Electric cars are a pipe dream

[Billy T]

Yes it is. Because looking around I could find NO square tubes used for containing high pressure gas, only round tubes. That's not a coincidence Billy. ...

You are wrong. I proved that mathematically.

The reason that almost all high pressure tanks are round is (1) that round steel pipes are readily available and (2) there is rarely a need for "flat panel" tanks. It is very easy to show / prove that:
A flat panel tank can easily hold several times more high pressure gas per Kg of tank material that a round tank can.

E.g. instead of using square cross section ADJOINING "tubes" make their shared walls taller. That does not increase the tension in it (or any other wall) so the wall thickness remains the same, but if the shared walls are three times longer (taller if he panel tank is horizontal) then the volume of stored gas is three times larger than for the square cross section tubes panel tank, which was already 60% more efficient (in volume to weight ratio) thn the round tank.

Of course there is an increase in the tank mass to hold 300% more. I.e. now the two horizontal wall of each "tube" are still of length 1 but the vertical (or "shared walls") are three units tall (long) so the 3:1 rectangular cross section tank has mass of each tube increased by (3*+2)/(1*+2) = 5/3 compared to the square cross section tube, but hold 3 times more.

Thus the 3:1 rectangular tank efficient (still in volume to mass ratio) than the square cross section flat panel tank by (3/5)x3 = 1.80 ; but compared to the round tan, this 3:1 tank is 1.60x1.80 = 2.88 or 288% better than a round tank. (But I have neglected fact one of say 100 "tubes" has an "unshared wall") so instead of 288% better it is about 285% better than the round cross section tube in terms of volume stored to weight ratio.

This is simple math - It should not be beyond your understanding.

* Only one of the shared vertical walls is counted for each rectangular tube. -That is what "shared"means.

 

[adoucette]

You are wrong. I proved that mathematically.

No Billy you didn't PROVE that joined square tubes could hold more pressure per unit of mass. Indeed you haven't shown ONE square pressure vessel so we could compare the mass vs volume/pressure to an equivilent cylinder, let alone joined pressure vessels.

Since squares stack more compactly and thus you could transport more in square tubes vs cylinders they would already be in use (even without shared walls) if that were the case.

But no such SQUARE pressure vessels are to be found.

Arthur

 

[Billy T]

No Billy you didn't PROVE that joined square tubes could hold more pressure per unit of mass.

Yes I did - fact that you cannot follow the proof does not invalidate the proof. Note I am speaking of a "flat panel" tank, not a square array of joined square cross section "tubes."

... But no such SQUARE pressure vessels are to be found. Arthur

That may well be true. I said one of the additional reasons why the flat panel tank would cheaper to make (per unit of storage capacity) is that it would be extruded. - All 100 or so adjoined rectangles made together.

Their material used would be a matrix of binder and short carbon fibers for great tensile strength. If the "shared walls" can only be 0.5cm apart instead of 1cm apart to keep the wall thickness small, so what? Just use 200 parallel tubes instead of only 100 to make a "flat panel" tank about the width of a car's floor.

PS when you do finally understand the math, please be kind enough to admit you were wrong and I was correct. (Or show some error in the math analysis.) Trippy said he "doubted" my claim. He surely can follow the proof, even the one which I later gave showing the 3:1 rectangular Flat Panel than gives ~285% greater volume for the same tank mass. I hope he will comment as to which of us is correct.

 

[adoucette]

Yes I did - fact that you cannot follow the proof does not invalidate the proof.

No Billy, your "proof" was only about VOLUME, not about PRESSURE.

Big difference.

http://www.peesi.com/Pressure Vessel Examples.htm

And it is why all pressure vessels you can find are spherical or cylinderical and why you can't show a single high pressure rectangular pressure vessel.

Arthur

 

[adoucette]

when you do finally understand the math, please be kind enough to admit you were wrong and I was correct. (Or show some error in the math analysis.)

Your analysis was NOT about PRESSURE Billy.

You made the assertion but haven't shown any math around shapes that can contain stress from pressure, and that's the issue, not just volume.

Indeed, since it can easily be shown that stacked squares hold more volume they would be the ideal shape, but it's obvious they are NEVER used in a pressure situation.

Form follows function Billy and the people who make these pressure tanks never make them in a rectangular shape for a reason.

