Condenser Cooling water flow

Agreed!

Chris

"Tony has had some good input, that's for sure"!! Amen to that!

Thanks Guys - I'm glad I asked the question because the more I thought about it the more unsure I became. However while you have been considering the question I have finally found the source of all wisdom concerning marine engines and steam plant which I thought had been lost in a house move.

The book has no title (!) but the foreword says it is a reprint (bound photocopy) produced by The Nabu Public Domain Reprints - made in the USA from a series of textbooks for marine engineers originally printed (as 11 separate volumes) printed between 1902 and 1907.

The answer for all airpumped tubed condensers is that water enters at the bottom or cool side/section and exits at the top or hot section - the book covers several designs but seems to be the general rule.

With my model I am more concerned in raising a vacuum than getting the best thermal efficiency so the condensate return will go into the top of the feed water tank (cylinder lubrication should float on top?). The engine has two feed water pumps which will draw from the bottom of the tank and whose discharge will tee into the boiler feed/hand-pump line. There is a manually operated feed by-pass that just returns to the tank.

The condenser will have a separate water tank the water from which I intend to circulate with a small submerged 12v aquarium pump.

I don't know how all this is going to perform but after spending lots of money and time in building the engine I thought it a bit of a crime to have it sitting as some sort of display item (my work is not exhibition quality although the casual visitor would be impressed) so I have built a stainless steel plated base in two parts - the engine section with all the tubes terminating at fixed manifolds and the boiler section housing the two water tanks, the boiler, its tubing and feed pump. Now if I can't get the system to work in a manner that is suitable and safe to 'show off', then I'll disconnect the two sections, mount the engine on a hardwood block and sulk.

Tony

I'm sure it will do as intended, but as a general point it can be worth keeping eyes open for old engineering text-books as guides to typical design practices contemporary with the machines we are replicating.

Many show not only the machine's principles like steam-engine valve-gears; but also common details like proportions of parts, based on various manufacturers' drawings.

Sounds like a great old book.

Got pics of your engine and set up so far?

Topical as just this very minute I found and ordered this

**LINK**

It has to be worth £3.90 to find out if it's what it says it is

Tug

Counterintuitively the vacuum is governed by the cooling water temperature and flow. Air pump only pumps out any entrained air coming out of the steam and condensate. It does not create the vacuum.

Condensate ends up much cooler than saturation temp because the cooling water temp must be considerably lower to promote a rapid rate of heat transfer. Steam condenses on the upper tubes and the condensate trickles down over thr lower tubes and is further cooled in the process. So a standard condenser ends up with considerably cooled condensate. But heat can be reclaimed from the cooling water by using it to heat raw feedwater or incoming furnace air etc.

Regenerative condenser reheats the cool condensate with steam and saves all that extra plumbing. But it still relies on large temperature differential in the first phase.

I am reverse-engineering the Mississippi steam ship side beam engine, and noticed that it has a condenser and air pump.

I have read that at such a low steam pressure (I think 18 psi), the condenser is required in order for the engine to operate and develop usable power.

The steam cylinder is 75" diameter, and the air pump is 46.5" diameter, so a 2.6:1 ratio as far as surface area.

It would seem that the air pump would use a significant amount of power to operate, but perhaps not.

I am not sure exactly what pressure the air pump piston has to work against, but that would determine the power used by the air pump.

I noticed there is a foot valve, so the air pump only has to move air and condensate past that foot valve? and then gets pumped back to the boiler, or pumped overboard by another forced pump?

I am new to condensers, especially of this old style where there was a jet of water forced into the condenser chamber, and no separation via the tubes commonly seen in more modern condensers, so trying to understand more about them.

.

Edited By PatJ on 24/02/2021 23:45:35

That sounds like an exceptional case if its running on 18psi. It would need all the vacuum help it could get.

Actually, scanning my old books some say vacuum is created and controlled by the cooling water and condensation. Others say rsther ambiguously that the air pump of the type you mention removes the condensate and air in order to maintain the vacuum. Maintain but not create? I dont know. There are many different designs. Most have separate condensate and air pumps. Modern power stations use a steam venturi ejector to remove the air.

That pump on your ship would typically be pumping only against gravity to pump the condensate into a storage tank or deaerator etc. A separate boiler feed pump would force feedwater into the boiler under pressure.

Although if its a spray type condenser as you mention, the condensate would be polluted with cooling water so they might have pumped the lot straight overboard and used fresh feedwater. But that seems pretty wasteful. Probably why that design fell from favour.

