Once more, after a long period of working for a living, I was struck with a burst of energy and decided to do a bit more fiddling with the turbine engine. Presented here, for your entertainment, are the current results.
The first thing I did was to add some odds and ends to the starting box mentioned in the last exciting episode. An external fuel tank from a lawnmower was found in a scrapyard, and cleaned up nicely. This was mounted on one end of the box via a couple of brackets made from some folded 3mm x 25mm mild steel bar. The tank conveniently had a 1/8" BSP port in the bottom, and this was connected to the fuel pump inlet via a length of 4mm ID transparent PVC tubing. The external tank is a departure from the original plan, as I had wanted to put the fuel tank inside the box. However, I was unable to locate the correct size and shape of paraffin-proof container, and when the lawnmower tank presented itself, decided to use it instead. The advantage is a somewhat larger capacity, with the minor disadvantage of a slight decrease in neatness. I also removed the fuel control needle valve, as it was doing absolutely nothing useful. I added a pressure gauge to the pump output, which measures up to 25 bar. The pump can easily produce this pressure, and puts a surprisingly small load on the motor in the process. I also drilled a small hole in the side of the box to allow adjustment of the fuel pump pressure via an allen key. This hole can be seen just to the right of the cable connector in the right picture below, about an inch up from the bottom. After a bit of experimentation, I finally managed to eliminate all the leaks in the plumbing, and the starting box no longer slowly fills with paraffin when in operation. I have decided that the natural habitat of paraffin is the floor. It manages to ease its way through the smallest of leaks, much like hydrogen, only slimier.
Having sorted out the plumbing, I turned to the electrical system. A switch panel was added, along with a 17Ah 12v sealed lead acid battery. The rather larger one I originally acquired for the system turned out to be faulty, and was replaced with the current one, which also fits better. The switches, from left to right, are:
The last switch is needed to allow the fuel line to be drained back into the tank at the end of operation, otherwise the fuel gets everywhere when the couplings are released. Another switch may be added to control the start propane flow via a solenoid valve at some point. In addition, there will be a fuel flow control which will vary the speed of the pump. There are also holes drilled above each switch for LEDs to indicate the switch status, but these have not been fitted yet, along with the final fusebox. The gap at the right-hand end of the switch panel is for an instrument panel I am in the process of building, which has two 4 digit LCD displays, one of which indicates kRPM and the other of which displays the temperature in deg. C from either the TIT or EGT thermocouples, selectable via yet another switch.
The wiring for the system terminates in an eight-way bayonet style socket on the end of the box, beside the fuel connectors. This is connected to the engine via about 2 metres of cable. The end of the cable is clamped to the engine frame with a metal P-clip, as shown below. From there the various wires go to the ignition system, start valve, etc. The engine is wired up as a 12v negative-earth system, like a car, to make wiring easy.
The next step was to modify the starting system. For some time I have wondered about the feasibility of using compressed CO2 as an energy source for spinning up the turbine, with the ultimate goal of building a completely portable engine. Carbon dioxide is readily and cheaply available, is as safe as a compressed gas can be, and is usually at a pressure that is fairly easy to work with, at around 40-60 bar. The only other alternative that wouldn't use prohibitively large or heavy cylinders is compressed air in the form of scuba tanks, but these are considerably more expensive, and difficult to get refilled, and are at a very high pressure of around 300 bar or more. While the energy storage is much more than that of a CO2 bottle of comparable size, the pressure is so high it is difficult , and not cheap or particularly safe to utilise it. The obvious problem with CO2, though, is that it makes a damn good fire extinguisher, and if used via my original air starter would flood the combustion chamber and I wouldn't get anywhere.
