Welcome! Log In Create A New Profile Recent Messages


Model boat Flash Steam Power Plant

Posted by lohring 
Model boat Flash Steam Power Plant
April 05, 2014 07:14PM
I'm considering building a steam powered radio controlled model boat with a monotube boiler. I've built many internal combustion engined boats and have held the 109+ mph gasoline engine powered speed record with my partner since 2004. Unlike the flash steam tetherboats, this new boat will need a control system. These engines probably have steam at about 1000 psi and run around 10,000 rpm. The boiler tubes may get to 1500 degrees F. They are non condensing uniflow engines with 54 degrees of steam admission and 102 degrees exhaust duration. I plan to throttle the engine with a variable valve lift and timing to reduce the cutoff. The boiler is around 30 feet of 1/4 OD .028 wall 316 stainless steel tubing. I've been studying the forum for boiler control ideas.

I think the Waterman pump system would work. The tetherboats run their fuel and water pumps off the crankshaft. If you build a duplex steam pump with one pump cylinder for water and one for fuel, you should have a water and fuel feed that's proportional to steam demand as set by the valve throttling. The layout would look like the picture from the Waterman patent. I'm considering the same size pump cylinders as the mechanical pumps that run at around 1500 to 2000 rpm with double acting steam cylinders having about 30 times the area of the single acting pump cylinders. That would be 1 1/4 inch diameter steam cylinders with 1/4 and 3/16 inch diameter pump cylinders.

The boilers and burners are very simple. Efficiency isn't a serious consideration, but power and control is. The boat needs to slow to an idling speed around 40 mph then accelerate to over 100 mph in 2 or 3 seconds. I hope to get around 8 net shaft horsepower from a 23 cc (1.4 cubic inch) displacement engine.

I'm very interested in your thoughts as well as a better idea of the pump sizing.

Lohring Miller

Re: Model boat Flash Steam Power Plant
April 05, 2014 08:10PM
Nice project.
Most of these type of boats built in the 30, 40 and 50 used piston driven water pumps right off the engine. Your generator looks like a copy right from the book Experimental Flash steam. By J.H Benson and A.A. Rayman. 1973 Great book.
Every thing needed to size the pump to the engine in that book. I trust you must have seen it or have a copy.
Re: Model boat Flash Steam Power Plant
April 06, 2014 10:56AM
I have that book as well as articles by Bob Kirk on his tetherboat and George Thompson on his radio controlled, flash steam mono. Paul Windross, the current steam tetherboat record holder, also sent me details of his engine. They all used fixed displacement, engine driven pumps. George's boat only had on - off engine control, though. The engines are pretty standard at this point. I only need to add a simple valve travel adjustment somewhat like what the Cyclone engine uses.

I want a simple, mechanical boiler control system. The dual duplex pump idea looked very promising. The pump pictured needs seriously upgraded components like the modern Gardner Denver pumps with piston valves in a cast iron and stainless steel steam end. I like the rocker valve actuation as well. My questions are about the cylinder size ratios and general operation of the system.

Does anyone have any more thoughts than were expressed in the Waterman thread? I was very interested in Andy's calculations. If the water and fuel are fed from the same pump, is any other temperature or pressure control needed? Will the system be fast acting?

Lohring Miller

Re: Model boat Flash Steam Power Plant
April 06, 2014 02:11PM
The waterman control system works well with a drum or marine type boiler not a Monotube boiler. It floods the valve box with water when the water is high in the boiler causing the pump to run very slow, then when the water is low in the boiler steam flows to the pump making it hot and work again, it needs to be self starting.
From my recollection on the Monotube race boats they seized the volume of the pump to the steam consumption of the engine. You need a small light weight piston pump connected to your engine. I did this with my tug boat. No controls it would maintain water level well over two hours. The two photos you just posted I also used on my duplex pump. See attached.

Re: Model boat Flash Steam Power Plant
April 07, 2014 03:11AM
Hi Lohring

My proportional pump idea, as far as I know, is an untried concept. I figure that things will change over a period of time and also with temperature. Worm up for example. Leakage for another. I figure that some feed back might be needed. Also a design so the water volume / steam volume ratio can be adjusted initially.

The basic idea is that it takes x BTU / lb of water to achieve the steam state desired. You have to put x BTUs into the fluid to raise it's temperature to the desired temperature at the desired pressure. Difference between the enthalpy of the steam and feed water adjusted by boiler efficiency. The boiler efficiency can change. Generally efficiency decreases with firing rate.

This would only work for a fixed steam volume to water volume setup having fixed thermodynamic states. Running the pump of a throttled engine on with variable cutoff is not going to work. The idea is a donkey pump with vastly different displacements for steam vs water cylinders. In the example there was a 42 to 1 volume ratio steam to water, The steam cylinder displacement would be 42 times that of the water cylinder. The donkey pump steam cylinder would be double acting and have pistons rods on both sides of the piston protruding out of the steam cylinder. Single acting water cylinders on those rods. When the exhausting side of the steam cylinder is a bit less than 41/42 the steam pressure of the inlet side the pump pushes water. I am figuring fuel would also be pumped in proportion to the water as well but having a temperature feed back adjustment. Actually the fuel cylinders are regulating the fuel in proportion to the water. I was figuring on an electric fuel pump feeding them. So they could be bypassed during start up.

