piston rings
Dorkmo
gets bought?
Mark Norton
They can also be turned on a lathe, usually from a cast iron billet. The main issue here is finding or buying the billet. If you have cylinder with a 4" bore, you're going to need a billet 4" in diameter, which if you think about it, might be pretty hard to find. Smaller bores seem more reasonable. Once cut from the billet as a ring, they are snipped to allow them to be opened a bit and fitted into the piston head grooves. Butt seam rings aren't the best, but if you use two and offset the seams, things won't be too bad.
Much of this was written up on the OSE wiki. Search on piston rings.
- Mark
Ken
A really nice solution are gapless rings:
http://www.totalseal.com/TechPage.aspx
Clupet rings have found a lot of antique auto and steam engine fans.
The company doesn't have a website but contact information is readily
googled. They'll make rings to spec and it's hard to see any ring
sealing better:
http://shopswarf.orconhosting.net.nz/clupet.html
Regards,
Ken
>
> - Show quoted text -
Dorkmo
http://www.jegs.com/i/Sealed-Power/844/E251X/10002/-1
would they be good?
how do u figure what size piston goes in a 4" bore?
+
how deep and wide do u cut into the piston for the ring to sit in?
i saw your cad drawings but i think it said TBD on some stuff.
Ken
inches total.
There is no correct answer to what size piston goes in a bore, largely
depends on the temperatures the piston sees and what materials the
piston is made of, for this you need to look up the thermal co-
efficent of expansions for both the piston and bore and determine what
the hot running temps will be. Calculate the hot bore diameter, then
subtract about .0015-.002 inches for the hot piston diameter, then
calculate the cold diameter from that. Always measure bore and piston
at same temperatures. You can check results by putting piston and
cylinder into near boiling water and letting heat thoroughly, check
that piston slides into bore, measure both to see if diameter
differences are OK.
The piston ring grooves should be deeper than the radial distance
between the ID and OD of the piston ring. If you place the ring
alongside the piston such that they look like this OO and then slide
the ring into the groove, a metal rule should not contact the ring
when the rule is laid directly onto the piston lands. No need to go
wild on depth, but a few thou would be nice.
Regards,
Ken
Mike Stone
A lot of the references I'm reading suggest a piston speed of about 600-1000 feet/minute, and that's a handy place to start figuring cylinder dimensions.. 2 * stroke length * revs per minute = 600-1000.
Is that speed still considered a good target, or have we moved up as materials improved?
Bastelmike
for combustion engines that need to last, 10 m/sec is considered the
upper limit. Racing car engines and such use up to 15m/sec, but that
are engines that last only a few hours. Long-living diesel as in
trucks, tractors etc, usually operate slower than 10 m/sec.
For a steam engine built without high precision tools I would consider
5m/sec as the upper limit to give it a long lifespan and reduce
vibrations.
5 m/sec are 300m/minute are 1000ft/minute, and I guess we should
remain below this value.
Mike
Dorkmo
Ken
Cylinder will need to be very polished to prevent rapid wear.
Temperatures will need to be moderate to prevent breakdown.
Being a plastic, teflon is going to be much more prone to deformation
under load and heat, potentially squeezing out.
This brings up a topic no one has discussed so far: lubrication.
Typically there are two lubrication issues with steam engines; the
lower end assembly such as crank, bearings, conn rod and so on and the
cylinder, the piston needs to be lubricated and so do the valves if of
the piston, slide, rotary, Corliss variety.
Some single acting poppet and bump valve engines eliminate the upper
end lubrication by allowing crankcase lubricants to splash onto the
piston and cylinder bore, this simplifies matters but brings up the
secondary problem of blowby steam contaminating the lube oil. Double
acting (and some SA) engines use a piston rod and lower cylinder
stuffing box, sometimes the crankcase is a separate enclosed unit with
another stuffing box on the piston rod....preventing contamination of
the crankcase. When lower end splash lubrication isn't available, a
steam lubricator is necessary to add oil to the piston and in the case
of valves which have rubbing surfaces the oil is injected into the
valve or steam chest. This provides lubrication but has a couple of
issues, the oil travels out the exhaust to the condenser (where it can
reduce performance) and then onto the boiler, where the oil can
potentially break down and form insulating layers capable of
accelerating tube burnout.
I suspect that there are enough issues involved in building a first
time steam engine without trying to solve the problems of plastic
rings as well...I don't believe any commercially available heat
engines use them and have to go with the theory that they might have
good reasons.
