Heilmann
The Steam Electric Locomotive
I will confess to being a novice when it comes to steam power. Electric motors and associated control systems I will admit are my field (no pun intended -for once!). The Heilmann is therefore the obvious first foray into steam locomotion for me. There are no complicated valve timings -no real problems in that area at all. All the main design problems relate to the electrical generation side of the locomotive -something I am far more happier with.
The basic principle is that a steam engine is used to turn a dynamo and the DC generated is then sent to the power bogies in the usual manner. I have performed several experiments with motors run as dynamos and have come up with the premise that I am going to have to wind one of these for myself.
The design for the dynamo comes from information gleaned from "The Institute of Intermediate Technology" who have an article on DIY wind generators. I have decided to build a dynamo rather than an alternator -simply because the original had one and well an alternator with silicon diodes is I feel 'cheating'. This is not to say that the model will not have 21st Century control electronics!!!
The steam motor is made from three "Unit Steam Engines" these can be slotted together to provide any real combination or shape of motor that is required. I got mine from Forest Classics.
Here is a picture of the 3 units slotted together -the sum length is 12.5cm.
To provide steam for these I looked around and found the article "Little Steamers and their Boilers" I decided to have a go at a porcupine boiler -but modify the design for my own needs. This involved shifting all the quills to the back of the boiler and mounting the combustion chamber there. I know that I will need quite a lot of steam -but I don't think that it will be at very high pressure...
The combustion chamber will have 6 burners at 2 differing heights -fed from two fuel lines. This will give me some element of control ( 2, 4 or 6 ) -rather than all or nothing. They will be of the tube wick type burning meths. The tube wick burner is noisy -but very efficient.
The fabrication of the boiler was something that I had intended to 'farm out' but after discussion with the person who was going to do it he said 'Look, I will watch and tell you -and you will do it. Otherwise you will never learn!' So, I am studying my little leaflets given to me and wading through types of brazing metals/spelters/ silver solders and gasses to be used with them. The alloys that I had chosen for my step brazing assembly I checked with the manufacturers who suggested others... (Always check!!!)
The alloys they have suggested allow me to use 2 steps rather than 3. I can safely use MAPP gas rather than propane to assemble my boiler -but the only easily available MAPP gas burner has no handle and I think the idea of a 454 gramme bottle with just a burner head on it -is a bit unwieldy...
The magnets for the dynamo are proving to be another source of design problems. They have a low Curie point of only 80C -thus is the steam comes near them they cease to be magnets. The magnets themselves are SUPERB and cheaper than I thought possible at £1 per square inch. The dynamo will have 16 of these in 8 pairs giving me a field of 12,500 gauss each at the surface -but they will be epoxied to the steel stator cage despite this!
The "A" level physics says that the voltage I should get per 100RPM is 2.7V the motors that I intend to use can run quite happily on 2V -so it is down to the dynamo and motor combination to supply this amount of 'juice'.
The dynamo consists of 6 PAIRS of windings and 8 PAIRS of magnets. This should give a pulse every 15 degrees of rotation. What the wave form looks like on a scope is anyone's idea -but I suspect the only thing that could be said for it is : it is not AC...
Here is the working drawing of the locomotive at the moment -although this may change!!!
I spent most of this afternoon constructing the commutators... These have been made from yorkshire plumbing fittings -keen readers will note that the last loco from 'Kitchen Sink Engineering' also used yorkshire plumbing fitting -there are 4 left in the bag!!! These have been split into two between the ends and the two pieces split again lengthwise to produce the two sleeves. These have then been epoxied to a length of 15mm deal dowel. The construction has been centre drilled to 6mm and a side hole bored to allow the cables for the last two commutators to be connected. The epoxy has been 'built up' between the gaps to give a smooth gap free ride for the pick up shoe. This will be made from a short length of brass pipe soldered to a length of springy brass strip.
I had intended to have a 'positive' end and a 'negative' end to my dynamo -but the wiring required and the space involved preclude this. I am having a set of three at the steam motor end and this shifts the dynamo farther from the source of heat -so hopefully still keeping the magnets below the Curie point! Another thing I feel that I will have to do is: when the steam motor is chugging along I will require a series of switches to enable me to step in and out the three pairs of windings to vary the load on the poor motor.