The steam to water cylinder displacement ratio is the specific volume ratio of the steam and feed water.

Re: Model boat Flash Steam Power Plant
April 07, 2014 10:35AM
You built a very nice duplex pump. I'm putting together the engine currently. I spent a lot of time thinking about variable stroke pumps before I stumbled on the Waterman concept. I still don't see why the Waterman system won't work exactly as you described with a monotube boiler. The steam cylinder of the pump is connected to the wet section of the boiler with the superheater coils connected to the exhaust. As the steam gets wetter, the pump it slows down. As the steam in the coils gets hotter, the pump speeds up. The valves need to be full admission and the pump can't lock up when run on water.

If the fuel to water ratio is approximately constant, the fuel supply will match the steam demand. There will be very different loads on the water side pumping against 1000 psi and the fuel side pumping against 50 psi. Will this matter? I suppost the only way to find out is to build a pump and try it. I can always run a geared pump from the engine. Controlling it is the problem. Below is one analysis from the thread on this forum.

Lohring Miller

"One could run the output of the evaporator section directly to the pump (no stand pipe). You would just have a separator water trap on the pump exhaust and the excess water would be routed back into the boiler feed.
At 1000 PSIA I figured the ratios of the volume of the steam from saturated mixture quality of 75% to superheated steam at 632.96F to feed water at 1000 PSIA and 180F

The ratio at the low end was 20.652::1 at 75% quality. At the other end was the super heated steam at 33.263::1. The saturated vapor (quality 100%) was 27.098::1.

It does look like a system of running steam from the evaporator into the donkey pump would be self correcting. However it would self correct to match the effective real world ratio.
After vaporizing steam at some boiler pressure, the vapor occupies n * the volume the feed water at boiler pressure does. I am using feed water at boiler pressure because in the water cylinder the feed water is being pumped against boiler pressure. Actually it needs to be a bit greater then boiler pressure. There is some drop in pressure due to flow resistance. I have seen measured pressures (in test data) across boilers of from 80 PSIA to 150 PSIA. These were boilers operating around 1000 PSIA.

Anyway by taking the ratio of the specific volumes you can get idea of the volume of water that needs to be pumped to replace the volume of water being used.

In my system: That ratio also is the pressure ratio needed to pump the water.

In my numbers above if you were trying to just get saturated vapor out of the vaporizer. Than the volume of saturated vapor would be just a hair over 27 times the volume of the water entering the boiler.

So let’s say we need to put 1100 PSIA pressure on the feed water to overcome the flow resistance. Then the steam cylinders only need 1100/27 PSIA to generate 1100 PSIA on the feed water or about 40.75 PSI. We would be putting steam in to the donkey pump steam cylinder at 1000 PSIA. The exhaust of the donkey pump would be at 959.26 PSIA and go to the super heater and out to the engine. This is a flow meter of sorts. For every cubic foot of steam going through the donkey pump we pump 1/27 cubic foot of water into the boiler. That is in the ideal analysis. There would have to be some adjustment to the ratio to make it work in the real world.

With reference to the self correcting; if we are getting superheated steam then its specific volume is greater than the saturated vapor and the pump would pump extra water and that would reduce the temperature. If it goes in to the wet steam region the steam volume is less then at 100% quality so less water would be pumped and the steam temperature would increase with reduced water rate. On the other hand, let’s look at the pressure change effect on the system. At lower pressure we would be maintaining steam around a wet point. While at higher pressure it would be in the super heated steam region. If we plot a line of constant 27:1 ratio against the saturated vapor line. The constant ratio line crosses the saturation at about 1000 PSIA. Above 1000 PSIA the constant ratio line is above the saturation line (super heated steam). Below 1000 PSIA it is below the saturated line (wet steam)."
Re: Model boat Flash Steam Power Plant
April 07, 2014 12:22PM
I wish you well with this project.
They call these experimental flash steam because that is what they are.
I have little experience with these high speed engines and flash systems.
I’m old school slow and easy and getting slower by the day.
I do know it’s best to use a pump that is gear reduced at least four to one, both water and oil pumps. I did this with both my full size boats and my big engine I built for my 35 foot boat. See attached. They work much better. It’s very easy to install a link bar to make the stroke adjustable. Plants with a control system you make the pumps 35 to 40 % over. Faster recovery. But with these systems you shoot for balance.
Most of these plants that I’ve read about run excessively hot and burn the generator out more then once before things are tweaked to last more then a few runs. According to the book some of these engines have the cylinders water cooled because they run so hot.
I don’t know of anyone working on s similar project. Hopefully you’ll find someone.
Good luck.

Re: Model boat Flash Steam Power Plant
April 07, 2014 02:26PM
Hi Lohring

You got the feed pump idea correct. If you used a lever linkage between the steam and water cylinders you would still have the same mechanical advantage and volume ratios. For example double the stroke of the steam cylinder and half the it's piston area. You have half the force with the same displacement. Connected to the water piston thou a lever system that gave it a 2 to 1 mechanical advantage would produce the same force on the water piston. it would have half the stroke of the steam piston. The volume ratio would be the same. The reason I bring this up is that the ratio could be adjusted with a lever system. It wouldn't be as compact. But the ratio could be adjusted.

