfully that isn't an obstacle to making a working example for very little money).

Having watched numerous "Hot air engine Expo" videos with what may be dozens of engines of virtually identical design, from tiny scale models, small, medium, and large models, of antique engines, running alongside the actual antiques, with 1", 2", 3", 6", 8", and up cylinder bore, the idea that any particular style heat engine could not be scaled up and down for any reason never really crossed my mind.

https://youtu.be/nl2CQDBUq3U

As far as the TMG being impossibly complex, difficult or prohibitively expensive...

I see in the TK engine, essentially the same engine as the TMG, but instead of using exotic, hard to engineer parts, it uses tin cans, balloons and coat hangers, but possibly runs even better than the TMG.

I've given consideration to the idea that scaling up could be an issue in that actual air molecules cannot be enlarged, so....

Using different gases, like helium or hydrogen, pressurization, molar heat capacity, certainly these kind of things relating to the properties of the working fluid itself, along with the properties of the materials the engine is built from, conductivity, heat retention, insulating properties, these things seem to have a much greater influence than size (or SA:V ratio).

Anyway, it's all theory and speculation to a degree

The situation with my workshop is, I've been collecting tools and equipment and materials for years, but no real room anywhere to set it up

Recently we purchased a commercial building nearby, very cheap, because the building was really on the verge of collapse.

The actual structural problems were not as bad as it appeared. So, a few small footings in the basement, and some extra support posts, and plenty of work space

I finished pouring the footings just last night, the damaged area of the building is otherwise supported on jacks until the concrete sets, and permanent support can be installed.

I can do some experiments with small models on the kitchen table, but anything involving turning on a lathe and machining in general, for bigger models or full size working engines will be delayed for some time.

In a few minutes I'll be heading over to remove the wooden forms around the support footings.

I can start setting up equipment in the basement once there is no danger of the building itself collapsing.

The actual structural problem turned out to be just a single mortis joint of a critical support beam around the staircase.

Tom,

the idea that any particular style heat engine could not be scaled up and down for any reason never really crossed my mind.

That is perfectly understandable. For example, all the Stirling engines in the Coolspring Power Museum video that you posted are all conventional - they have the usual two pistons (power, displacer) kept to a phase angle by linkages, and they all fit into one of the three recognised types (alpha, beta, gamma) and - as you say - come in any size you like.

But Warbrooke's and Thermal Lag engines in general are very different. They don't fit into alpha, beta, or gamma categories and don't have a physical displacer and they appear to maintain their phase angle naturally by stable resonant cycling controlled by the regenerator heat transfer time (consequently they don't need flywheels). Since that (hot end) part of the heat transfer duration is partly a function of SA/V of the regenerator matrix performance, scaling might be .........interesting.

Which is why, I suspect, you don't see larger versions. At least I haven't.

Alphax wrote: ↑Sun Feb 13, 2022 2:06 am Tom,

the idea that any particular style heat engine could not be scaled up and down for any reason never really crossed my mind.

That is perfectly understandable. For example, all the Stirling engines in the Coolspring Power Museum video that you posted are all conventional - they have the usual two pistons (power, displacer) kept to a phase angle by linkages, and they all fit into one of the three recognised types (alpha, beta, gamma) and - as you say - come in any size you like.

But Warbrooke's and Thermal Lag engines in general are very different. They don't fit into alpha, beta, or gamma categories and don't have a physical displacer and they appear to maintain their phase angle naturally by stable resonant cycling controlled by the regenerator heat transfer time (consequently they don't need flywheels). Since that (hot end) part of the heat transfer duration is partly a function of SA/V of the regenerator matrix performance, scaling might be .........interesting.

Which is why, I suspect, you don't see larger versions. At least I haven't.

Or, perhaps, maybe no one (or very few) ever tried.

The only sustained effort along those lines was/is derwood, that I've ever seen, and that, perhaps may have been too innovative. His "scaled up" thermal Lag went through several iterations, but I did not, I don't think, actually end up seeing any engine that proportionately looked like a thermal lag. His regenerator tube in particular seems disproportionately long, the piston diameter disproportionately large.

Possibly he did attempt a simple direct scaling up and it didn't work, I don't know. But other than that... I pretty much draw a blank, and I've pretty much looked at nearly everything over the past 12 years.

It is such a simple engine, it is difficult to understand why. I think that may just be because nobody every really understood or agreed on how it works. IMO it seems to violate the rules.

