Stirlingengineforum.com

First things first. I think we still need to find out if it even works at all. Maybe stephenz is right and my thermal imaging device is faulty, so now I'm rebuilding the engine so it can be tested with a thermocouple.

After the epoxy set I found that the flange on the glass cylinder interfered with the flywheel, so I had to take it to the shop and grind it down with a diamond bit on the drill press.

Just returning home now

Not a fan-boy, but I like Tom well enough — I can’t imagine why some Jackass would say he needs to “prove” what he has observed. Prove the obverse or shut the H. up. And if Matt has concluded an ambient engine is possible (I haven’t seen that, but obviously that’s what a no heat rejection engine leads to,) then let him speak for himself.

Thank you for that response Tom. It would be great if you were willing to share what you had come up with all those years ago. If not that is understandable. I also believe Carnot is wrong, but unfortunately the burden of proof IS on us, and the more scientific we can be, the better we can make our case.

Bumpkin wrote: ↑Sun Jul 09, 2023 7:48 pm Not a fan-boy, but I like Tom well enough — I can’t imagine why some Jackass would say he needs to “prove” what he has observed. Prove the obverse or shut the H. up. And if Matt has concluded an ambient engine is possible (I haven’t seen that, but obviously that’s what a no heat rejection engine leads to,) then let him speak for himself.

You're right Bumpkin, claiming any engine without 'cooling' does conjure up fanciful ideas. My voodoo cycle pitch was a way to explain no external cooling, but this may be wrong. Depending upon regen realities, voodoo may require external isothermal cooling similar conv'l gamma, but also might not require external cooling. I need to return to voodoo, but currently bogged down with conv'l gamma and six day work weeks. For now, I'll go with Vincent's voodoo conclusion.

BTW, no one replied to a post I made a while back that should have caught any ambient engine lover's eyes: within the same volume/ratios, 2 masses of gas expanding at 300k have the same power as 1 mass of gas expanding at 600k. The trick for any ambient engine is supplying the cold hole or in conv'l buzz, 'supplying' the sink vs the source.

I haven't read any Tesla stuff in decades, but I've always found this idea intriguing, since Wneg decreases with decreasing temperature. I remember seeing a few cryo studies that game this.

Just for the sake of completeness, I made some additional modifications to the engine.

I cut a disk of stainless steel screen from one of those cooking splatter screens for covering bacon and put it in the bottom against the hot plate as an additional regenerator, but mostly just to take up room as the regenerator on the sides was a little too shallow and loose, the screen on the bottom took up the slack without restricting air flow

I also found that the new glass piston was recessed on the bottom and wanted to fill this dead air space with something so I mixed up some JBWeld. While mixing the epoxy I got the idea to mix in some glass hollow microspheres I had been sent as a free sample from the manufacturer. I thought this would help block heat loss in this critical area.

After reassembling the engine completely, I tried various timing adjustments but the engine was barely running.

I found that when I tightened down the set screw it kept pulling the metal sleeve back into the old position. To keep it where I thought would be best I had to fight with it a bit, holding it in place with plyers and turning the set screw with another set of plyers.

I finally got the timing where I wanted it, and turned over the engine, it started up slowly at first.

I already had the thermocouples attached. One to the top of the engine and the other to a spare engine top to read "ambient"

The thermocouples were initially reading the same, I think 72°F

I had several thermocouples and found two that seemed to be very close. Read very near the same temperature when tested,by holding between my fingers or left near to each other in the open air.

As the engine was running, I looked down at the meter and the temperature readings were not the same anymore

The reading of the running engine had changed.

After a moment I saw the temperature reading of T1 was ticking down. (Reading colder and colder) maybe 1/10th of a degree every few seconds.

I was shocked and looked around for my phone but I had left it upstairs.

I looked back at the meter. It was still ticking down steadily so I tried to quickly get upstairs to grab my phone to record what was happening. I left the engine running.

I found the phone and got back downstairs, which took a few minutes. I was curious as I approached, if the engine had stopped or would still be running, it was still running, quite fast.