The other part is to supply fuel in proportion to the water pumped with temperature feed back control.

Re: Model boat Flash Steam Power Plant
April 17, 2014 10:50PM
A basic problem in flash steam is matching the fire and feedwater, with either overheated coils, or flooded coils, when the matching porportions are not just right.

A fixed orifice with saturated steam or saturated water discharging to atmosphic pressure has characteristics usable for control. At 100 PSI, the mass flow rate of saturated water is about 10X the mass flow rate of saturated steam for any sharp edged orifice. A separator tank receives the steam/water mixture from the flash coils, and the orifice is located at the bottom of the tank. If too much water is entering the tank, it quickly discharges thru the orifice, keeping the tank from flooding. When steam is reaching the orifice, the mass flow goes down by a factor of 10, so much less loss of steam. The discharge from the orifice can be directed to a counterflow heat exchanger, with the other fluid (being heated) is cool feedwater flowing on its way to the flash steam generating coils. This arrangement recovers much of the heat in the blowdown from the separator tank. The engine receives steam from the top of the tank, which always sees just steam.

For example, say we have enough fire to generate 100 Pounds Per Hour (PPH) of steam in the flash coil. The feedwater pump is 200 PPH, so the separator tank orifice passes 100 PPH saturated water back to the feed tank, and about 100 PPH saturated steam is available for the engine.

If the fire is made twice as powerful, then 200 PPH steam is produced, 10 PPH steam bleeds back thru the orifice, and 190 PPH is available for the engine. Any firing rate between these two extremes can be used, and the separator tank will never flood, the engine always gets steam, not water, not superheat. A separate superheat coil after the separator tank can add superheat if desired, but the main generating coils will always see steam/water mixture, so no concern about very high temperatures in the flash coil.

Having this large latitude in firing rate with a passive control systen (no switches, no floats, no moving parts) can solve the flash generator control problem. The system described above allows a significant mis-match without flooding or overheating of the steam generator output. A crude porportional system, linkling firing rate and feedwater flow, can allow turndown ratios of 10:1 quite easily. I have used this system on a flash boilered steamboat, it works well.

The description above is somewhat simplified, but serves to illustrate the principles to achieve reasonable control of a flash steam generating system.

Edited 1 time(s). Last edit at 04/17/2014 10:51PM by fredrosse.
Re: Model boat Flash Steam Power Plant
April 18, 2014 11:55AM
I'm not sure I fully understand your system, but I think that's part of the basis of the Waterman steam pump system. There the orfice between the pump and the boiler output along with a relief valve is part of the output regulating system. Are you are saying that the orfice alone in a bypass circuit would adjust the output to the super heater coils depending on the state of the steam in the coils before the superheater?

These power plants run at least 1000 (maybe 2000) psi with the fuel and water rates directly proportional to engine rpm. The trick at full throttle is adjusting the fixed pump stroke for the right amount of water. I'm sure the fuel rate is more than really needed especially if the boiler was fancier. The 316 stainless steel coils seem to be able to be run dry without failure, but flooding is an issue.

The water pump is 1/4" diameter and the fuel pump is 3/16" diameter. Both have a 3/8" stroke. My calculations show around 1.3 pounds/minute of water and .5 pounds/minute of fuel at 2000 pump rpm and 10,000 engine rpm. The water rate gives around 7 to 8 hp, about right. The fuel rate is 9000 BTU/minute of fuel or over 200 hp at 100% efficiency. Maybe the rpm based pump output will be enough with the intake valve timing/throttling adjustment. The boat only really needs stop, low speed, and full throttle.

In any case, I'm going to build the engine with a throttle system and a gear drive pump first. I'll see how this works on a bench against a water brake dyno before trying other systems. Thanks for all your ideas. For models, simpler is always better.

Lohring Miller
Re: Model boat Flash Steam Power Plant
April 18, 2014 03:53PM
Fredrosse - thanks for that interesting scheme. Has it actually been used in a working plant, and how do you decide the size of the discharge orifice? Is the response time quick enough to cope with large variation in steam demand as would be found in a steam car or is it best suited to a more steady demand situation such as in a boat? Or only for racing model boat plants?

Re: Model boat Flash Steam Power Plant
May 22, 2014 10:04AM
The steam generator described was used in a 1/4 horsepower steam "scanoe" (Coleman square back canoe) and worked well at relatively constant loads. I had no automatic burner control, but I see it would work well with modulating burner and feedwater running in unison. The old systems that tried to match feedwater and fuel flows (some early steam cars had this, Serpolet?) were inherently more troublesome. The separator and water dump back thru a counterflow heat exchanger (heating feedwater) is much more forgiving and stable. If superheat is wanted then another level of complication is present.

The orifice sizing is straightforward according to Moody Critical Flow tables.
Sorry, only registered users may post in this forum.

Click here to login