Few understand how ordinary Stirling engines work. Even fewer, maybe nobody understands "Thermal Lag" engines, and I suspect, many who think they do are wrong, so this lack of knowing how to proceed may be a barrier. Should the regenerator use more or less material? What size/length bore? Questions, questions, questions.

Could it scale up?

Nobody knows until somebody tries.

Did Ted Warbrooke himself ever try?

In the patent for the thermal lag, not exactly the same but close to the Stirling 1, I don't recall anything about any size limitations.

For some reason, I've also encountered an awful lot of negativity about the possibility of ANY Stirling engine EVER being worth the effort of scaling up. Too inefficient, too expensive, not practical, no power, no torque, on and on and on, so many millions of dollars spent, so many promising inovative new Stirling engines just around the corner, have come and gone, so people write it off as impossible or not worth bothering. Too much time and expense for nothing, and not much incentive to invest the necessary R&D, because a centuries old technology is likely unpatentable, or the patent wouldn't be worth much.

The BIGGEST, or one of the biggest disincentives is the advocacy of the MYTH of "Carnot efficiency", that claims it is an unavoidable LAW of the universe that ALL heat engines are inherently inefficient, and nothing can change that.

IMO it's all a bunch of bologna. A big lie, or perhaps just a supposition that went unquestioned for over a century and is assumed to be true. But if the idea of Carnot efficiency is applied to the thermal lag engine, it would be deemed "impossible".

You can still find scientific articles explaining why it is impossible for a heat engine to operate without a flywheel, The engineers that might be responsible for scaling up an engine for a big corporation that could afford to put the money into it, all believe in, and base their calculations on the presumption that the second law of thermodynamics, as it is claimed to apply to heat engines is valid.

So, proceeding based on false assumptions would likely fail. If "Carnot efficiency" is NOT a false assumption, then failure is guaranteed.

Ted,

You've made some interesting points, as usual.

I'd like to respond to them, as I think I agree with much of what you say. I'll respond one point at a time (and number them if anyone wants to join in).

[1]

Or, perhaps, maybe no one (or very few) ever tried.

I agree. Few have tried - or at least if they have they haven't reported findings.

Tom,

[2] About derwood's engine - agreed. I have one (a conventional gamma scaled up, not a thermal lag) and it too only works when heated to ridiculously high temperatures (cherry red). So in my case, the cumulative "faults" in the engineering are to blame, but I think any Stirling engine that struggles to run without colossal heat inputs has problems of one kind or another.

Tom,

[3]

It is such a simple engine, it is difficult to understand why. I think that may just be because nobody every really understood or agreed on how it works. IMO it seems to violate the rules.

Certainly is is beyond all doubt that there is no scientific consensus for how Thermal Lag engines actually work - as is clear from the academic literature (and strong contradictory interpretations from Tailer/West on the one hand and Organ on the other).

Personally I don't think it violates the rules - I just don't think the rules have been agreed or defined.

Tom,

[4]

Did Ted Warbrooke himself ever try?

It seems he did make several different versions and recommends to the readers (of his popular book) that they should experiment for themselves. But whether he was interested in scaling up remains unknown - none of his numerous followers seem to have either (see point [1] above).

Tom,

[5]

For some reason, I've also encountered an awful lot of negativity about the possibility of ANY Stirling engine EVER being worth the effort of scaling up.

I think that is because we live in a world where power is currently easy to get from IC engines, and direct comparisons between Stirlings and IC are (sadly) the default way of looking at things. The problem is that the specific power output per unit swept volume from a Stirling engine can never match that of a diesel or petrol engine. At least I know of no specific cases where they do.

If the objective is to get compact power delivery then Stirling's won't be the best choice (usually). Electric power and battery storage offers vastly superior energy density to Stirlings for moving things like vehicles. But Stirlngs do have lots of other applications, and I think we'll start to see more of these in the not too distant future.

Tom,

[6]

The BIGGEST, or one of the biggest disincentives is the advocacy of the MYTH of "Carnot efficiency", that claims it is an unavoidable LAW of the universe that ALL heat engines are inherently inefficient, and nothing can change that.

I wouldn't worry about Carnot. It isn't actually useful in designing working Stirling engines. It is useful for starting arguments, though, and is often the weapon of choice for those who like to criticise Stirling engines - their other weapon of choice being the comparison with ICEs.