At start up I will have none of the winding connected to the capacitor bank, then one, then two, then finally the third one.
I started fabricating the copper boiler based on notes from "Little Steamers and their Boilers". I downloaded the on line catalogue from a metal stockholders and began to pour my way through it. Suprisingly the hardest thing about it was understanding the specifications. I grew up in a firm Centigrade Centimetre Gramme Second environment -thus I have problems 'seeing' Imperial feet and inches and SWG specs might as well be in Cuneiform -for all the good of understanding them was to me. But despite this, and armed with the conversion chart at the end of the catalogue, I was able to produce dimensions that I could understand and deal with. The boiler will be fairly low pressure (2 bar) so I worked out that 1.5mm copper was capable of taking this amount of pressure if re-enforced with stay bars. So, I ordered the first installment and then set off with my Imperial measurements and metric tools...
This shows the boiler cylinder with the two end plates and the three stay bars through it. I normally draw everything out on graph paper before ordering or starting to build a project -but this is the first one where the drawing has had to have two sets of measurements written onto it -one for the designer to use and one for the supplier to use... Despite this everything has (so far) gone together quite well.
The real problem that I expected to happen did not materialize -that of marking out the required number of holes for the quills and drilling them. I draughted my datum line with a 2B pencil and then wrapped a sheet of graph paper around the boiler and drilled through the intersections with my pillar drill and a 1mm drill. These were then opened up to 3.2mm (1/8 inch) with a wheel brace. The quills are simply 3cm lengths of copper rod roughly rounded at each end.
The next shot shows the nearly completed boiler quills ready for silver soldering. I will use MAPP gas and a solder with a melting point of 710C as the first 'step' and then a lower point solder at 610C for the second 'step'.
There are generally two main methods of burning meths in a model locomotive. The wick method and the vapour method. I had been leaning heavily toward the vapour burner method until I ran head long into the problem of getting the vapour to the burners. In principle this is simply a hot tube in which the meths boils -but the distance and condensation of the vapour seems insurmountable at the moment -so I have decided to go with the wick method. My wicks will be home made by my son. I have already nailed 4 pins into the top of a cotton reel and he can help daddy making string tubes!!!
The method I intend to employ to get the meths to the burners is 'the chicken feed drip method'. This has a great appeal to me because it mimics one of my childhood wonders -my grandmothers "Magic Chinese Teapot". It was spherical and had no top but from its spout poured; first the milk -and then the tea. I later learned how it worked. There were two containers one for milk the other for tea -both filled from the inverted teapot and a cork bunged in the hole. A finger across either of the dragons eyes (one of the dragons formed the handle) blocked air to that chamber and allowed air to the other -thus either milk or tea came out of the spout -the other dragons mouth.
Here is a rough drawing to show you the idea -except this dragon will actually breath fire -not 'Gold Top" and "Earl Gray"!!!
It also means that the system will be air tight and will have to be filled in the same manner as grandmothers teapot -upside down...
I have been busy over the past few days and I have made both of the fuel tanks -the main tank and the sump tank. I have also rough cut the parts for the combustion chamber -I cannot really call it a firebox... The burners are arranged in three groups of two in a hexagon patten. I did use a set of dividers and mark them out that way! There is a central 25mm air hole and the spent exhaust gases will be vented from a twin 15mm tube assembly that will meet in front of the boiler and then vent to air. I hope that a wire mesh at the main air intake and exhaust side will confine the fire to the combustion chamber. The main fuel system will be filled with meths -as a safety precaution. Ridiculous as it sounds this makes the fuel system fireproof -as there is no air. The principle is the same as the Bosch fuel injection system in which everything is filled with petrol.
It does look rather like the back end of the rocket engine for the ME163 -but having three sets of burner 'should' give me some form of control for the heat input. There is the question of the tuned exhaust effect -but whether I could ever get this to work properly is doubtful.. If I could then the entire assembly should pulse at lower 'C'. The nearest thing to this would be the double disc venturi that I intend to fit on the smoke stack. Again -how much forced draught it would give a small boiler is questionable -but it should 'look the part'!!!
