small steam engine for a model boat.

[optima21]

now I have the linkage to drive the pump, I needed to make a pump, its one of those jobs that at first doesn't sound to bad until you think about what it needs to do, keep the weight to a minimum and keep it compact. It took me a while to decide if I was best machining it from a scrap gunmetal casting or to make with the main body being made from flat bar and for the bits that have the one way valves being made from round bar, and sliver solder them together. I went for the second option as it would be better wearing if the part that has the one way valves were made from phosphor bronze with the main body being made from brass.

I started off making the main body, which can be seen with the drawing that I was working to. its a working drawing for me to be able to follow and understand, and easy to alter too. the body at this stage has the two mounting holes for the legs in it, and a hole where the valves would go though it.

then its turned in the lathe to create the legs and create the round part of the cylinder and the end was threaded for holding the o ring that will seal the piston into the cylinder. this end was also centre drilled so that I could finish the bore off later in my milling machine, as I though that would be the easiest way to hold it. this was machined with a parting tool with very little overhang.

the sides of the main body were then milled down to the finished size, and the top of the cylinder were then rounded over with a file

the part that becomes the one way valves had half of it machined, this half didnt need to be as accurate but needed a small notch milling with a 1/16" milling cutter to allow water past the ball when it was lifted off the valve seat. this was done on the milling machine

and the two parts ready to be silver soldered together, the M10 nuts are just to keep the two parts in place

the other valve part being machined, being threaded to fit a union or banjo bolt (Im going for banjo bolts as it reduces the height.)

and creating the seat for the ball using a hardened steel ball (on the end of a 2.4mm tig rod) gently hit with a hammer on the tailstock. if this was done before silver soldering the oxidation and pickling would have affected the accuracy of the seat for the ball

the cylinder bore was then drilled out I did try a reamer on some scrap brass but the fit was better with with a drill as the stainless rod I'll be using as a piston was slightly undersize. this would have been easier to do when in the lathe for the first time, but could have distorted slighty when silver soldering, and silver solder could have gone in the bore which would have ruined it.

a bit of oil on some stainless rod and the bore was "adjusted" to have a nice sliding fit

and the finished part

Im quite pleased with how it worked out and its a bit of a long post about such a small part, but its worth having a think about how you'll make things as with these things that the first option isnt necessarily the best. also dont be put off by things looking to be complicated, break it down into logical steps and it becomes easier to take on. if you look at the piece of brass in the first picture and the last one you'll see what I mean and they dont look they are the same part.

 

[Carl Wilson]

This is a great project. I really like how you work and how you do things. Drawings that you can understand and work from and enough precision to do what you want.

What I mean by that is precision where it's called for and appropriate to the application.

Does the intake for the pump come from under the hull of the vessel?

Looking forward to more updates.

 

[Wallace]

An excellent thread with superb work.

 

[arther dailey]

same as,loving it .Thankyou

 

[Carl Wilson]

Re the problem of getting enough flow from the pump for the motor speed/power, could you build a double acting pump similar to the one you've already made? Then you'd get twice the flow per stroke.

 

[Exuptoy]

I'm blown away. I'm just learning to use a lathe, and whilst I'm enjoying the experience, you have just turned everything on its head. Amazing!

 

[optima21]

thanks for your kind words guys

Does the intake for the pump come from under the hull of the vessel

The problem of getting enough flow from the pump for the motor speed/power, could you build a double acting pump similar to the one you've already made? Then you'd get twice the flow per stroke.

no I might just use a 50 or 75ml plastic bottle for supplying the water, that should be enough for 30 minutes running (at a guess).

a double acting pump would have been possible but I think its too much work for this project. the cylinder in pump body is longer than necessary so I can just make a new linkage and piston if necessary, buy I'll see how it goes. the reason for having the pump is to keep the boiler size down, if the pump doesn't work as I want I can just make a bigger boiler, and fit the engine in a bigger boat and build the boat and make it electric powered.

I'm blown away. I'm just learning to use a lathe, and whilst I'm enjoying the experience, you have just turned everything on its head. Amazing!

well thats part of the fun learning how to do things. the reason Im posting on here is that I know that there are a quite a few people with lathes and milling machines on here and its just to show you can do alot of things with a bit of thought, and if it doesnt work first time, think about where you went wrong and do it a different way next time.

lathes are very versatile, with a vertical slide they can can do most complex work like on a milling machine. when I bought my lathe I then decided that I needed something to make, and my first project like many others was a Stuart 10H engine, Hitch is building a vertical version of this engine. this is the engine that I built. when I made it, I didn't have a milling machine, just my lathe and a small 5 speed pillar drill

 

[Exuptoy]

Been looking at Stuart this week funnily enough when one of the lads in work mentioned he had built one years back. He reckons they will purchase a built unit from you once complete? that's mad.