Probably the most obvious proof of this is the design of all modern passenger jet aircraft.
the fuselage is a pressure vessel.
The design goal is clearly to obtain the strongest shape with the least amount of material and it is always a cylinder.
Rectangular fuselages (ford tri-motor etc) went out of aircraft design as soon as they became pressurized.

As you can see in this cross section, not having square sides causes loss of usable space, but still there are no rectangular pressurized aircraft designs:

Arthur

 

[Billy T]

Your analysis was NOT about PRESSURE Billy. ...
Arthur

Stop being ignorant of the math (or just ignorant in general?) - if you don't agree with the math analysis, show where it is wrong.

I gave two reasons why the much more efficient flat panel tank is not used already in post 2143. Here is a third:

It is not as easy to more around as a round tube with the same mass (less volume), but it would be the floor of the car and never need to be carried any where separately.

Of course I am considering pressure when I speak of a pressure tank. My assumption is that both the flat panel tank and the round take operate at the same pressure - perhaps 20% below the pressure that would rupture them. At the same pressure, tank volume to mass ratio is the important factor.

Later by edit: I do have an error. While the 3:1 rectangular tank's "shared walls" would have the same thickness as they do in the square cross section tube tank, the outside flat surface walls would need to be three times thicker. I don't have time now to correctly re-calculate the advantage of 3:1 vs. square cross section flat panel just now but is not the great 285% advantage.

Just guessing: Instead of 60% better than the round tank, the 3:1 cross section tube may be only 75% more efficient than the same mass round tank.

 

[adoucette]

Stop being ignorant of the math

I'm not the one who is being ignorant Billy.

You only show math related to volume but you haven't shown any math related to the STRESS from PRESSURE and that's the key to the design of a PRESSURE VESSEL.

Which is why all pressure vessels are spheres or cylinders and you can't find any with rectangular sides.

Indeed the engineering calculators available for determining the wall thickness for pressure vessels don't even offer a calculation for anything but a cylinder.

http://www.engineersedge.com/material_science/hoop-stress.htm

But this little tidbit found in an engineering article on Pressure Vessels shows clearly why you are wrong.

All pressure vessel shells must be closed at the ends by heads (or another shell section). Heads are typically curved rather than flat. Curved configurations are stronger and allow the heads to be thinner, lighter, and less expensive than flat heads. pg 68

http://www.arab-eng.org/upload/uploading/PV.pdf

So that's a very clear statement as to why curved sections are used.

To give an idea of the magnitude of the difference there is this second instructional piece:

2.2 Hemispherical Heads - The required thickness of a hemispherical head is normally one-half the thickness of an elliptical or torispherical head for the same design conditions, material, and diameter

Which shows that just going away from a round to an semi-round eliptical shape, not even all the way to a flat shape, requires TWICE the material thickness.

Big OOPS Billy.

Arthur

 

[Aqueous Id]

I was just wondering if it is feasible to suppose that having generators (of some kind) attached to the wheels, rims, axle or driveshaft can be used to supply kinetic energy to the batteries. Here we have two ideas, the breaks and the energy from the naturally present rotation of the contact points to the movement of the vehicle itself. TBH, i dont know how the break idea would work, though i kind of understand the concept.

It is almost never feasible to recapture energy like this, because the losses in converting back and forth will simply waste what you could have used to get down the road. An exception is braking, where the momentum of the vehicle is changed, releasing energy. Then, instead of wasting it (as heat of friction on brakes) it is recaptured and used to recharge the battery. It is not efficient, but it can be done in such a way to produce a very good braking system.

 

[Billy T]

I'm not the one who is being ignorant Billy. ... Arthur

Then point out the error in the following "baby step" proof:

I need to do some corrections, especially for the 3:1 case, so make this analysis in “baby steps” for you to follow. I number each step so if you find mistake in this proof that the flat panel storage tank is much more efficient (holds more NG per unit tank mass) than the round tank does (both at the same pressure) you can easily tell where I made a mistake in this proof.