I am now taking a rest from all the metal bashing and returning to the quieter -if more irksome world of the dynamo....
The core of the dynamo is a piece of 55mm AF Aluminum hexagon bar with a 4mm hole bored down it's length. The shot also shows the 16 25mm square NIB magnets. To 'pose' this shot it needed a piece of blu-tack to hold the magnets to the bench top and another two pieces to hold the dynamo core -as the magnets attracted the steel shaft. Yes, some invective was required to separate the two....
To this has been epoxied six lengths of 15mm channel section.
The entire assembly has been rough balanced by sticking offcuts of copper sheet into the still soft epoxy and 'twirling' it until there is no "real" heavy point.
Well a couple of months have passed since I updated the progress on this -so I thought I had better continue. The reason is the fact that I am in the process of moving house and will in a few days time have my own shed to work from, The dynamo is completed and has been tested in the chuck of my pillar drill at the five belt speeds available to me. It took a little bit more initial balancing than I expected but it is quite happy at 900rpm. The design speed is 2,000rpm -so there is a little more needle file work to go yet.
The test data for my dynamo is as follows:
520rpm 2.80Volts rho=5.53
900rpm 4.04Volts rho=4.48
1,370rpm 5.55Volts rho=4.05
1,880rpm 7.64Volts rho=4.06
2,620rpm 11.49Volts rho=4.38
If you plot the graph you will find that rho (the voltage generation loss factor) is lowest between 1,400rpm and 1,900rpm. In practice the dynamo will hopefully turn between 1,500 and 2,000rpm. This fits in nicely with the design of the boiler which is designed to provide 12 litres of steam at 3 bar pressure. The calculations show that the boiler should be able to provide 4 cylinders at 3 bar at 2000rpm. I know I originally set out to use 3 cylinders -but the fourth one was added as a safety margin...
I have decided to go with the low voltage but high current Johnson motor. This is a 5 pole motor rated at 9 Volts and 10 Amperes. It will work quite happily at 1.5 Volts and 1.5 Amperes. The design for each power bogie will use two of these motors each powering two axles. Thus the minimum "draw" on the dynamo will be 1.5 Volts 6 Amperes.
I have also been working out how much power (wattage) I should be able to get from my 4 cylinder Unit Steam Motor. Working from "First Principles" I arrived at the following conversion method.
This method is incorrect as it does not take into account Adiabatic and Isothermal calculations-but it works for "ball park" figures...
1 Bar = 100,000 Pascals
= 100,000 Nm2
= 0.1 Nmm2
Area = 63.6mm2 (4.5 x 4.5 x 3.142)
Energy = 6.36 N over a distance of 18mm
= 0.114 Joules per Bar pressure
4 Cyl = 0.576 Joules per Second (Watts)
THUS: at 2,000rpm at 3 Bar the energy developed by my simple 4 cylinder motor should be:
X Watts = 0.576 (Joules) x 3 (Bar) x 2,000 (rpm) / 60 (seconds)
= 57.6 Watts
The next piece of design work entails the power supply system - which typically enough of me - is a hybrid design of old respected analogue technology and digital electronics...
The initial system is a simple linear voltage regulator fed from a capacitor "Surge Bank" consisting of 8 x 2,000uF electrolytic capacitors. The reason for the large number of small capacitors rather than one large one has to do with the amount of "ripple" on the main bus bar from the dynamo. If I had one large capacitor -then I would have to find one of 16 Amperes ripple rating. These do exist -but the price is simply hideous!!! Far simpler (and cheaper) to have a large number of small capacitors. The next stage is a very old fashioned 2N3055 used as a voltage regulator. This then feeds into a modern Pulse Width Modulated power supply using a NAND gate as the pulse length device. This is controlled by a simple variable resistor. The control FET is a very desirable HEXFET...
Thus the power supply system should allow upto 6 Volts and 3 Amperes per motor -or 72 Watts
The original had two eight wheel bogies -which this will have. However in order to get four axles around 2 foot 6 inch curves I will have to cheat a little and make the two bogies into four "sub bogies". This will also be the first time I have transverse mounted my motors -normally I mount then in line and use a worm and spur gear arrangement. Some people may blanche at the next drawing -but to be honest it is the best way of doing it. I had contemplated 16 SME motors each driving one end of an axle -but the voltage (hence dynamo rotation) required would have been well into the 12V or 3,000RPM band.