 

[optima21]

yeah the thing about the model engines is that you have made them yourself. if you dont want to buy a kit, you can make them from bar stock, and Elmer Verburg made a wide range and style of engines in a book of his. his drawings leave some bits to be worked out yourself. they can be seen on the link below

http://www.john-tom.com/html/ElmersEngines.html

I was going to be making a no.24 beam engine

but as a twin beam engine as I like the motion like this one

so that'll be another project when Ive finished this boat engine but then next year I'll be working on my cars again

 

[Exuptoy]

Lovely!

 

[Carl Wilson]

I don't really understand the process of forming the seat of the check valve in the pump by tapping the ball in, could you explain this for me please?

Thanks very much!

 

[optima21]

for the one way valves in the pump I needed to modify the ends of a couple banjo bolts and here I am cutting the notch in the end of one, I guess I could have just used a hacksaw and file though

I seem to much more time thinking than doing on this project, creating solutions to problems that dont exist, and this happened again on the bracket to hold the parts of the water pump together. this was made from 0.5mm thick stainless steel. here is the blank with the holes drilled and edges filed, and the aldi step drill dill a good job of the 12mm hole.

and then this was bend in a vice between some metal plates and clamps

and the completed parts of the water pump.

and the completed pump, my thinking that by using a channel section I can keep the weight down and stop it from flexing. I need to make the main framework again when I've finished and the legs and screw hole in the base will need to be in different places. the whole assembly weights 28 grams and its stall current of the motor on 5 voltsis 0.2 amps.

so not it comes to do some playing with electricity, water and fire, so what could possibly go wrong. so this a quiet corner of my kitchen.

it comes in handy that I don't have an electric toaster, instead I have a fold up camping one, which is a good base for heating the boiler up. the electric wire are silcone insulated so are fine at the temperature they are exposed to.

and here I am playing with the electricity.

its pretty basic stuff though, a signal from the electrodes in fed into an inverting op amp that is used a comparator, with the output being fed to a transistor to work as an on/off switch for the motor. I did try using transistors alone for doing the switching, but the conductivity of the local tap water at room temperature wasn't giving consistent results. I was orginally hoping to run this off a single lipo battery (3.7v) but the lower the voltage, the more variable the results so I'l be using a 3.7 to 5.0v DC - DC step up converter with is 1/2" square so is small enough, and will give me a regulated power supply.

https://www.ebay.co.uk/itm/372899832490

our tap water round here is very soft so maybe I need additional impurities to improve the conductivity. our water has very low conductivity at room temperature and then it increases steadily until about boiling point and then it drops again. so far things are too variable to get consistant results, the room temperature issue is easy to get round, I just need to start off with warm water, Im wondering if changing electrode position might help too. its easy enough to get on/off switching a fixed temperature anywhere apart from near boiling point. If I cant get this part of the puzzle to work, the engine will still be made but I probably go for a simple pot boiler and meths fired burner like an old mamod steam engine.

are there any experts on here about the electrical conductivity of water?

 

[Carl Wilson]

Really fantastic work. The pump looks fantastic, well designed and fit for purpose. Maybe the final piece of the puzzle is to use some means other than the conductivity of the water to trigger the level sensing.

Not sure what though. And I can see that the conductivity of the water is the simplest way to go.

 

[Dieselman 63]

I think I may have to get one of those kits one day. The good idea fairy has just visited and I’m thinking of how a working model of Leawood pump house would be a cool little project

 

[arther dailey]

learning a lot on this thread

 

[optima21]

learning a lot on this thread

you're not the only one, its a steep learning curve for me too.

completely off topic, but I forgot to post this on my last post.

I recently saw a film from the mid 1960's about how they were making computers for the Apollo space craft. now I know what Im doing isn't rocket science but when I saw this and you see how intricate and fiddly some parts of the early computers were to make its a bit of an eye opener. the first 10 minutes explain how you find you way in space, the original sat nav, then it gets onto the computer bits.

 

[optima21]

I've now made a new gas valve, this time the opening of the valve is controlled bu the spindle screwing into the union for holding the sealing o ring on the spindle rather that screwing into the main body of the valve.

this is turning the point of the spindle, its 20 degrees internal angle and at the point is 0.6mm wide.

for holding this in the collet chuck it was necessary to make a bush to hold it, it was 2.5mm o/d and 2.0mm i/d, showing the i/d being reamed

and the tapered hole in the valve body was cut with a 20 degree engraving D bit.

and then the spindle was lapped to the body for a gas tight seal

I wonder if anyone else here seems to make parts that are smaller than the drawings rather than parts that are larger than the drawings.

and I needed some banjo's to test the water pump to see if it would work. the blanks were turned to size and the centre was couterbored so not to restrict the water flow, shown here dilling the hole for fitting the tail so I could fit some silicone tubing to it.

these were then parted off for delicate parts like this that can't be rechucked to clean up the back face, I sharpen the end of the parting tool at an angle so you dont end up with a burr etc. this parting tool is 1.5mm wide

and then preparing the banjo's for silver soldering. the lighter was moved to a safe distance before warming the area up

ohh and the pump works showing 50psi on the gauge, but it takes 10 minutes to pump 20cc of water which is how much water the boiler can hold its running on 5V and drawing 0.1A. I think it might improve with a few tweaks though and that is using the 100rpm motor and draws 0.1 amp. its likely that I'll end up making a new scotch link with a longer though and trying the 150rpm motor.