Edited By Hopper on 25/02/2021 01:05:56

Edited By Hopper on 25/02/2021 01:12:00

Edited By Hopper on 25/02/2021 01:16:02

Here is the information on the Mississippi steam ship.

This data mentions 12 psi, which is even lower than I stated above.

Constructed around 1840, using a copper boiler I think.

Boilers in 1840 were apparently rather crude from a standpoint of withstanding very much pressure.

Hopper-

Everything you mention sounds reasonable.

I guess when you consider that science class experiment where they would introduce steam into a flexible container, and then submerse the sealed container into cold water, and it would collapse.

So the vacuum force would be generated without an air pump, and my guess is like you mention, the air pump just removes excess air and condensate.

Here are a few items I found in some old books.

The Mississippi condensers were jet condenser type.

I hope I am not cluttering up or hijacking this threat too much with all this.

Edited By PatJ on 25/02/2021 02:05:49

Edited By PatJ on 25/02/2021 02:07:46

Wow thats some primitive old stuff. Should be fun to model. Kind of the other end of the scale from the highly developed triple expansion engines of the 1920s etc.

Edited By Hopper on 25/02/2021 03:00:26

Definitely on the other end of the spectrum from where steam engines were in 1920.

Predates the screw propeller (The Mississippi used paddlewheels).

I am told that this was the first attempt to mechanize a US ship with a steam engine, and it did appear to be a functional ship, which sailed/steamed all over the world.

It was the first and only side beam (lever) steam ship the US Navy ever built, and so that makes it rather unique too.  It seems like they took a sailing ship design, carved out the center, and dropped in two very large engines and a bunch of boilers.  Sort a shoe-horn affair it would seem, but apparently a successful attempt.

I was quite surprised when I read that they used sea water in the boiler.

The frame is Gothic style, and so is rather a pain in the rear to model in 3D, but I am getting there.

But I saw the post title, and so decided to try and learn more about condensers.

.

Edited By PatJ on 25/02/2021 05:06:00

Edited By PatJ on 25/02/2021 05:06:44

Edited By PatJ on 25/02/2021 05:07:34

Tony: you have a PM from me.

Having had a good think, Hopper is right and I was wrong. Not unusual. As soon as any steam is condensed, the tubes must have a film of water, so the heat transfer is steam to condensate to tubes to coolant. Thus the condensate needs to be cooler than the steam. How much cooler depends on the surface area and steam flow.

As well as being cooled by the tubes, the condensate is being heated by the steam, but as you go further down the tube bank there is less steam, so the lower tubes will be cooler than the top. I wonder if anyone ever tried having the steam flowing East-West across tubes running North-South with trays between the rows of tubes to catch the water and stop it pouring over the lower banks.

I still don't think that having the condensate temperature lower than the exhaust saturation temperature increases efficiency.

I've never quite got my head round this air pump doesn't create the vacuum argument. This seems to be conventional wisdom, but no air pump means no vacuum, and the characteristics of the air pump (clearance volume and swept volume) must dictate how low the pressure of air in the condenser can be maintained. When I worked on steam turbines many moons ago they had steam feeds to the shaft seals so that any in-leakge would be steam rather than air. The valves would be adjusted until a wisp of steam came out of the gland.

How all this works with a jet condenser wants more thinking about.

Which would be why Weirs developed the regenerative condenser that has an area full of "fresh" steam below the bottom tube bank so that droplets of condensate fall off the tubes and pass through the steam, raising them to within a claimed one degree of the turbine exhaust/condenser inlet steam temperature and extracting heat from the steam at the same time. It combines the advantages of rapid cooling due to low cooling water temp but sends the condensate back to the boiler feed supply at optimum temperature to minimise reheating. DETAILS HERE  Of course, the cooling depends on the flow rate of the cooling water as well as the temp so can be regulated that way.

Googling around at steam condenser diagrams, it looks like there are cross flow condensers as you suggest, with baffles to direct the steam through in a zigzag path and baffles for condensate flow. It looks like just about every possibility and combination has been tried over the years.

Also stumbled on some studies on turbine condensers in nuke power stations that use ambient river water or sea water to cool them and then discharge it back into the river/sea. They are concerned that with global warming, if the cooling water source temp rises by one degree C, efficiency of the generator unit will be reduced by almost half a percent. So it seems in those installations, cooling water temp is important, and coming out of a river or sea it would be relatively low.