Regards,
ken
> > > - Show quoted text -- Hide quoted text -
Mike Stone
Are there any issues with using graphite as stuffing? It's chemically inert, self-lubricates, has a low coefficient of thermal expansion, and doesn't suffer thermal breakdown until about 3000* C.
It's soft and brittle, so I wouldn't want to use it for anything structural, but stuffing doesn't need to be strong. I assume you'd want to polish the wearing surface that runs through it to reduce wear, but it's easy to polish a shaft.
As I think about it, I doubt you'd even need solid graphite. A putty made from powdered graphite and a little high temperature grease could be packed into the stuffing box and put under moderate pressure by the gland. That would squeeze it into shape and distribute the pressure more or less evenly across the whole contact surface.
That sounds like a lot of potential benefits, so what are the problems?
Paul Passarelli
--Paul
Ken
shaft packings.
http://www.mcmaster.com/#graphite-packing/=f0o6z9
I used to use graphite tape as packing seals in the navy 30 years ago,
worked just fine.
Regards,
ken
Paul Passarelli
In another thread the question of graphite stuffing was asked. Once again we are in a diverging situation between the low-tech options of graphite and the high-tech engineering plastics.
This is a paradigm that we'd do well to choose one side or the other but probably invite failure to try to straddle and combine. While a graphite slug might perform superbly in a high-tech ultra-modern design, the need for a "whisker strengthened, vapor deposited", gonkulator widget in an "any man" design will surely stomp on the brakes.
I feel a decision in this area is worth debating.
Mike Stone
> In another thread the question of graphite stuffing was asked. Once again we are in a diverging situation between the low-tech options of graphite and the high-tech engineering plastics.
I agree that it's worth discussing the relative merits of contemporary materials versus traditional ones. It might even be worth creating separate categories for designs as we go down the road.
I'd debate the modernity of graphite though. Good ol' plumbago is a common ingredient in recipes from the late 19th century.. right up there with mercury, the various oxides of lead, nitric acid, and potassium cyanide. ;-)
Dorkmo
really reasearch how it would hold up over time. they coat cooking
pans with it? :P thought they say u shouldnt use metal utensils with
it.
with regards to the oil issue, would having a gear pump and then some
sort of squirting device to shoot up from below the cylinders work? is
that feasible?
and for water contamination if there was a pipe that went down from
the oil reservoir then turned back up to the oil line would the oil
separate out and the water could exhaust out. i've herd this called a
frog leg?
Russell Philips
expensive.
How much would a 20mm chrome piston ring cost?
What would be a good matching cylinder sleeve material? hmmm
What are the characteristics and dynamics of chrome on chrome?
Is there an oil-less piston/cylinder solution?
Will steam "lubricate" some(any) combination?
Max Kennedy
It can be done
Mark Norton
- Mark
Ken
I only know of two lube free steam engines. Enginion (a VW
subsidiary) patented some materials allegedly capable of doing the
job, but the foundry work looks well beyond backyard capabilities.
Harry Schoell's Cylone uses water lube and plastic rings and bearings
but he has patents, isn't likely to give away trade secrets and the
since the engine hasn't been subjected to third party long term dyno
testing we can only guess at how successful he's been.
Chrome rings would be good on nitrided steel liners.
Chrome liners would be good with moly rings.
Squirting oil up the cylinder bore is common automotive practice.
Regards,
ken
On Nov 21, 7:38 am, Russell Philips <russellphil...@hotmail.com>
wrote:
Ken
far lower than oil, oil can support a large load and prevent metal to
metal contact while water is much less able to do so unless huge
bearing surfaces are contemplated. The water lube in the Cyclone
can't provide hydrodynamic support, the concept is to aid the low
natural friction of the plastic and carry away heat. This is a really
fascinating research subject, but one that will likely take time and
money to solve, which I think is antithetical to the design goals.
Regards,
Ken
On Nov 21, 9:14 am, Max Kennedy <mekennedy1...@gmail.com> wrote:
> Can one use a thin film of high pressure water being ejected from the
> piston head sides to replace both rings and lubricant? Think of an air
> hockey table but using water instead of air? Eliminates oil contamination
> problems. Just floating an idea if you'll pardon the pun.
