regenerator design
regenerator design
Hi - new to this website . Have got V twin compressor as starting point & wondered if anyone has any advice on regenerator design/ideal transfer pipe size w.r.t. displacement volume
Re: regenerator design
According to many, stainless foil, placed in layers, spaced apart with shim stock, and bolted and sealed together. Copper would be nice, but thermal short circuits can occur. It must be thermally isolated from both the heater and the cooler, or thermal short circuiting on the outside of the device will happen, hence the use of stainless steel which has the worst thermal conductivity of any metal and a few ceramics, too. I picked this up by studying Andy Ross' book "making stirling engines".
Better to wait for me to make my share of mistakes on the design, for I am working on the problem. I will try to keep up my pace, but many obligations distract me from my project.
Better to wait for me to make my share of mistakes on the design, for I am working on the problem. I will try to keep up my pace, but many obligations distract me from my project.
Pssst! Hey you! Yeah, you. Over here....Re: regenerator design
Copper has better thermal conductivity (exceeded only by silver and gold) but at high temperature deforms easily, changing the gas physical spaces, and creating thermal short, also: oxidize on air, and the oxide does not lead well.
Is used stainless steel low alloy.
For your knowledge consult the tables of the internal conductivity of materials, and consider the availability in wires or thin sheets.
The foils or wires or wool forming the regenerator must be extremely quick to pick up the heat or pass it, so you need a high coefficient of thermal conduction; must also have a small mass so that the heat alters its temperature significantly in both directions in short time (heat exchange with the gas occurs naturally better with greater temperature difference).
For said reasons, it is clear that the thin foil or wire or wire wool, have little mass and large surface area, that also has a huge importance in the heat exchange with the gas.
The scientific calculation of a recuperator has in fact shown to be one of the most difficult.
Is used stainless steel low alloy.
For your knowledge consult the tables of the internal conductivity of materials, and consider the availability in wires or thin sheets.
The foils or wires or wool forming the regenerator must be extremely quick to pick up the heat or pass it, so you need a high coefficient of thermal conduction; must also have a small mass so that the heat alters its temperature significantly in both directions in short time (heat exchange with the gas occurs naturally better with greater temperature difference).
For said reasons, it is clear that the thin foil or wire or wire wool, have little mass and large surface area, that also has a huge importance in the heat exchange with the gas.
The scientific calculation of a recuperator has in fact shown to be one of the most difficult.
Re: regenerator design
Conductivity is the last thing that is needed in a regenerator on a hot air motor, what is required is the hot end to stay hot and the cold end cold, and the air going past to drop off and pick up heat.
The place I first learned about regeneration was studying open hearth steel furnaces, this is how the temperature is built up to melt the iron ore, and scrap mix. The regenerator in this case consists of chambers under the furnace, loosely filled with bricks. Ian S C
The place I first learned about regeneration was studying open hearth steel furnaces, this is how the temperature is built up to melt the iron ore, and scrap mix. The regenerator in this case consists of chambers under the furnace, loosely filled with bricks. Ian S C
Re: regenerator design
Ian: in this case I'm not in accord with you.
I think that in regenerator the conductivity is extremely important.
The conductivity is the property that admit the passage of the calour throug the mass of the material, raising quikly the temp. of all the mass, and equally extracting all the calour (and lowering the temp.) in equal very small time.
If the conductivity is small the heat do not pass "into" the material; the mass interested is only the superficial and for so little calour, the internal heat, costantly stored, produce a thermal inertia that prevents the rapid change of temperature, essential for heat exchange.
Aiding elements for this conductivity is small mass (small wires or wool or very thin sheets) so calour penetrate easily in all mass, and (of course) with really elevate surface of passage.
Instead I am in perfect accord for you about the material to be used, I think that copper is not adapt becouse oxidize easily, (oxides are insulating and compromises the conductivity), and overall copper is not mechanically strong at medium high temp., have a "negative hardening", (at a glance "softens"), underposed to the pression waves in hot air, collapse creating thermal bridges between wires or gauzes, (remember; have to be in VERY thin wires or wool) these accumulates not well removable calour that reduces the yeld; also circulation of the flow is compromises.
