Expansion in Reciprocating Steam Engines: Hyberbolic/Isothermal, Adiabatic, Mix? & Does It Matte

[crshears]

Hello again all,

As per recommendation I'm re-posting this from Turbine & Engine Design.

I volunteer aboard a retired steamship; the following are pertinent threads I've started on the subject...

http://www.eng-tips.com/viewthread.cfm?qid=341551

http://www.eng-tips.com/viewthread.cfm?qid=342485

http://www.eng-tips.com/viewthread.cfm?qid=343198

http://www.eng-tips.com/viewthread.cfm?qid=378864

I'm finally at liberty to identify the vessel in question; her website is @

http://www.sskeewatin.com/

.

If the type of rehabilitation and re-purposing described in the above threads were ever to take place, it would be very desirable to make the facility as energy-efficient as possible within the limits imposed by not compromising the vessel's historical integrity. In consequence I'm undertaking some of the preliminary footwork so have to the right kind of info as would be needed for FEED, considering this would be a very brownfield development.

Moving on:

Numerous classic texts on the subject, and even the 1970s materials I used during my college and in-career studies, state more or less categorically that the expansion of steam within the cylinders of a steam engine is hyperbolic and/or isothermal. I find it of interest that no qualifications, codicils or caveats are appended to this assertion...

Other less-numerous sources state that in the case where steam engines harness superheated steam [the necessary lubrication, cylinder wall, piston and piston ring material considerations having been addressed], minimal heat transfer will occur between the cylinder walls and the working fluid and the expansion of the steam will in this specific type of instance be closer to the adiabatic than the isothermal.

My instincts are to believe the latter sources...but I've been wrong before, hence the wish to poll greater minds than my own.

Depending which path you take, you can end up at very different places in the steam tables, which leads me to believe I'm reasonably warranted in saying that the answer to this query would have profound effects upon the heat balance and hence the design of the plant, whether the sizes of any of the cylinders would have to be adjusted due to the change from saturated to superheated, and perhaps even reheated, steam, to what extent this would be altered by the addition of an LP exhaust turbine in pursuit of increasing overall plant efficiency, and so on.

Any thoughts?

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[EdStainless]

I have seen that ship.
Do you know the original design, temp, flow, quality of steam?

I trust that you have talked with the guys at IST about an HRSG, they are good at 'odd' applications.

But to your question, I would consult "Heat-Power Engineering, Part II, Barnard, Ellenwood, and Hirshfeld". There are actually new printings of this early 1920's book available cheap.
And this

https://archive.org/details/unaflowsteamengi00stumrich

And a couple of attachments.
As I recall most large recips started with just a little SH, then reheated to almost the same conditions. So you ended up with two very similar HP cylinders and then two larger LP cylinders. Under study state conditions I believe that you are correct, adiabatic comes closer to describing the conditions, with corrections for the real world.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[crshears]

Thanks for the response, Ed.

There were four coal-fired, hand-stoked Scotch boilers producing saturated steam [I can find no evidence or mention of superheaters] @ a maximum pressure of #220 psig, 396°F . . . but I can't see where cyclonic or other separators were ever installed, so the dryness fraction I can't speak to. Pounds of steam per hour I have not yet been able to ascertain.

Will search the Web for IST's homepage...

Found the book you mention as a downloadable pdf at

https://archive.org/details/elementsofheatpo02hirs

- looks an excellent resource to use alongside the books by Seaton, Ripper, Sennett and Sothern I've already found the same way. Some more light weekend reading ahead of me...

The Kee's engine is a quadruple expansion double-acting 3300 IHP engine, therefore HP > IP1 > IP2 > LP. I'm working on the number-crunching for applying both superheat and reheat, as well as adding a low-pressure exhaust turbine to harness the delta P between the LP cylinder's terminal/release pressure and the condenser.

Details, details...

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[georgeverghese]

Would imagine the operation of a steam engine where mechanical shaft work is extracted, that the expansion would approach adiabatic full isentropic. Similar to that for a radial entry gas expansion turbine, except that isentropic eff is much higher for the recip engine.