The power is taken from the Johnson Motor via a toothed belt to a central bevel gear. This then powers a shaft to the two worm and spur gears. The entire assembly pivots within the bogie and thus the bogie can take a sharper curve than would be expected.
Well a couple of weeks have passed and I have moved house, and I am working in the (somewhat) draughty corridor between the my work shop and the coal house (as was). The soldering of the boiler has to be done outside -I cannot take any chances with the models etc that are inside the workshop -now unpacked and on shelves...
I have been using MAPP gas from a Bernz-o-matic torch -which is very nice and easy to use -the main down side is the smell of MAPP (Methyl Acetylene Propadene) -which I can only describe as a combination of sweaty socks and ripe drain(!) I have experimented with two types of silver solder alloy Easy Flo2 and Silver Flo55. Either I am using it wrong(?) or Easy Flo2 is far too liquid and I am getting a better (looking) joint with the Silver Flo55.
One unexpected problem came with the assembly of the boiler. I couldn't get it to solder!!!
Having sat down with the book "Model Boilers and Boilermaking" I calculated my heating area and my evaporation of water in Cubic Inches per Minute evaporated at 100PSI for 100 sq inches of heating area. This the famous Henry Greenley Formulae.
The heating area of the boiler barrel is:
0.5 x 3 x 3 x 3.142 = 14.1 sq inches
The heating area of the quills is:
63 x { ( 0.0625 x 0.0625 x 3.142 x 2 ) + ( 0.125 x 3.142 x 1.25) }
= 63 ( 0.025 + 0.49 )
= 32.5
Thus giving a total heating area of : 46.6 sq inches.
As I am using 45PSI this equates to:
X = ( 100 / 45 ) x ( 46.6 / 100 )
X = 1 Cu inch of water per minute
So far so good. BUT what the book didn't tell me (and to to be fair it didn't even occur to me) was that heating area of a porcupine boiler also equates to COOLING area.... Being in the computer business I should have looked at my boiler and seen a copper flower CPU cooler -rather than a set of heating quills.
I borrowed a LARGE propane burner from a friend played the flame on the quills and this pumped heat into the boiler and I returned to soldering my boiler. When complete it will have to be pressure certified which will entail it being pumped up with water to TWICE its operating pressure for 15 minutes.
I have no illusions that my first attempt at certification will be a failure -and it leaks.
There were afterall 73 holes in the boiler...
Well at this point there are 2 holes left in the boiler, and I have suprised my self by not burning myself. Scalding yes -but burning no!!! The fittings for the boiler have been ordered and I await their delivery. I have standardized on 5/16th inch 32 TPI ME with 3/16 pipe for all my fittings. I did this in the fond hope that it would make my life easier.... However I have since found out that the tapping drill needed for this (very) common size costs £7.70p and is the very strange dimension of 7.1mm. None of the tap tools that I have will take this small a tap -so I have had to order a TEE tap to hold it in.
They came about 10 minutes after I had posted the paragraph above.. (typical)
I have begun construction of the firebox and its "exhaust system". Whether I will feed the waste steam from the steam motor to the exhaust flue (to produce a draught on the firebox) or rely on the electric fan running full time I am still not certain(?) Curiously enough the one problem that had not really entered my head was the problem of ignition.... I have been playing with electical ignition via heated nichrome wires. But based on simplicity I have decided to go with a match shoved into a hole in the side of the firebox!!!
When I was debating what sort of body work to give "the Hooded Swan", (as it did not at all resemble the original "Electric Fusee"), three things struck me. I had never built a streamliner, I had never built a fantasy, and I love Art Deco..
These are the original drawings of "Electric Fusee" from "Unusual Locomotives" by Ernest F Carter.
See Picture 18 LARGE HIGH RES IMAGE 3.5Mb
The elements that make up the raw inspiration for this are:
The "Hiawatha" Atlantic locomotive.