1) “Efficiency” is defined as the ratio of stored gas to the mass of the tank, but since both tanks have the same pressure in them, efficiency is the tank volume to mass ratio, or if they are uniform in cross section along their length; Efficiency, E = A/Pt where A is the cross section Area, P is the perimeter and t is the wall thickness. “t” is assumed to be a constant unless stated otherwise. (I’ll also take pi to be 3.14)

2) For a specific flat panel tank case, I’ll initially assume 101 small, square cross section “tubes” with the flat panel tank horizontal (like the floor of car). (I’m tired of putting quotes around “tube” just because the tube is square in cross section, so I’ll stop doing that. “Horizontal” as in a later post for the 3:1 rectangular flat panel tank case, with tubes adjoining along the 3 side, I can say things like “The top of each tube is 1 wide and it is 3 tall.”)

3) Comparing the efficiency of one square tube, one unit of length on an edge, to a round tube one unit in length in diameter. Note the tension in the walls of the square tube and the round tube is the same so the wall thickness is t for both with the same safety factor and same internal pressure. Their A/Pt efficiencies are 1/4t vs (3.14/4)/(3.14t) = 1/4t. I.e. identical. (The actual value of t will depend upon what the wall material and pressure are.)

4) Now note that if the diameter were 10, instead of 1 the volume would go up 100 fold, the P goes up 10 fold AND the tension in the circular wall goes up 10 fold too. So the bigger circular tank need wall thickness, T =10t. Hence A= pi (r ^2) = 3.14x5^2, and P = 3.14x10 and T =10t or A/PT = 25/100t = 1/4t.
SUMMARY OF POINT (4): Circular pressure tanks ALL have efficiency 1/4t, independent of their diameter. Thus I only need to compute the efficiency of the flat panel tank and compare to 1/4t.

5) Each of the 100 square tubes has only three sides (Think of adjoining square Us turned 90 degrees from normal. (The “fourth side” belongs to the adjoining square tube.) Thus for these 100 square tubes, 1 on an edge, A=1, P=3, and t is still t, so its E = 1/3t vs. the E of the diameter 1 circular tube, which point (4) proved was 1/4t. So for every square tube, part of flat panel set (except the one on one edge, which has identical E to any round tube), the square is more efficient by the factor 4/3.

The average efficiency of the 101 tube flat panel compared to any diameter round tube is thus:

{100x(4/3) + 1x1}/ 101 = (133.333333… + 1)/101 = (134.33333…)/ 101 = 1.330033…
Or slightly more than 33% more efficient.

Compared to a single big round NG tube of the type used today sideway in the trunk of a car, due to the highly curved ends it has, the square tube flat panel would be at least 35% more efficient as its tube ends are nearly flat.

I hope to do, accurately too, the 3:1 rectangular cross section flat panel tank of 101 tubes in a post soon, but want you to admit you were wrong first. – I.e. the flat panel storage tank is significantly more efficient (holds more NG per unit tank mass) than any round storage tank.

 

[adoucette]

Then point out the error in the following "baby step" proof:

Because you are only dealing with VOLUME and not STRESS.

Read up about it Billy and the design of pressure vessels is all about STRESS.

Manuals go on and on about it and what's clear is that as you use any shape that isn't round, the level of stress goes up and the wall thickness has to be increased.

Which I tracked down and for which you ignored:

All pressure vessel shells must be closed at the ends by heads (or another shell section). Heads are typically curved rather than flat. Curved configurations are stronger and allow the heads to be thinner, lighter, and less expensive than flat heads. pg 68

So the only question Billy is what part of that statement about Curved configurations being STRONGER than FLAT ones did you not understand?

Arthur

 

[Billy T]

Because you are only dealing with VOLUME and not STRESS. ... Arthur

False. I fully considered stress, but spoke of the wall tension to thickness ratio*:

...
3) Comparing the efficiency of one square tube, one unit of length on an edge, to a round tube one unit in length in diameter. Note the tension in the walls of the square tube and the round tube is the same so the wall thickness is t for both with the same safety factor and same internal pressure. ...

4) Now note that if the diameter were 10, instead of 1 the volume would go up 100 fold, the P goes up 10 fold AND the tension in the circular wall goes up 10 fold too. So the bigger circular tank needs wall thickness, T =10t.

Again instead of waving your hands, showing picture of air planes, etc. Tell what is wrong with the now numbered "baby step" analysis I made for you to be able to understand and tell where you think there is an error.