Believe recip engines like these were used in the old days ( before expansion turbines came into vogue) for creating low temps in air separations plants for producing liquid O2 and N2. These fell out of fashion due to reliability being lower than expanders.

 

[crshears]

Hmp...

Had a look at "Heat-Power Engineering, Part I, Barnard, Ellenwood, and Hirshfeld;" pgs 378-379, article 229 states that "[fig 204 shows how well] the expansion and compression curves follow the polytropic law PV superscript n = K."

Now I'm really confused...

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[EdStainless]

The rounded corners indicate deviation from ideal, and the slope of the lines as well being related to steam quality. Notice that this is P vs V (not what we are accustomed to) and that it is all from real engines.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[georgeverghese]

For full isentropic adiabatic, n = k ( poly eff = isen eff = k )
As polytropic( for compressors) (or isentropic for expanders )efficiency drops, n < k.

As gas pressure increases, and compressibility factor Z < 1, then k deviates from ideal gas values.

 

[georgeverghese]

Pardon me, Perry 7th edn. in page 11-101 says isentropic eff for recip expanders ranges from 75% for small machines to 85% for the larger machines.

So isen eff is similar to that for turbine expanders.

 

[crshears]

I trust that you have talked with the guys at IST about an HRSG, they are good at 'odd' applications.

I've asked for access to the White Paper section of IST's website...wish me luck.

Any other resources you'd suggest looking at to gain some familiarity with flash boiler / OTSG technology? [I do intend to Google for it, but there's often so much chaff mixed in with the nuggets...]

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[crshears]

FWIW, fourth working day in and IST hasn't either granted me White Paper access or responded to a separate e-mail I sent on the same topic; I guess I either don't rate, or am not perceived as providing enough likelihood of future business...

Any further recommendations as per previous post?

Thanks.



CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[crshears]

Got the IST approval; will check their website on nights next week.

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[EdStainless]

IST builds HRSGs that are fairly unique. They are designed to be drained when you don't want the steam, so do diversion gates for the exhaust and no dumping of steam when you don't need it.
This requires building out of more expensive materials, but it greatly simplifies the system and controls.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[racookpe1978]

Let's assume the steam engine is up at operating temperature, and the cylinder itself is fairly well insulated.

Expansion inside each piston is then isothermal, with the volume expanding as the piston is forced down. But the final volume is absolutely limited (by piston volume: low power = same piston volume at high power) so the "outlet" steam will have more or less energy (enthalpy) and condition (water percent) based on ho much energy is NOT used at that stroke.

So, if you needed low power, then the steam output/lb steam flow would be at higher energy than at higher power, but the volume the same, and the temperature nearly the same as inlet temperature, so efficiency is less. If saturated steam inlet, then water-laden steam at output.

 

[crshears]

Thanks, all, for the responses; very useful stuff. The number crunching continues...

Had a look at IST's website and White Papers: very informative, and at first blush their once-through steam generator looks like just the ticket for the application. Of particular interest is the relative rapidity of starting and shutting down, as well as the very quick transient response time. Also of note is that IST is less than 200 km away by rail from the shipyards where the work would likely be undertaken...

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[crshears]

I trust that you have talked with the guys at IST about an HRSG, they are good at 'odd' applications.

Thanks again, Ed; I have established a personal contact with someone at IST. We'll see where it goes...

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[crshears]

Long-overdue follow-up...

IST was kind enough to send me the specs on a once-through steam generator they made the design for but that the customer decided not to order...preliminary calculations suggest that two CTG's @ ~2 MW exhausting to one HRSG each would give about the right volume of steam @ an outlet pressure of ~440 psia & @ 1,000°F that, when expanded through a topping turbine to an exhaust pressure of ~240 psia, would enable the quadruple expansion reciprocating engine to operate near its design power. Capacity/power of topping turbine to be determined, although its product would without question be electricity for sale. Thought is to use two separate admission nozzles so if running on only one CTG / HRSG combination the topping turbine would not operate at reduced efficiency due to throttling loss.