The Southern Electric "Head of Mercury" logo from the 1930's
The "Statue of Hera" from the Louvre
Several Italian "Futurist" posters that are amoung my collection
But most of all the 22 strong fleet of Metropolitain Railway BOBO locos that managed the changeover from steam to electric power at Rickmansworth!!!
Having a nice computer system to play with, the calculation of compound curves and their rotation and plotting was not a problem. What was, was transferring the plotted templates on A4 to sheets of cornflake box to make cardboard mock ups to see which one was the most acceptable in real life...
In the end none of my beautiful ornate curves made it to the final box of cornflakes. But rather a simpler (and I think better) series of facet curves. The idea is that of the head of an Art Deco woman with her hair flowing behind her and a head band/tiara in it.
The hood was made by forming sections of 3mm ply and then glueing alternate strips of 5mm sq pine and 10mm strips of balsa. The lot was then laminated on the inside with glass fibre "tissue paper" and resin. The hood has to be a tight fit as the front fan will cool the dynamo (hopefully) and the waste steam from the motor will all have to vent towards the rear on the hood.
There is as yet no exhaust system for the boiler combustion chamber and waste steam from the steam motor -these will have to be fabricated to vent either side of the boiler. The system I am proposing to use is a single venturi version of the "pepperpot" used by Rhodesia Railways during the Period of Sanctions. However this one lies on its side... The blast from the dynamo coolant fan and the waste steam are directed over the venturi which sits on top of the exhaust stack -itself with a blast fan. The boiler section is separated from the rest of the loco by what will be an air tight wall with just the steam pipes and the vent blast hole between them.
The electronics are well underway. I will not bore you with them other than to say they are very easy to build and so far I have constructed one of the two required PWAM units on Veroboard.
The fuel tank (300ml) is located at the face end of the swan and the box supporting the frame is where the electronics will be located. I am expecting HUGE RF shielding difficulties as there is a very large amount of electrical noise locally!!! Everything is "star earthed" in the best practice -however I am still putting a lot of Faraday cages around as much of the electronics as I can...
Well I finished the soldering for the boiler, fitted all the fittings and carted it on wet windy night to be tested, and it failed. However I was not disappointed as I had rather expected this! I borrowed the test equipment and with the aid of an UltraViolet dye (quinine from a left over bottle of "Indian Tonic Water"!) and set up the test gear by the cooker...
Here you can see the boiler with the BGRA official test equipment. It is a modified baking tin with a hand ram pump, a gauge and a stop cock. There is a flexible high pressure hose to a union which screws into the adaptor fitting on the boiler. The system is filled with clean cold boiled water and the baking tray topped up to the level of the release vent pipe at the stop cock. The stock cock opened and the system is primed with a few strokes of the ram. The stop cock is closed and the entire system is them gently brought up to pressure with the hand ram. I know it is very hard to see here but the pressure dial is registering 50 PSI and will go up to 250 PSI -which I think is unlikely in a model loco...
The quinine glowed in the UV light from the insecticutor showing where the leaks were.
I had one of the stay joints leak and two leaks were on the end seam of the boiler where I had to change hands to get around the other side. Fixing the leaks took two days and five attempts. The delay was caused by running out of MAPP gas and silver solder... The boiler is now "tight" and has taken 90 PSI for 15 minutes which would have certified it for 45 PSI working.
Well I at down a few months ago and did my fluid dynamics bit -read up on old 1920's books and began to design my burner. The burner plate is a hybrid tube wick burner. i.e. burning some of the meths causes it boil elsewhere in the burner plate and the hot vapours ignite and burn with a nice gentle hot flame.
That was the idea....
What was never explained in any of the books is how to fine tune your burner. With a gas one -you turn the valve, with coal -you stop shovelling. With meths it seems that thermal runaway is not an option!!! I made my wicks from cotton tubes (Matthew helped Daddy with the cotton reel and 4 nails knitting machine...) I injected 20ml of maths into the burner wells put it on the centre hob plate of the cooker, struck the match and very nearly lost the ceiling!!!
"Expect some flare when the wicks are new -but this will soon settle down".