Excuse if I'm not correct in words of correct English.
Fer
I think that in regenerator the conductivity is extremely important.
The conductivity is the property that admit the passage of the calour throug the mass of the material, raising quikly the temp. of all the mass, and equally extracting all the calour (and lowering the temp.) in equal very small time.
If the conductivity is small the heat do not pass "into" the material; the mass interested is only the superficial and for so little calour, the internal heat, costantly stored, produce a thermal inertia that prevents the rapid change of temperature, essential for heat exchange.
Aiding elements for this conductivity is small mass (small wires or wool or very thin sheets) so calour penetrate easily in all mass, and (of course) with really elevate surface of passage.
Instead I am in perfect accord for you about the material to be used, I think that copper is not adapt becouse oxidize easily, (oxides are insulating and compromises the conductivity), and overall copper is not mechanically strong at medium high temp., have a "negative hardening", (at a glance "softens"), underposed to the pression waves in hot air, collapse creating thermal bridges between wires or gauzes, (remember; have to be in VERY thin wires or wool) these accumulates not well removable calour that reduces the yeld; also circulation of the flow is compromises.
Excuse if I'm not correct in words of correct English.
Fer
Re: regenerator design
About my bad use of English, I correct myself:
"Aiding elements for this CONDUCTION is small mass...."
"Aiding elements for this CONDUCTION is small mass...."
Re: regenerator design
I would not recomend steel wool, although it will work OK, it breaks down, and small bits end up all through the motor, stainless foil seems to be the material of choice by the big boys, those ones that publish their designs and ideas. Ian S C
Re: regenerator design
A few thoughts on this.
You need both high and low thermal conductivity.
The body of the regen needs to have a low conductivity and I can only come up with one answer to this and it is a high temp resin composite. Works in jet engines.
The higher the thermal conductivity of the substrate the better. But if your substrate is in contact with its self like steel wool or the foil examples I have seen you are setting the stage for a battle between hot and cold and given time one side will win. You need a thermal barrier between the connecting points of your substrate and a complete air gap between each piece. I think this takes us right back to the body being a high temp composite.
When trying to squeeze every last bit of efficiency out of the high speed transfer we are talking about shape of the substrate is critical.
Round screen looses out here because it is a compromise. Heat transfers into a material most efficiently on a flat surface but radiates out the best off of an edge. Screen wire is round. A weave of flat strips roughly as wide as the diameter of the screen wire should out perform the wire. It presents a flat face but also 4 edges.
There is no escaping Newton either, just as with any fluid the best place to extract heat/power is the outside of a turn.
At first I was looking at this from the stance that the bi-directional flow was a problem to be overcome but actually it offers a viable solution a single flow direction doesn't.
2 ideas I have
The first is problematic because it might restrict flow to much. But utilize flat planes sharp edges and directional change.
A flat rectangular box with flat plate vanes attached to the top and bottom in rows so that when you put the top and bottom together the top vanes fit down between the those in the bottom leaving a uniform air gap between all of them. The vanes would not go all the way across the box, but instead keep the same air gap used between them.
The problem I see with this is that while you maintain the same volume you drastically increase the distance the air must travel.
The flat sides also are problematic but it would be more difficult to make this as a cylinder and that would also decrease the surface area of the vanes.
This 2nd one is a better solution in my opinion butwill be the devil to explain in text form so I apologize in advance if I blow it.
It would have to be constructed split into a top and bottom oriented lengthwise.
The body would look like an s made up of 2 short radius u bends. It would also use vanes inside but the would face opposite directions in each end to work the same for the hot and cold ends.
The vanes would be curved pieces with a curled lip on the end facing that sets outside end. The would resemble a nautilus pattern and would be anchored in inside of the bend.
The idea is that as the air enters it first contacts the face of the curled lip and then into the spoon shape of the vane being pushed toward the inside of the bend of the body where it will run off of the vane and be carried back toward the outside to hit the next vane in the spoon. The curled edge turns toward the outside of the bend but does not touch the body it serves to scoop the air from the outside up into the curved inner spoon of the vane.