>
> Max
>
> On Mon, Nov 21, 2011 at 7:38 AM, Russell Philips <russellphil...@hotmail.com
> It can be done- Hide quoted text -
Mark Norton
Ken
previously on the market that didn't need some kind of cylinder
lubricant. Oils are handy, economical and understood. Water presents
an idea solution, but the goals of this group would seem to mitigate
against lengthy and costly development in favor of the readily
obtainable; so at the moment I really see no viable alternative.
Regards,
Ken
Russell Philips
that oil-less was possible/probable. So it appears that lubrication is
essential and therefore any exotic material combination moves out of
project scope. Splashing crank case oil is 'good.enough' for the ring/
cylinder lubrication. Standard materials are fine.
This bends me back towards a very conventional style of engine (single
piston instead of dbl acting). So, it appears that standard off the
shelf rings will do the job.
I'm good with adding oil lubrication as a design requirement. (Mark/
others? - decision time?)
Mike Stone
> Can one use a thin film of high pressure water being ejected from the piston head sides to replace both rings and lubricant? Think of an air hockey table but using water instead of air? Eliminates oil contamination problems. Just floating an idea if you'll pardon the pun.
Your basic idea is good.. such bearings exist and can support huge loads. The oldest Michell/Kingsbury fluid bearing in the US weighs 2-1/4 tons in its own right, carries about 200 tons of combined load, and has been in service since it was installed in 1912. Its estimated maintenance-free lifespan is somewhere around 1300 years:
http://en.wikipedia.org/wiki/Fluid_bearing
Air bearings also exist and are popular for high-precision applications.
The devil's in the details though, and for this specific application the details suck.
First, a fluid bearing is a machine in its own right. You need machinery to make it work, and space to put everything. A feed system that works from inside a fast-moving piston would be nearly as complicated as the engine itself.
Second, keeping the pressure balanced within the gap is tricky. The whole reason fluid bearings work is that if one part of the load gets closer to the bearing surface than another, the pressure in the thin region becomes higher than the pressure in the thick region. For small to moderate imbalances, that pushes the load back to its proper position. If the difference gets too big, the thin area pushes the load away really hard, momentum carries the load past the correct position and too close to the opposite side, and you get the same problem all over again. If things get really bad, the swing gets larger every time and the load 'crashes'.. comes in contact with the bearing surface itself. In general, fluid bearings like rotation or smooth linear travel. They don't like shock loads or reciprocating motion.
Third, you never want free water inside a steam cylinder. It steals heat from the working steam, costing you pressure.
Mike Stone
> I'm good with adding oil lubrication as a design requirement. (Mark/others? - decision time?)
The higher-output engine is outside my scope, so I'll leave that decision for the stake-holders.
At the low-output side, I'm thinking about a DA cylinder, a piston rod that runs all the way through the cylinder, and stuffing boxes at both ends. That puts most of the load support at the stuffing boxes, which at least stay in one place.
Moving the support/alignment issues out to the stuffing boxes should (hopefully) allow me to make the piston simpler. Right now I'm toying with the idea of a sacrificial piston ring made of softer metal.. probably copper or brass. My "wouldn't it be nice if" design involves wrapping a few turns of 12-gauge copper wire around the piston and letting that wear to fit. Maybe bed it in a layer of graphite impregnated grease. The grease would be there to reduce pressure loss between the cylinders more than to limit friction.
Yeah, it would wear quickly compared to automotive grade piston rings. Hopefully a good polish inside the cylinder combined with moving most of the heavy load off to the stuffing boxes would mitigate that to a reasonable degree.
I think it's a design that will degrade gracefully. Excessive wear should show up as a loss of efficiency rather than a ruined cylinder, and let's face it, the thermodynamics of a small cylinder (1x1 or less) aren't that good to start with.
Finally, it has the virtue of being cheap and easy to fix. Twenty minutes, a foot of Romex, some simple hand tools, and you're good for another MTBF.
Jason Learned
Jason
Max Kennedy
Max
Jason Learned
http://en.wikipedia.org/wiki/File:Scotch_yoke_animation.gif
Seems like a great design for steam.
Jason
Jason Learned
Ken
Not all Scotch yokes use two opposing SA cylinders, I'd hesitate to
say that might even be the more rare instance, there have certainly
been any number of attempts at DA cylinders and Scotch yokes.