I had not prepared for making a video so the workbench is a wonderful mess, covered with Clay impregnated perlite left over from earlier high temperature displacer experiments.

I have to say, the thermocouples had been through some abuse, taking readings inside of a microwave and microwave kiln and other things, I had completely fried at least one, so there is a possibility the one attached to the engine was malfunctioning.???

Anyway, I started recording.

https://youtu.be/2oXIFSIdOfU

The phone was running out of memory.

The steam generator was running out of water which could have burned up the element but I wanted to see if the temperature would get as low as freezing, so kept recording though the video was becoming choppy.

I felt the top of the engine with my fingers a few times and it did feel cool, but I'm not sure about freezing cold.

When I accidently touched the bottom plate it was scorching hot.

Also I haven't put in fresh batteries in a while.

After the steam generator ran out of water and It off and the engine slowed down and stopped, I detached the thermocouple from the engine and it again returned to room temperature to match the other one after a few minutes.

I think maybe I need a new digital thermometer as this one is even more unreliable than the thermal imaging viewer.

Maybe I had the thermocouples plugged in backwards? Is that possible? So it was reading in reverse?

I tried to find the thermal camera through all this but I had left that over at the other workshop or someplace.

If you have any doubt on your thermocouples you should try to the read the temperature of boiling and melting water.

The average humidity in NY state today is 56%, which at 72F should give you a dew point of 53F.
I didn't see any ice forming but it might be hard to see.

I've seen plenty of people attempting to run LTD engines as refrigerators (or heat pump) by driving the flywheel to an electric motor, and I've never seen anyone get a temperature drop of more than 1C. Including that research paper I linked yesterday.

Anyway, if those readings are correct you'd be looking at over 1 W of cooling power.

how close my diameter and thickness estimates are?

A few questions.

What do you mean by "energy lost over time"?

Is that 1 watt per second, minute, the entire eleven or so minutes?

You have "energy lost over time" as 876 joules.

That represents about 8 seconds of heat input from the boiler at 100 joules/second (100 watt heating element)



Also, the boiling water below the engine is applying from a 100 watt heating element 100 joules per second.

You don't seem to be including that input heat in your calculations. To have a temperature fall would require first "cooling" the 100 watts/second input heat.

The way I figure it, there is an input of 100 joules/second. Probably less, but the steam is enclosed in a PVC pipe which is pretty good insulation and no visible steam was escaping so the majority of the steam was condensing on the bottom of the engine and giving up the same amount of heat needed to keep it at a steady boil.

At let's say 15% Carnot efficiency, (likely a very high figure, 20% at best) let's be generous and say 20% that leaves 80 to 85% of the input heat unconverted, so that amount of "waste heat" will need to be "rejected".

So, the cold side plate by that reckoning should be receiving for "rejection" about 80 joules per second.

It takes 4.18 Joules to raise the temperature of 1g of water by 1°C.

80 joules per second should, I would think, raise the temperature of the cold plate, maybe, I don't know, I'd guess probably 5° or 10°/second?

Even if the thermocouple was reversed and the change in temperature (seemed like about 1 or 2 tenths of a degree/second) represented a temperature increase, that increase is far below what is predicted by the Carnot limit.

Watt = Joules per second.

I am sorry for the lack of explanations. What I mean by energy lost over time is specifically the energy lost by the aluminum plate over the course of those 661 seconds (11 minutes or so). Why 11 minutes? I browsed through the video trying to find a spot in the beginning and in the end where the temperature is displayed.

i.e. at timestamp 0:16 I see 14.2 C (converted from whatever degree F value was displayed)
and at timestamp 11:17 I see at -1.1 C

I then estimated the mass of aluminum (as first estimation I ignore all losses including conduction losses through bolts and other parts of the engine) using density, diameter and thickness.

From there you can calculate the energy gained/lost through Q = m.Cp.DT which gave me a 876 joules.

Which averages to 1.32 J/s (Watt) over the course of 661 seconds.


That "energy lost over time" technically equates to the cooling power observed which makes absolutely no sense to me.
I encourage you to check your TC against melting and boiling water.