By judicious cyclonic separation [to spin out cylinder oil] could also add a bottoming LP exhaust turbine between the quad's LP cylinder outlet and condenser; prelimary numbers suggest about 500 kW of additional generation could be derived by this means. "Twin" nozzles here, too.

Depending on temperature of topping turbine exhaust steam, either part of it or live steam could be fed to a contact reheater partway through the quad steam cycle, both to reduce/eliminate cyinder condensation and ensure moisture content to/within the LP exhaust turbine does not rise so high as to cause blade erosion there. This would also avoid altering an already-set HRSG design, something we can't afford - plus a sizeable discount on the HRSGs would probably be offered provided we didn't ask for any alterations.

In parallel with topping turbine would be a steam accumulator to not only absorb excess steam production but to store enough of it to keep the auxiliaries hot overnight.

Additional supply of 440 psia steam would be provided via electric boiler overnight when bulk electricity prices are often very low.

As was mentioned previously by someone else, going this route would dispense with bypass ducting entirely, although intent would be to still have HRSG gas outlet 'blast gates' to conserve heat overnight and shorten HRSG start-ups the following day.

Next job is to find [a] free or cheap access to PEPSE, or the appropriate Thermoflow program, or some other software that will enable much more accurate number crunching, and someone to assist me along the way. Wish me luck!

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[crshears]

Hello again, all!

I recently received some information [via scanned PDF of original 1907 sea trials, no less!] of the actual performance of the Kee's engine at the time of its acceptance trials, to wit:

Heat Balance/Engine Specifications

As designed:

1. Quadruple Expansion Steam Engine; steam admission @ 220 PSIG saturated

2. Cylinder [piston?] diameters: HP - 23.5”; MP1 – 34”; MP2 – 48.5”; LP – 70”

3. Stroke of all cylinders: 45”

4. Horsepower: 3300

5. Rotary speed: 120 rpm

As determined by sea trial:

1. Power developed: 4 hour trial – 3194 IHP @ 96.185 rpm

2. Maximum Power developed - 2 mile run - 3457 IHP @ 97.82 rpm

The notable difference in rotational speed between design and actual is likely due to a coarser pitch of propeller having been fitted, although this is uncertain.

There is quite a bit more about furnace grate size, heating surface area, et cetera et cetera, but still no information about its water rate, which is frustrating; click on the link to see the scan.




CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[gruntguru]

At 120 rpm, the power could be as much as 20% higher.

Note the power figures are "indicated" ie work done on the piston crown. "Brake" power at the prop shaft after friction losses will be somewhat less so perhaps the 3300 rating is BHP. That would make sense if IHP is 3457 @ 100 rpm then BHP might be about 3457 + 20% - friction = 3300.

Anyone out there know what is a typical value for FMEP in these old steam engines?

je suis charlie

 

[crshears]

Hi gruntguru, thanks for the input...although I'm not clear on what FMEP is; I've seen lots of abbreviations, but this one is a first for me.

As for the original question...

I've come to the conclusion that whichever text it was that described the expansion of steam in a steam engine cylinder as isothermal was flat-out wrong; Skrotzki's "Basic Thermodynamics" says much more accurately that [not an exact quote] "although the expansion characteristic within steam engine cylinders often lies close to PV = C, the expansion is not isothermal..." and it makes good intuitive sense to me that for engines using saturated steam this statement is closest to actual experience.

Sothern's statement [again, not a direct quote] that "engines of the first class in good condition, using superheated steam and operated correctly, will produce an expansion line that approximates quite closely to the adiabatic" is also logical, with the major cause of departure being the irrecoverable loss due to transfer of heat into the cylinder walls at the beginning of of the stroke and transfer back into the steam at the end.

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[gruntguru]

FMEP = Friction Mean Effective Pressure = Loss of MEP due to mechanical inefficiencies.

BMEP = IMEP - FMEP

je suis charlie