This works on both the flat plane and outside edge of a bend to maximize the scavenging of the heat/cold. When the air reverses it will now be pushed behind the vanes where it will contact the outer side of the spoones bend and then have to pass over the edge of the curled lip. A texture of small bumps ridges or spikes on this backside of the spoon ov the vanes would increase the transfer as well.
The bends would also help impede a short circuit thru the body though as I stated before the most logical material I can think of here is a high temp resin composite most likely using carbon re-enforcement.
Hope that was understandable.
I have never built a regenerator so this is just speculation based on experience with other applications that share some of these same issues.
You need both high and low thermal conductivity.
The body of the regen needs to have a low conductivity and I can only come up with one answer to this and it is a high temp resin composite. Works in jet engines.
The higher the thermal conductivity of the substrate the better. But if your substrate is in contact with its self like steel wool or the foil examples I have seen you are setting the stage for a battle between hot and cold and given time one side will win. You need a thermal barrier between the connecting points of your substrate and a complete air gap between each piece. I think this takes us right back to the body being a high temp composite.
When trying to squeeze every last bit of efficiency out of the high speed transfer we are talking about shape of the substrate is critical.
Round screen looses out here because it is a compromise. Heat transfers into a material most efficiently on a flat surface but radiates out the best off of an edge. Screen wire is round. A weave of flat strips roughly as wide as the diameter of the screen wire should out perform the wire. It presents a flat face but also 4 edges.
There is no escaping Newton either, just as with any fluid the best place to extract heat/power is the outside of a turn.
At first I was looking at this from the stance that the bi-directional flow was a problem to be overcome but actually it offers a viable solution a single flow direction doesn't.
2 ideas I have
The first is problematic because it might restrict flow to much. But utilize flat planes sharp edges and directional change.
A flat rectangular box with flat plate vanes attached to the top and bottom in rows so that when you put the top and bottom together the top vanes fit down between the those in the bottom leaving a uniform air gap between all of them. The vanes would not go all the way across the box, but instead keep the same air gap used between them.
The problem I see with this is that while you maintain the same volume you drastically increase the distance the air must travel.
The flat sides also are problematic but it would be more difficult to make this as a cylinder and that would also decrease the surface area of the vanes.
This 2nd one is a better solution in my opinion butwill be the devil to explain in text form so I apologize in advance if I blow it.
It would have to be constructed split into a top and bottom oriented lengthwise.
The body would look like an s made up of 2 short radius u bends. It would also use vanes inside but the would face opposite directions in each end to work the same for the hot and cold ends.
The vanes would be curved pieces with a curled lip on the end facing that sets outside end. The would resemble a nautilus pattern and would be anchored in inside of the bend.
The idea is that as the air enters it first contacts the face of the curled lip and then into the spoon shape of the vane being pushed toward the inside of the bend of the body where it will run off of the vane and be carried back toward the outside to hit the next vane in the spoon. The curled edge turns toward the outside of the bend but does not touch the body it serves to scoop the air from the outside up into the curved inner spoon of the vane.
This works on both the flat plane and outside edge of a bend to maximize the scavenging of the heat/cold. When the air reverses it will now be pushed behind the vanes where it will contact the outer side of the spoones bend and then have to pass over the edge of the curled lip. A texture of small bumps ridges or spikes on this backside of the spoon ov the vanes would increase the transfer as well.
The bends would also help impede a short circuit thru the body though as I stated before the most logical material I can think of here is a high temp resin composite most likely using carbon re-enforcement.
Hope that was understandable.
I have never built a regenerator so this is just speculation based on experience with other applications that share some of these same issues.
Re: regenerator design
Vicsmirf, a practical form of regenerator can be found in Andy Ross's book "Making Stirling Engines" (free down load), one example is an annular shape, and filled with .001" stainless foil, dimpled with a seamstress's tracing wheel (a disc about 1/2" dia with a star patern around it), not pressed hard enough to pierce the foil, this spaces the foil as it is wrapped on it's self. the space ends up at .008" to .010". Ian S C