One thing to keep in mind about reinventing the wheel...a lot of
people have spent much time and money trying to do so...and yet the
wheel basically remains unchanged because it is very good at what it
does. The Scotch yoke has some nice theoretical advantages including
the elimination of secondary unbalance shaking forces. When you get
right down to it, though, the Scotch yoke is an attempt to somehow
justify the act of trading rolling friction for sliding friction. We
get an inkling that this is maybe not a real popular endeavor when we
look at the wide variety of ball bearings, needle bearings, roller
bearings and so on used in linear transport systems to eliminate
sliding friction...serious engineers are putting a lot of effort into
doing exactly the opposite of what a Scotch yoke does. As a trivial
exercise, make sure your hands are dry and try rubbing them together
with all the pressure you can muster. Now put a pencil between your
hands and do the same thing.
Here's part of the problem, even when using oil lubricated Scotch
yokes...as the piston approaches 50% stroke the pin velocity
approaches zero, and according to the Langmuir theory of lubrication
is is the motion of the parts that creates oil wedges that float the
parts upon one another. So, for a large portion of each stroke you are
going to have large loads and minimal lubrication. By contrast, with a
connecting rod the piston speed might approach zero at TDC and BDC but
the rotating velocity on the rod is still high and the oil films are
undisturbed.
As shown, the Wikipedia animation is fairly unworkable for a engine of
any size, they show the crank pin riding directly in the yoke. This
means there is very tiny contact surface where the pin and the yoke
meet, certain recipe for mechanical failure. More successful Scotch
yokes had a slide working in the yoke, with the pin turning inside the
slide. After these few hundred years, however, connecting rods
dominate by such a huge margin as to relegate other mechanisms to
mechanical curiosities simply by the fact they are so good at what
they do.
Regards,
ken
On Nov 21, 8:44 pm, Jason Learned <jasonlear...@gmail.com> wrote:
> In a scotch yoke engine two pistons are directly opposite each other. The
> power stroke on one side compresses the other piston. So in essence you
> have a double action piston that has been split into two putting the drive
> shaft in between.
>
> http://en.wikipedia.org/wiki/File:Scotch_yoke_animation.gif
>
> Seems like a great design for steam.
>
> Jason
>
> On 22 November 2011 02:15, Max Kennedy <mekennedy1...@gmail.com> wrote:
>
>
>
> > Why does a scotch yoke eliminate a need for packing?
>
> > Max
>
> > On Mon, Nov 21, 2011 at 7:50 PM, Jason Learned <jasonlear...@gmail.com>wrote:
>
> >> I still think using a Scotch yoke design would be great for a DA engine
> >> with no packing required.
>
> >> Jason
>
Mike Stone
> In a scotch yoke engine two pistons are directly opposite each other. The power stroke on one side compresses the other piston. So in essence you have a double action piston that has been split into two putting the drive shaft in between.
>
> http://en.wikipedia.org/wiki/File:Scotch_yoke_animation.gif
>
> Seems like a great design for steam.
The animation in question is kind of misleading. You don't actually need push rods on both sides for it to be a scotch yoke. As shown, it's an 'opposing piston' engine, which is indeed a worthwhile idea.
The scotch yoke per se is the 'pin sliding in a slot' part at the center of the image. It converts reciprocating motion to rotation, and is really defined by having two sliders (one at the piston and one at the crank) and two rotating joints (one at each end of the crank).
The packing has a different job. It's there to let the piston rod slide back and forth while keeping the steam inside the piston.
I'm thinking about making the packing strong enough to support the piston, but that's in addition to its primary job.
Jason Learned
I was using the Bourke engine design as my base. His triple slipper bearing seamed to take care of the lubrication problem and his engine's purported performance is very high. It was a shame he died just as he started to manufacture. Maybe his version of this engine would suit our needs, if his bearing will not work because of the lower rpm in a steam engine versus his 6,500-15,000 rpm then I guess it is back to packing. Either way I thought it worth seeing if it could be made feasible. I have a set of Bourke's plans to use as a guide.
Best regards,
Jason
Ken
The Bourke engines reported performance is indeed higher than those of
conventional manufacturers. I suspect every engine discussed or
advertised on the internet is also better than conventional engines.
One of the SACA fraternity classifies these are 'cartoon engines',
lots of drawings and specs but no real hardware.