 

[optima21]

this may be a bit technical for some, and remember is wrote by someone who isn't technical , so there will be more pictures after this post.

they say you shouldn't mess about with water and electrics, but things get worse when you add a bit of heat into the mix. there are a few things that have an impact on using the conductivity of water for indicating and maintaining the water level.

1) the conductivity of water varies depending on the quality of water. pure water doesn't conduct electricity, so it needs a trace of impurities to work.
2) the conductivity of water varies according to the temperature of the water
3) the level of the water will vary slightly if the electrodes are making or breaking contact with the water.
4) as the water starts to boil the water level will increase slightly as the water is converted to steam and the bubbles disperse in the water.
5) the contact area of the electrodes needs to be as large as possible due to the low conductivity of the water
6) even though only tiny amounts of current pass though the electrodes, one will slowly be eroded due to electrolysis.

I hadn't really thought about any of them until it came round to sorting the electrics for maintaining the boiler water level, and I think Im bringing the steam age into the late 20th century.


from the above list there are a few variables to throw into the equation, and thankfully as long as it turns the water pump off and on at the right time I'm happy, so it was just a case of trying different things until you find something that worked and I found very little on the internet to help me.

so here is my circuit diagram that seems to work so far for me. I'm not an electronics engineer so if it looks wrong to anyone who knows about these things let me know . this is the first time Ive ever tried to use mosfet transistors. so my first thoughts about the circuit design was:-

1) I would be using a 3.7V lipo battery and thought a regulated 5v supply would be better as it would simplify the voltages dividers used for the input to the input of the op amp used as a voltage comparator.
2) limit the current consumption to a maximum of 150mA as the Lipo only has less than 500mah capacity and needs to operate the radio control gear to operate the servos for steering the boat and operating the throttle.
3) use the smallest number of components I could and to keep the size of the components down.

the circuit is really simple, comparing two voltages one is fixed and the other is variable depending on the water level, so if the water level is low in the boiler the pump will be turned on and if the water level is high the pump will be turned off. the fixed voltage I decided on was around 0.65V. and if the voltage between the boiler probes was less than this the pump would be turned on and if the voltage was higher the pump would be turned off. The op-amp I used to compare the voltages was a CA3160E as their output voltage goes from 0V to the full voltage, and they have fet input so dont need much signal to compare. I did try transistors first, but it was more complicated and the output switching had less of a voltage range. the output from the op-amp has a led to show when the pump was off so that if that is lit or the motor is running you know the circuit is running. The op-map is also used to drive a power stage to switch the motor on and off. I originally tried aZTX650 NPN transistor as the can handle upto 2 amp (I was only wanting 150mA) and are in a tiny package but at higher currents the output voltage would go to less than 3.5V so was wasting a significant amount of power (and potentially torque) for the motor. I decided to have a look at using mosfets and the BC170's seem to be working well with less power loss and drawing 150mA it didnt get warm. The output current for the motor was limited to 150mA using the 39K resistor. Im waiting for ZVN4306A Mosfet transistors to try as they can handle higher current, can disipate more heat and have less of a power less, but its all relative, but if you can have free power its worth looking into.

so this my my trail setup, the boiler was fired up on my cooker. the think at the top of the photo is the step up power supply, Im was amazed at how small it was and it works pretty well too. (shown in the bottom pic with the electrodes just for scale).

and a close up

I found the vertical electrode needed to be as wide as possible to give consistent results and the length of the electrode dictates the water level in the boiler. the electrodes are stainless steel but the discolouration will be due to the elecrolysis.

and a few stats about the boiler and electrics

1) the boiler water capacity 20CC in the main barrel, not including the water tubes.
2) with room temperature water in the boiler the pump turns off with 15cc of water in the boiler and the voltage between the electrodes was 0.75V
3) with 50C temperature water in the boiler the pump turns off with 14cc of water in the boiler and the voltage between the electrodes was 0.9V
4) with 80C temperature water in the boiler the pump turns off with 14cc of water in the boiler and the voltage between the electrodes was 1.1V
5) if the 80C water was heated up so it just started to boil in the boiler (you could hear it) the voltage increased between the electrodes to 1.8V
6) when the boiling water started to emit steam from the boiler, the voltage dropped to 1.5V
7) if the boiler continues to be steamed, the pump starts to intermittently on with 12cc in the boiler.
8) with the pump motor running freely it drags 0.06A and the supply voltage to the circuit is 5.32V and the voltage dropped across the motor is 5.2V
9) with the pump motor stalled it drags 0.15A and the supply voltage to the circuit is 5.18V and the voltage dropped across the motor is 4.4V

and why do I post things like this here.....well if I loose the bit of paper I write it on, I'll never find it again

 

[tom2207]

exquisite ,,, let me know when you start into clocks .

 

[Carl Wilson]

It would be good if the electrodes could be fed with an Ac signal. That would stop the electrolysis.