You didn't answer, was there any condensation or frost on the top plate?




edit: there is nothing worth calculating on the input really. All we know is what you pointed out: an approximate value of the heat input. By the way, was it a constant 100 W output, or would it just occasionally heat up based on internal temperature/thermostat?

Actually the steam generator I was using previously was 100 watts but after it ran out of water the ceramic heating element overheated and cracked and it stopped working.

I took it apart to find out what was wrong with it. It appeared to be very simple. Just a switch and the ceramic heating element attached to the bottom of the aluminium reservoir. It is a "facial sauna" intended for producing steam continuously.

I went and bought another similar facial steamer, but a different brand. I assumed it would be the same wattage but I just checked, it is actually just 85 watts.

I also read up on thermocouples and it is possible to plug them in in reverse but this will not be apparent as they do not measure temperature directly.

The meter has an internal thermometer that measures ambient which is used as a reference. The thermocouple is used to calculate the degrees of temperature above or below that and polarity does matter.

So while testing at ambient for two thermocouples that read the same the polarity had no effect.

I checked. There is a very small difficult to see (+) on one side of the yellow plug and another (+) sign on the meter. I did indeed have the T1 thermocouple plugged in backwards, so when it looked like the temperature went down a degree below ambient it was actually going up a degree.

I don't know why in the world they don't make these things with plugs with different size prongs that could not be so easily reversed, or color code them or something.

No, there was no ice formation. No chance whatsoever of that at 110°F or thereabouts.

stephenz wrote: ↑Mon Jul 10, 2023 4:26 pm Watt = Joules per second.

I am sorry for the lack of explanations. What I mean by energy lost over time is specifically the energy lost by the aluminum plate over the course of those 661 seconds (11 minutes or so). Why 11 minutes? I browsed through the video trying to find a spot in the beginning and in the end where the temperature is displayed.

i.e. at timestamp 0:16 I see 14.2 C (converted from whatever degree F value was displayed)
and at timestamp 11:17 I see at -1.1 C

I then estimated the mass of aluminum (as first estimation I ignore all losses including conduction losses through bolts and other parts of the engine) using density, diameter and thickness.

From there you can calculate the energy gained/lost through Q = m.Cp.DT which gave me a 876 joules.

Which averages to 1.32 J/s (Watt) over the course of 661 seconds.


That "energy lost over time" technically equates to the cooling power observed which makes absolutely no sense to me.
I encourage you to check your TC against melting and boiling water.

You didn't answer, was there any condensation or frost on the top plate?




edit: there is nothing worth calculating on the input really. All we know is what you pointed out: an approximate value of the heat input. By the way, was it a constant 100 W output, or would it just occasionally heat up based on internal temperature/thermostat?

I don't doubt your calculations, but the energy lost, if the thermocouple is reversed then becomes the energy gained by the top plate, which you say was just 876 joules over the course of 11 minutes or 1.32 joules/second.

Assuming the engine was receiving 85 joules of heat/second (85 watts heat from steam condensation input)...

It looks like about 80 joules in at the bottom plate and less than 1.5 at the top plate.

Previously in these kind of experiments I'd use a cup of hot water as the heat source, which would cool down quite a bit after being poured into a cold cup, then cool down over the course of the experiment.

This time I kept the water boiling under the engine because I imagined that the more power in, the more cooling, like a refrigerator. Turning off the heat input would be like unplugging a refrigerator, so I imagined.

The only thing I don't think I can agree with is where you say: "there is nothing worth calculating on the input really."

The electrical input is 85 watts continuous. Of course there are loses, but just 1.32 joules of heat arriving at the top plate for every 85 joules applied to the bottom plate seems to me like a very very small amount of heat "rejection".

BTW I'm pretty sure that the metal plates on this engine are not aluminium. I'm guessing they may be a low grade stainless steel, as I've had them a long time and they have gotten wet from running on steam or on ice many times, or have been stored in a damp basement but don't show signs of any rust.

Tom:

The calculations I put forward merely depict the temperature variation (in this case decrease) of the aluminum plate over some time. That's all this calculation is: a simple calorimetric characterization. Don't over analyze it, there is nothing there.