The flare was about 1 metre high and lasted for about 10 very long seconds... Then it settled down to about 5cm high and burned to a nice black crown in the burner tubes. The next stage was to install the burner in the combustion chamber of the Heilmann. The cell was padded with ceramite (proof to 1200C) and a lot was expected of it. A large jug of water and a fire blanket were in readiness. The Heilmann was perched on bricks; the fan ON -the pump running the meths to the burners ON -but no ignition (?) Checking the wiring showed fat blue sparks of 5mm length -but NO flame. Lifting the sight glass and poking a lit match into the hole ignited the burners -but nearly cost me a few sq centimetres of skin.
The test 500ml of water boiled to 2 bar in 5 minutes 34 seconds and the initial test was deemed a success -then we opened up the combustion chamber and saw what we saw....
The aluminium mesh used as a fire screen was in a puddle in the bottom of the combustion chamber. The fan blades had an artistic look to them -they are blowing COLD air into the chamber so this must be from radiant heat...
The burner for the Heilmann is designed for the job.... It is NOT like the classic Mamod burner -but more resembles a gas turbine burner!!! I have tested the burner away from the boiler on two occassions now and yes the amount of heat is increadible. I am going to have to buy more ceramite "felt" and shield a few more parts... When running the fan forces the flame into a ring and the flame colour changes from yellow to blue -showing the combustion is very nearly complete.
The book mainly used to design this burner is Karl Noble Harris : "Model Boilers and Boilermaking".
I explained my problems to my father who had a look at the burner running and smelled the output. He asked me what I was burning and his youngest son produced a bottle of B+Q purple meths. My father then walked off to his back room and returned with a winchester of clear liquid. He looked at the pipes being heated by the wicks nudged a few with a pair of pliers and refilled the fuel pump from his winchester.
"Sparker?" he asked and I pushed the button "crack snap" and there were flames... Nice Quiet Blue Flames and no smell(!) There were slight puffs as the liquid boiled in the heater tubes and the burner stepped up to high gear. Soon a nice corona of blue fire surrounded the burner...
The clear liquid in the winchester was of course SVI "Spiritus Vini Industrii" or 95% Ethanol and 5% Methanol. Not Ethanol, Methanol, Pyrridine, Methyline Blue et al....
How it works...
The original design described in the book operates in the same manner -but his version is a large torus with a central meths wick space. I have shrunk it down to a wick around a central pipe to which fills with meths and is boiled by the flame from the wick around it. The aluminium mesh covered the central air hole in the copper plate and was (ha ha!) to protect the fan beneath it...
Wether fan assisted draught is required is now debatable -because the drops of aluminium that fell onto the fan blades must have caused it to stop working quite soon during operation(?)
A few more weeks have passed and it is now the Easter Holiday. A major design change has been agreed with my wife and we have converted to Butane gas -although I might go to Propane/Butane mix later. The burner matrix is a ceramic square 3 inches across.
Never ever having "run" a gas system before I arranged for the Club Secretary to give me a couple of lessons. There were a few pops and bangs and eventually we worked out the correct settings for my loco.
I was not happy with the smell of the exhaust products...
The first attempt was to increase the size and number of air holes in the mixing tube and move the position of the gas jet. This did improve matters somewhat but I still had a "luminous yellow" flame not the "bunsen blue" colour that I was looking for. So I modified the air intake to use a small pressure box fed from a CPU cooler fan. This runs at 6V and the smell of the exhaust is now what it should be -just hot fumes -without any of the acridity it had before. To illustrate how much improved the effect is -the matrix goes from white to orange in seconds.
NOTE: the fan blows air downwards onto the air intake -drawing warmed air from the exhaust stack pipes above it. The Butane tank fits at the rear of the loco and is quite hard to fill.. This is because I am very deaf and (of course) cannot hear the tank filling! So, I have to wait for the Butane to spurt out everywhere....
There was the first initial trial of the boiler and I ran it up to boiling point to see if anything was seriously amiss. The steam valve held pressure and when (ouch, ouch ouch) turned steam vented into the cylinders and the motor kicked over.