* Note I spoke in terms all readers could understand. Probably 3/4 could not tell you the difference between stress and strain so I did not use those terms.

 

[adoucette]

Here you go Billy.

From Non-Circular Pressure Vessels.

Comparison of the Rectangular Vessels with the equivalent Cylinderical (Circular Cross Section) vessels indicates the former are rather inefficient. Cylinderical Vessels will sustain considerably higher pressures for the same wall thickness and size. See Fig 20.

http://www.gowelding.com/pv/square.pdf

Fig 20 shows how considerable the difference actually is.
They aren't even in the same ball park.

And now we know why all the existing CNG gas storage tanks are cylinders.
Because rectangles are horribly inefficient in comparison.

Arthur

 

[Trippy]

Yes, there is: If the passengers remain in or near a car with the back seat (plus other parts?) with such an intense fire that enough heat passes thru the wall dividing the rear seat area from the trunk area and continues long enough to cause overheating (non-uniformly also so pressure relief value fails to function) of the NG tank, and die, that would be blessing for the human race - kept some real idiots from reproducing.

I'm going to give you the opportunity to rethink this statement before I reply. I'm also going to point out that the South Korean bus injured people, including bus passengers, and that the Honda Civic incident caused the recall because the Seattle Fire Department filed a 'near miss' report - an emergency responder could have been killed in that, and almost was (as a result of that incident they now have a set of counter measure in place in an effort to prevent such happening again.

SUMMARY:[/b] Amazingly the flat panel NG storage tank holds 60% more NG than the round one when both use the same amount of material, weight the same, and have the same bursting pressure limit.

Hmmm, manufacturing issues to one side, I can see a number of other potential problems with this - for example, the behaviour of pressure fields in corners.

 

[Billy T]

... Because rectangles are horribly inefficient in comparison. ...Arthur

More irrelevant hand waving instead of consideration of my math PROOF!

I'm NOT comparing one square tank to a round one. That is irrelevant! I am comparing a round tank to a flat panel tank of say 101 ADJOINING square tubes both at the same pressure and holding the same mount of NG.

The flat panel tank has 33% greater efficiency !!! Uses 33% less material.

I even explained to you intuitively why the flat ADJOINING square tubes of the flat panel tank are more efficient: - All but one of the 101 tubes SHARES ONE OF IT FOUR WALLS with the adjacent tube. Effectively 100 of the 101 tubes have only three, not four, walls. How dense can you be?

BTW, the single square tank with edge of 1 unit length is NOT "horribly inefficient in comparison" to a circular tank of diameter 1 unit of length. Point 3 of my post 2151 proves they have EXACTLY THE SAME EFFICIENCY, neglecting end effects, which I did not compute. You can let "1 unit of length" be mm, or an inch or a foot, a yard - whatever you like and this remains true.

 

[Billy T]

/To trippy:

I'll reply to the non-safety part of your post after you tell whether or not, in your opinion, my main point is correct.

I.e. am I correct that 101 square cross section adjoining tubes in a Flat panel tank holds 33% more NG at the same pressure than a round tank of the same material and weight?

This nonsense with Authur has gone on far too long, but he will not admit he is wrong. (I don't think he ever has.)

 

[Trippy]

/To trippy:

I'll reply to the non-safety part of your post after you tell whether or not, in your opinion, my main point is correct.

I.e. am I correct that 101 square cross section adjoining tubes in a Flat panel tank holds 33% more NG at the same pressure than a round tank of the same material and weight?

This nonsense with Authur has gone on far too long.

For the moment, I'm inclined to agree with Arthur, even if for no other reason that my experience tells me that corners focus stress fields, and so for that reason, if no other, it seems reasonable to expect that tanks that have a non-circular cross section would require additional strengthening.

 

[X-Man2]

Well first of all this is not a home unit, it's extremely expensive.

Secondly, 30 minutes is about 5 times as long as it takes me to fill up my car with gas, so while you think it's great news, it's not that thrilling to the average consumer.

Third, rapid charging is not recommended for more than 80% of capacity, so you won't get a full charge with one.

Fourth, rapid charging is not recommended if you want the longest life from the battery, and remember the battery pack is only expected to retain 70% to 80% of its capacity after 10 years .




Which would lower the LEAF's range to ~34 miles (@80%) in the last years of it's life.