What I would normally expect from any stirling engine running off 100W of electrical power, would be a few Watt going into mechanical power and everything else as losses. I would expect the cooler plate to rise 10 maybe 20 C degrees over ambient and stabilize at some value. Since the cooler is only "cooled" by natural convection (which is a fairly weak phenomena). Doing a calorimetric characterization as the cooler temperature is rising would help characterize the heat the engine is rejecting, ideally this measurement would be done with the cooler plate well insulated.

But again, the main issue with your numbers here aren't so much about the fact that heat rejection is small but negative! In other words the cooler plate being lower temperature than ambient, the engine is not rejecting any heat but receiving some. Where is this energy going?

Lastly, keep in mind a non negligible chunk of your heat input will not make it to the working fluid and will be wasted before it gets it. These are losses through conduction of the various parts of the engine mechanically (thus thermally) connected to the heater plate.


Did you not see any frost or condensation forming during that 10+ minutes test? Based on your location (humidity), and temperature condensation should have started forming within the first 2 minutes of the video.


As for the top plate not being aluminum, I can update the calculation if you'd like using stainless steel density and specific heat. Can you confirm the dimensions? diameter, thickness?

Another BTW,

I plugged in all four thermocouples I have (checking polarity) and checked them in the boiling water. Three were reading very close to the same, the fourth one was way off, so I could check three locations on the engine.

I also have another meter with just one thermocouple, so that is four altogether, though I have not checked that meter but it is almost new and not used or abused.

I would like to try and get readings of the plates both inside and outside the engine. Not sure what could be learned from that but I'm just curious.

You suggested earlier that the ∆T of the working fluid inside the engine is very little.

Problem is the meter refreshed the readings infrequently compared to the RPM so rapid temperature fluctuations or differences would be pretty "washed out" but who knows, we might learn something.

I might also insulate the top and take readings under and above the insulation.

stephenz wrote: ↑Mon Jul 10, 2023 7:31 pm Tom:

The calculations I put forward merely depict the temperature variation (in this case decrease) of the aluminum plate over some time. That's all this calculation is: a simple calorimetric characterization. Don't over analyze it, there is nothing there.

What I would normally expect from any stirling engine running off 100W of electrical power, would be a few Watt going into mechanical power and everything else as losses. I would expect the cooler plate to rise 10 maybe 20 C degrees over ambient and stabilize at some value. Since the cooler is only "cooled" by natural convection (which is a fairly weak phenomena). Doing a calorimetric characterization as the cooler temperature is rising would help characterize the heat the engine is rejecting, ideally this measurement would be done with the cooler plate well insulated.

But again, the main issue with your numbers here aren't so much about the fact that heat rejection is small but negative! In other words the cooler plate being lower temperature than ambient, the engine is not rejecting any heat but receiving some. Where is this energy going?

Lastly, keep in mind a non negligible chunk of your heat input will not make it to the working fluid and will be wasted before it gets it. These are losses through conduction of the various parts of the engine mechanically (thus thermally) connected to the heater plate.


Did you not see any frost or condensation forming during that 10+ minutes test? Based on your location (humidity), and temperature condensation should have started forming within the first 2 minutes of the video.


As for the top plate not being aluminum, I can update the calculation if you'd like using stainless steel density and specific heat. Can you confirm the dimensions? diameter, thickness?

I think you missed my previous post:

viewtopic.php?f=1&t=5547&p=19690#p19686

The T1 thermocouple was actually plugged in backwards so there was no drop in temperature below ambient. As far as previously running on just hot water poured into a cup, I'm not so sure one way or the other. The engine has been modified at this point but I'll keep tinkering and testing and make sure my thermocouples are plugged in the right way next time!

You would not be able to measure sub-second temperatures without more advanced equipment. Those handheld meters and tips have a response time barely faster than 1 second anyway.

You should put 1 probe on the top plate, 1 on the bottom plate and suspend the last one about 1 inch away from the engine as to get an ambient reading (that will respond quicker than attaching it to a piece of metal which will smooth out the numbers but slow down the response time). Since I presume you will be filming, then you do one a fast response time.