The only practical solution was to use the gas jet to spin the turbine wheel, rather than the compressor wheel. This is actually more efficient, as the shape of the blades is such as to extract the energy of the high-speed gas more effectively than if it is directed onto the backs of the compressor blades. The airflow through the combustion chamber driven by the compressor pushed the CO2 out the exhaust, without affecting the combustion flame. The biggest problem is the fact that the inside of the turbine volute is a very hostile environment during operation, with very high temperatures and high gas speeds, which would tend to melt and erode most easily obtainable materials. However, I managed to obtain some very high temperature stainless steel hydraulic tubing from a scrapyard, which originally came from a military aircraft. It even had a compression fitting on the end to which I had an adaptor for the 1/4" BSP fittings I have standardised on. I drilled a 10mm diameter hole into the side wall of the volute, slanted in the direction of rotation. At right-angles to this I drilled and tapped an M5 hole to accept a grubscrew to fix the tubing in position. After a hell of a lot of effort (that tubing is incredibly hard and tough) I was finally able to form the pipework into the correct shape to fit through the hole to within a few millimetres of the turbine wheel, pointing in the correct direction. The other end of the tubing is bent around to connect to a solenoid operated valve mounted on the frame of the engine. The tubing can be seen in the lefthand picture below, going from the valve assembly at the lower right side of the engine frame, over the top of the frame, and down into the volute behind the oil line. The air line I was using for testing purposes is connected via a PCL coupler to the inlet of the valve, and is the blue hose going off the right side of the picture. The wiring at the bottom of the valve connects to the cable going to the starter box.
The picture on the right is of a Sodastream tm CO2 bottle from a UK drinks making machine, with a home-made outlet valve/adaptor connected to it. These contain 250g of gas at about 50 bar pressure, which I used to test the start system after initial tests with compressed air had been successful. The performance on CO2 is excellent. The turbine spins up much faster than on compressed air, due to the considerably higher pressure, with a lot less noise and to a higher speed. The gas does not interfere with combustion, and in fact helps a bit with damping out any flaring from the exhaust during starting. Starting is now very easy, being just a literal push-button operation. There is, however, one major problem. The bottle shown above proved to contain enough gas for TWO starts only. I need to obtain a larger bottle, and I'm going to get one of the ones used in pubs and the like for beer pressurisation, which contain about 10kg of gas. This should be sufficient for many starts. The other thing I am still a bit worried about is cold gas hitting a hot turbine, when restarting after a run. Hopefully the inlet tubing will act as a preheater and prevent any nasty results, but only time will tell. I have sampled the sound of the new start system, which can be downloaded here (98k). Please note that to save space I am now storing all samples as MPEG Audio Layer 3 files, which gives the best compression ratio. A good player for such files on the PC is WinAmp, which can be found at ftp.cdrom.com , amongst other places.
A picture of the complete system is shown below. The large propane bottle is what I am using for current tests, since I still haven't had time to sort out the problems I've been having with the liquid-fuelled system. I switched to a smaller bottle, though, because I had to take the 23kg one back to work for the heater there over the winter, since otherwise we all froze. I also obtained a high-pressure (2 bar) adjustable regulator for the bottle, which allows a lot more gas through than the original 0.5 bar one I had before. The system as shown below is essentially self-contained, allowing noise-making and neighbour-disturbing to be carried out in remote locations very easily.
In addition to the modifications to the existing engine, I have started acquiring parts to build a new one. Below is a picture of a Garrett T3 turbocharger I recently got for 25UKP, from a local car breaker. It's from a BL Metro, and is in quite good condition except for the bearings, which are totally shot. It obviously had a lubrication failure at some point, and the turbine end bearing was running almost dry. The end result is that the turbine bearing is worn to an amazing degree, giving about 2mm radial play, as opposed to the correct .2mm or thereabouts play. The thing was completely clogged with carbon when I got it, and the bearings were seized, even though it had been recently removed from a running vehicle. I spent some time stripping and cleaning it, and scooped out an amazing amount of gunk from inside the CHRA. The turbine end oilway was packed so totally with carbon deposits that I ended up having to drill the stuff out. After all that, the turbine appears almost completely untouched. It needs new bearings, but should run correctly. The good news is that since the thing is of fairly recent origin (1986) and of a common make, a bearing kit is fairly cheap. In addition, the turbine inlet manifold is exactly the same size and shape as the one on my existing KKK turbo, and all the holes are in the same place. This means that I can use the same combustion chamber on the new turbo, only needing to make an adaptor for the compressor output to connect it to the combustor air inlet. It's a bit smaller than the current turbo, but otherwise very similar in proportions. The main difference is that since this one was designed for a petrol engine, rather than a diesel one, it should be able to take a higher turbine temperature safely.
Larger versions of the above pictures can, as usual, be downloaded below.
As a final note, the picture below is proof that everyone has an off day sooner or later. After months of working with flammable and explosive fuels, high speed rotating things, high temperatures, irritated neighbours, etc, completely unscathed, I finally succumbed to a moments carelessness and nearly removed my thumb with an upsettingly sharp drill bit. Life can be a sod sometimes!
Back to the beginning of turbines. Back