HOWEVER I had the most colossal of steam leaks at the junction pipes of the cylinders. The pipes to the cylinders were silver soldered, but the connection pipes between the cylinders were epoxied. I took a deep sigh, scraped all the epoxy off and cleaned and de-greased and tried again. The second time showed no leaks. The next test was to couple up the motor to the dynamo and see what happened. On the first attempt I did get some revolutions before it stopped. The key element here seems to be the tension on the springs holding the cylinders to the face plates. Too tight -it doesn't turn, too slack the steam escapes -and it doesn't turn.... The second "power run" has been better and this time the dynamo hit 3 Volts, (which equates to 600RPM) for nearly 20 seconds. The motor is still very "tight" and I expect it will losen up as it "runs in".
The next step is to strip everything down again and clean out any swarf that has been produced during the "running in" process.
A couple of weeks have past and it is now late April. I have added a FOURTH cylinder since, (as suspected), the three just did not have the necessary torque to turn the dynamo at speed. I initially "timed" them at 1-2-3-4 or sequentially -but there were vibrational problems and, (once again), all the epoxied joints cracked. I nearly lost a few parts of my fingers from live steam burns.... After a soothing bag of frozen peas I stripped the steam motor down and then did what it tells you to do in the assembly instructions -use lengths of silicone tube to join the cylinders together. These were wired into place with offcuts of 1mm wire.
I then tried what is called "Odd Fire" 1-3-2-4 but although that turned very smoothly it still did not have the torque required. So, I tried "Even Fire" 1&3 + 2&4 with PAIRS of cylinders firing -thus giving me twice the torque -but nowhere as near a smooth a running.... The even fire system has worked and I am now getting reliable power runs of 4.5 Volts (1,400 RPM) for bursts of 90 seconds or more. The last test produced a sustained power burst of 4 minutes -but at 2 Volts. Interestingly enough it seems that the initial torque needed to set the system in motion is far higher than the continual steady state. The last run required 45PSI to begin -but was running quite happily on 20PSI for the duration of the run.
The next problem that has arisen is the coupling of the electric motors to the DC power supply provided by the dynamo. ERRRMMM.. This is anything but flat DC at the moment and the electric motors move "somewhat" at odds to the dynamo. There is a slow feedback oscillation effect. When the motors are running at constant speed the load = braking force on the dynamo is constant and everything is fine. HOWEVER, the speed of the dynamo varies as the load produced by the electric motors causing the work done by the steam motor to rise and force it round again. The lag on the system is around 10 seconds. I have reduced this by having a huge capacitor bank between the dynamo and the electric motors. Eventually of course an R/C ESC will sit between them (see above).
Lubrication (oil) has been a problem. The stated oil for the Unit Steam Engine is a std motor oil, but it is too thick to get to turn when cold -so I have been trying various "steam oils". None of which has seemed to work. However I do have an alternative oil that has (so far) worked superbly: from cold -to smoking hot.
Castor oil.
Attempts at some form of displacement lubricator have not been successful, so at the moment I am simply squirting it before the run and during the run -with my mouth firmly shut!!!
I think I am now at the point where I can put the boiler and steam motor side of it to rest -while I start to build the power bogies.
Well after a few back breaking weeks in the garden Summer has arrived (thankfully!!!) The power bogies are well under way and I have begun the intricacies of the power gearing and distribution. I have decided to go with a Klein Linderner axle at the front and rear of each power bogie, but given the wider gauge of the bogies, (this is my first Gauge 3 loco), this will enable me to fit most of the work externally -rather than having to cram it into a 15mm pipe fitting in between a 29mm gap. Hopefully this will produce a better handling loco.
Here is the Heilmann on its bogies for the first time.
It is somewhat of a beast!!! The construction track is 2 metre lengths of angle iron screwed to a few off cuts of 2x1 on a length of MDF. I am convinced that if I tried to lift this (it weighs in at 10Kg) on std Gauge 3 track -it would simply bend...
As of yet the power control equipment is not behaving itself. The HEXFETs each seem to want to do their own thing and despite the fact that both draw the controlling voltage from the same variable resistor the output voltages are off by as much as 5%(?) This may not sound much -but it translates into one additional turn of one set of wheels for one bogie compared to the other for every twenty turns.