The typical scenario for all these things goes something like this:
Inventor A comes up with a fabulous idea but the (stupid, greedy,
corrupt) corporations won't build it because (they are mean, will cost
money, stupid, prevent them from leeching off society) and therefore
lacking funds he builds it on his own. The initial tests looked good
but (he died, went broke, the engine developed so much power that it
went into orbit, they bought him out) and thus the potential has never
been realized. Usually there is some hook for money in here somewhere
unless wer're dealing with actual true believers. If there is a
strong emotional content to the narrative, check your wallet, fast.
Truth is, Bourke's wonderful detonation idea is something that people
have been trying to prevent for over 100 years. It's why octane
ratings are important, and why higher compression engines need higher
octane fuels. Detonation is typically associated with blowing things
up, which is precisely what happens to Bourke engines, they don't work
much before they turn into scrap metal. IC engines aren't 'explosion
motors' like the old timers called them, but instead the fuel/air mix
is used as a propellant by regulating the propagation velocity. The
rest of the Bourke engine isn't much better mechanically, there is
nothing there that a connecting rod doesn't do cheaper, more simply
and reliably.
Might as well get the rest off my chest (sorry for picking your post
to dump on, it isn't personal, but when you work where I do you get a
lot of people telling you about this great thing they saw on the
internet and questioning why you guys don't do better).
The Tesla turbine is a joke, a bad joke at that, the efficiency is so
low that it is arguable whether the engine can drive its own feed
pump.
There never was a magic carburetor at GM, no one from the company came
and took the carb off someone elses uncles car that up to that point
was getting unheard of mileage. R and D is done at the Tech Center,
miles from a factory, using cars assembled or modified on site.
Development is not done in the factory, and it isn't done on cars sold
to the public.
Barrel engines, wobblers and the like all look good, but the sliding
forces or extreme angles at which you need to apply forces make them
less than competitive.
The Green steam engine...better just leave that alone before I get
really worked up.
Anyhow, the key thing to look for in unusual alternative technology
engines is dyno tests.. Lots of them. Testing for efficiency,
performance and reliability. Said testing done in front of impartial
witnesses who have no fiscal stake in the outcome and are also
technically conversant with dyno testing (like the SAE).
I hope I didn't come off as abrupt, that sure isn't my intent. I tend
to get irate because I continually see people being talked into
putting time, effort and money into things that don't have the track
record which has been portrayed. Building an engine on your own is a
big enough challenge without excess baggage being tossed in.
Regards,
Ken
On Nov 22, 8:22 am, Jason Learned <jasonlear...@gmail.com> wrote:
> Hi Ken,
>
> I was using the Bourke engine design as my base. His triple slipper bearing
> seamed to take care of the lubrication problem and his engine's purported
> performance is very high. It was a shame he died just as he started to
> manufacture. Maybe his version of this engine would suit our needs, if his
> bearing will not work because of the lower rpm in a steam engine versus his
> 6,500-15,000 rpm then I guess it is back to packing. Either way I thought
> it worth seeing if it could be made feasible. I have a set of Bourke's
> plans to use as a guide.
>
> Best regards,
>
> Jason
>
> > > - Show quoted text -- Hide quoted text -
Mark Norton
I have some more to say on this issue, but I'm going to move it to a new thread, since it touches on how to go forward from here.
- Mark
Dorkmo
Ken
part size with the change in temperature and the coefficient of
expansion. Make sure the part size and temperature are in compatible
units with the coefficient of expansion.
Mike Stone
Check my math guys:
R : T.hi : P.hi : MEP
==================================
1 : 170.3 : 5.0 : 5.0
2 : 326.8 : 12.5 : 10.6
3 : 435.7 : 21.3 : 14.9
4 : 522.1 : 31.2 : 18.6
5 : 594.7 : 41.8 : 21.8
6 : 658.1 : 53.2 : 24.8
7 : 714.6 : 65.2 : 27.5
8 : 765.9 : 77.8 : 30.0
9 : 813.0 : 90.9 : 32.3
10 : 856.6 : 104.5 : 34.5
11 : 897.4 : 118.5 : 36.6
12 : 935.7 : 132.9 : 38.6
13 : 971.9 : 147.7 : 40.5
14 : 1006.2 : 162.9 : 42.3
15 : 1038.9 : 178.4 : 44.1
16 : 1070.2 : 194.3 : 45.8
17 : 1100.2 : 210.5 : 47.5
18 : 1129.0 : 226.9 : 49.1
19 : 1156.7 : 243.7 : 50.6
20 : 1183.5 : 260.8 : 52.1
R - expansion ratio
T.hi - supply steam temperature
P.hi - supply steam pressure
MEP - mean effective pressure
The table shows the supply steam conditions necessary to get 5psi absolute at the exhaust, for various compression ratios, assuming adiabatic expansion. It also shows the mean effective pressure for the stroke.