Arthur

As usual you miss the point of all this,you see its an improvement in the right direction.Comparing charging to gasing up your car is immature as IC autos and gasoline,oil etc has had a century or more to improve and mature.Again no matter if its EV,NG,Compressed Air etc that replaces FF's and our transportation it must happen.And it happening will take multiple decades to improve and mature.Changing over our National transportation is a MAJOR undertaking and we will take 2 steps forward and one step backward well before we get there.

Your gas and oil that you idolize is killing us and our environment not to mention its running out (the easy cheap stuff) So it doesn't do any good to stay with the status quo.

 

[Billy T]

For the moment, I'm inclined to agree with Arthur, even if for no other reason that my experience tells me that corners focus stress fields, and so for that reason, if no other, it seems reasonable to expect that tanks that have a non-circular cross section would require additional strengthening.

OK - I will reply to your concern about corner effects:

We are speaking of a high pressure gas. The mean free path is very small so more than a few microns from the corner the "pressure field" you spoke of is the same as in the center of the tank.

I think it may be wise, even required to have a small radius of curvature on the inside at the corners to smoothly transition the stress from vertical to horizontal at the corners. Say a 1 mm radius of curvature "fillet" - that is very tiny bit of extra weight on 1 inch flat wall or greater tube. - Certainly less than the extra mass the more curved end caps of a single circular tank would have.

The two extreme outside sides of the flat panel probably need to "bow outward" some too -perhaps they are 10% longer than the internal divisions which are strictly flat except for the tiny "fillets" where they join the external walls. I am just guessing, but I will lower my claim to say the 101 adjoining tubes flat panel tank is only 30% more efficient than the round tank. I ignored several of these minor effects but strongly doubt they make more than a 10% reduction in my computed 33% better efficiency.

This is proving to be a good example of where relying on one's "experience" and intuition and ignoring math analysis is sometimes very misleading.

Please look at the math analysis, the "easy to follow "baby steps" of post 2151, instead relying on only your "gut feel."

 

[Trippy]

OK - I will reply to your concern about corner effects:

We are speaking of a high pressure gas. The mean free path is very small so more than a few microns from the corner the "pressure field" you spoke of is the same as in the center of the tank.

I think it may be wise, even required to have a small radius of curvature on the inside at the corners to smoothly transition the stress from vertical to horizontal at the corners. Say a 1 mm radius of curvature "filet" - that is very tiny bit of extra weight on 1 inch flat wall or greater tube. - Certainly less than the extra mass the more curved end caps of a single circular tank would have.

The two extreme outside sides of the flat panel probably need to "bow outward" some too -perhaps they are 10% longer than the internal division which are strictly flat. I am just guessing, but I will lower my claim to say the 101 adjoining tubes flat panel tank is only 30% more efficient than the round tank. I ignored several of these minor effects but strongly doubt they make more than a 10% reduction in my computed 33% better efficiency.

This is proving to be a good example of where relying on one's "experience" and intuition and ignoring math analysis is sometimes very misleading.

Please look at the math analysis, the "easy to follow "baby steps" of post 2151, instead relying on only your "gut feel."

My 'gut feel' and 'experience and intuition', in this instance, is based on math analysis (stress-strain analysis has a big role in structural geology).

I'll try and respond in more depth at some point later.

 

[adoucette]

More irrelevant hand waving

No hand waving Billy.

I posted an actual engineering chart that shows how much thicker the walls have to be for a square tube vs a cylinder and a clear explanation of the inherent value of cylinders over rectangles for storing gasses under pressure:

Comparison of the Rectangular Vessels with the equivalent Cylinderical (Circular Cross Section) vessels indicates the former are rather inefficient. Cylinderical Vessels will sustain considerably higher pressures for the same wall thickness and size. See Fig 20.

It clearly says that Rectangular vessels are rather inefficient and that cylinders will sustain CONSIDERABLY higher pressures for the same wall thickness.

What part of that don't you understand?

Now if you want to you can use THAT chart in the document I linked to (fig 20) to figure out the width of the walls needed to construct your rectangular tubes sharing one wall, but you will quickly see that even with that form they are nowhere near as efficient as cylinders because for the same thickness, cylinders can handle CONSIDERABLY higher pressures.

Have fun.

Arthur