All of which has nothing much to do with model engines but interesting in principle. It would be more interesting to know if model triple expansion engines with a condenser but no air pump do in fact pull a (partial) vacuum in the condenser shell. I wonder if anyone ever bothered to measure it? Maybe a job for you, Tony, when you get your rig up and running?

Edited By Hopper on 26/02/2021 00:40:38

On this question of whether or not the air/condensate pump creates the vacuum or not..

My experience is that you won't see any vacuum until the pump is working. At the beginning of the process, the condenser will be full of air, which is rather hard to condense, so until that has been pumped out you won't see any vacuum. Actually of course steam from the warm up might well have displaced a lot of the air, but if the cooling tubes are full of cold water that steam will condense and the condenser is liable to be flooded with condensate.

However, the point that someone no doubt was trying to make way back in history, is that the whole process would not work if steam/water did not have the lovely property that the volume of the water is about 1600 times less than the volume of the corresponding steam. So to maintain the vacuum it is only necessary to pump out a relatively small volume of water, plus of course any unwanted air that has made its way in.

The same thing is also an advantage on the feed pump side. Compare the relative size of the feedpumps to the main engine, then compare the size of the compressor on a jet engine to the turbine. Sadly for steam we also need a relatively large boiler to heat the working fluid, much bigger than the combustion cans on a gas turbine. Perhaps a better comparison is the much larger size a typical hot air engine needs to generate a similar power to a steam engine

Practically, on my own plant on "Dancer", it is necessary to open the vacuum breaker for a bit on starting out. This allows starting out without pressurising the condenser and allows the circulating pump(driven by the main engine) to start the flow of cooling water. After a short time, the cloud of steam over the side from the vacuum breaker diminishes greatly and condensate starts to be returned to the hot well. At that point, the vacuum breaker can be closed and the air/water pump starts pulling down the vacuum. When all is well it takes maybe 10 to 20 seconds for the vacuum gauge to climb around the scale.

Running on salt water was relatively common in the early days of marine steam, even into the 1850's.. It is very wasteful since the boiler tends to get loaded up with solids so has to be blown down frequently or even continuously, and the salt content raises the temperature required to boil the water. A boiler full of salt water is also very prone to priming.

All this is of course pretty irrelevant to a model, where efficiency is not a great concern, but it is kind of satisfying to do things properly even if it doesn't matter. Condensing the exhaust, even if you don't care about the vacuum, will reduce the cloud of steam that would otherwise result, which might make it nicer to run on steam for a public display.

My own favourite reference would be "The Efficient Use of Steam" by Lyle Watson.

John

John, good to hear it from someone who has done it at model size. That makes sense to use the one pump to remove condensate and air on a small job. Air being heavier than steam at these temperatures will sink to the bottom of the condenser. So it would seem that as long as the condensate pump is designed to suck condensate faster than the engine supplies exhaust steam, air will be removed along with it to make up the shortfall.

Like Tony I worked on full sized turbines in power stations many years ago but never took close notice of the condensers. All I vaguely remember is each condenser was about the size of a house and the exhaust steam trunking from turbine to condenser was big enough to drive a car through but the air pump was about the size of a Myford lathe with a small one or two horsepower motor, maybe a five at most. So it always seemed to me that the 1600 to 1 reduction in volume between steam and condensate would create the vacuum. Couldn't see that tiny air pump keeping up with the total steam output of a 120MW turbine at 3,000rpm. But as you say, if the condenser is full of air, no vacuum can exist there so the pump plays its vital part in maintaining the vacuum.

Looks like archive.org has a free copy of "The Efficient Use of Steam" so I shall have to have a squiz.

Edited By Hopper on 26/02/2021 06:57:14

I have a problem in locating a small positive pressure/negative pressure dial gauge. I may try to rig up a U tube to do the job. I have been troubled by condenser problems during my working life while working for WH Allen. Finding a steam leak is very easy but finding a vacuum leak is a nightmare. I am about 3 weeks away from a first trial run and will report on my findings.

Tony

No idea where you would get a vacuum gauge that small. Maybe John Olsen above can help?

Looking forward to seeing the video of it running, vacuum or no vacuum!

No idea if it's practical for your purpose but I remember from my Tesla Coil days that they are used to detect vacuum leaks.

Basically a probe with high voltage at low Radio Frequency power on it is brought close to the test object until sparks jump. As the sparks concentrate on areas on low pressure a strengthened spark indicates where the vacuum leak is. Description of such a device here.

Dave