For those of you who are new to the calculations, 'adiabatic' means 'no exchange of heat with the surroundings'. The steam doesn't lose heat to the walls of the cylinder, and doesn't take heat from them. It doesn't exist in practice, but is a handy abstraction when you're doing the theoretical stuff. That's also an older term (hey, my reference books are from the 1800s). The modern term is 'isentropic', which is ever so much clearer. ;-)
The ideal gas law defines relationships between the pressure, volume, and temperature of a gas. The classical form is:
PV = nRT
where 'n' and 'R' are constants. You can compare two conditions by dropping the constants and saying:
P1V1/T1 = P2V2/T2
In practice, PV tells us how much gas we have, and T tells us the amount of energy stored in each unit of gas. If we have 1 cubic inch of gas at 100psi and 100* F, we can expand it to 2 cubic inches at 50 psi without changing the temperature, so the internal energy stays the same. If we put that gas into two equal-sized containers then separate them, we change the value of PV/T by changing the amount of gas we're talking about. The internal energy per unit of gas stays the same though.
We can also change PV/T by heating the gas or cooling it. When we do that, the amount of gas stays the same, but the amount of energy per unit of gas changes. When we do that, either P or V will change, but they do so in different ways.
If you heat a gas but keep the volume constant, the pressure will rise. If you heat a gas but keep the pressure constant, the volume will increase. Thing is, changing the volume causes the sides of the container to move. That means we have a force (the gas pressure) acting across a distance, and that's the definition of work. It takes more energy to heat a unit of gas 1 degree under constant pressure than under constant volume because under constant pressure, you have to heat the gas *and* do external work.
The amount of energy needed to do either one is called the 'specific heat' of the gas. There are two kinds, both of which have symbols. Kv is the energy necessary to heat a gas by a certain amount under constant volume, Kp is the energy necessary to heat a gas by the same amount under constant pressure. The ratio Kp/Kv is important to further calculations, so it gets a symbol of its own. It's usually called 'gamma', but for ease of typing I'll call it 'g'.
When you expand and compress gases without caring whether you change the internal energy, the ideal gas law no longer holds. Instead, the relationship between pressure and volume looks like so:
P1(V1^n) = P2(V2^n)
where 'n' has different values depending on how much energy you add or remove in the process. I'll spare you a whole truckload of math, and just tell you than when the change in internal energy (represented by the change in T) exactly equals the amount of work done by expanding the gas from one volume to another, n = g.
That condition -- energy lost as temperature = energy done as expansion -- is called 'adiabatic (isentropic) expansion'. If you run it the other way, you get 'adiabatic compression'.
So.. under adiabatic conditions:
P1(V1^g) = P2(V2^g)
and with some additional hand-waving:
T1/T2 = (V2/V1)^(g-1)
V2/V1 is the expansion ratio, which warrants its own symbol, 'R', making the equations:
P1 = P2(R^g)
T1 = T2(R^(g-1))
The expansion ratio also shows up in mean effective pressure calculations. Again, I'll spare you the calculus and just give you the result that actually matters:
MEP = P1 * (1 + ln(R)) / R
Those are the equations I used in the table above.
I set the equations up to figure the initial pressure/temperature from a known final pressure/temperature because the curve for adiabatic expansion grows faster than the curve for steam saturation. Translating that to human terms, if you start with saturated steam and expand it adiabatically, it will condense. If you want saturated steam coming out of your cylinder, your supply steam has to be superheated enough to fill the gap.
I've chosen an exhaust pressure of 5 psi absolute (10 psi below atmosphere) as a target that shouldn't be too hard to hit. The saturation temperature for steam at 5 psia is 165* F, which I fudged up to about 170* F for a couple of reasons. First, it gives me a small margin for error, and second, 170.33* F = 350* K. You have to calculate temperatures from absolute zero, not from the zeros of the farenheit or centigrade scales, and 350 is a friendlier number than 347.04.
I think the math is correct, and the first line shows a 1:1 expansion (no expansion) for the sake of calibration. Still, catching mistakes early in the design process only costs ego damage, catching them in fabrication or deployment costs money and/or minions. If anyone sees problems or has questions, fire away.