If it doesn't work, fiddle with it...

'Spinning up' the turbine with an air blast is a pretty effective way to start it, but when it comes to making careful measurements, the fact that something is forcing air into the engine gives a problem.

I can't tell if the compressor is adding anything to the airflow. Certainly the speed increase I get when burning fuel would be accounted for simply by the heat expansion of the air being forced in. (About a factor of three in temperature gives a factor of three in volume - equals a factor of three speed increase)

So: I dug a small electric motor out of the junk box and mounted that to drive the turbine shaft via a rubber band/pulley.
With the engine spinning, I now had only the air being supplied by the compressor and could watch the effect of burning fuel inside the combustion chamber.

When the fuel was ignited there was a distinct and measurable increase in speed as the turbine started to generate it's own power and 'help' the electric motor.

The increase in speed was significant: from 4500 rpm cold, to 7500 rpm at the limit where the NGV begins to glow bright red.

So I'm definitely getting a power increase. Question is: How much?

Observation

It's interesting to note that power is still being supplied after the fuel gas is turned off. The combustion chamber remains hot for some time and continues to heat the air, supplying power to the turbine. Speed gradually returns to the original 'cold' value, but it takes a couple of minutes.

Another motor

Repeated the experiment with a different, more powerful motor. I geared it up so that the motor was heavily loaded and wouldn't max out in terms of revs (that's important - once the motor begins to reach it's own speed limit, internal resistance begins to dominate).

This time I got 7,500rpm cold and 12,000rpm hot. At these speeds the compressor was supplying a lot of air - I could open the gas bottle tap much further before the exhaust reached red-heat. Interesting that the ratio hot speed to cold speed is about the same (1.6 : 1). It's probably telling me something... but I don't know what.

The test self-terminated when the rubber band broke.

Another way to test the compressor

Bloody infuriating!

It still refuses to run -just!.

Once I get it spinning and ignite the gas, I can back off the hairdryer until I am standing two feet away, just playing the wimpy little hairdryer on the air intake from that distance. The engine keeps running with just that gentle draught, but if I switch off the hairdryer it slowly winds down.

Now I've run out of propane and have to wait for a delivery.

Later that same day...

It occurred to me that since I know how much airflow it takes (1 hairdryer's worth) to spin my turbine at a particular speed (4000 rpm close-up), if I spin the compressor at that speed, in a perfect engine I should be able to get the same airflow. (In a perfect engine, the massflow produced by the compressor would be precisely equal to the that required to spin the turbine - it's a perpetual motion thingy)

Obviously, the engine isn't ever going to be perfect, but I'll get a figure for the relative 'goodness' - an overall efficiency figure.

So I removed the turbine wheel and set up an electric motor to drive the engine shaft at 4000rpm Then I taped a big plastic bag over the exhaust and switched on the motor.

It took 30 seconds to completely inflate the bag. Motor speed was a little higher than I intended (about 4500 rpm), so call it 35 seconds to compensate for that.

Then I repeated the experiment, using the hairdryer shoved up the air intake instead of the electric motor.

That took 15 seconds to inflate the bag.

So the compressor supplies a little less than less than half the airflow that would be required to make the engine perfect.

That's not as bad as it sounds, an overall 50% efficiency figure would actually be acceptable. If the turbine and compressor are 70% efficient each, that's the resultant efficiency you get (70% of 70% is 49%). According to the books, those are typical figures. It looks as if my gut feeling that it's almost there is bolstered by the measurement - it comes out at 43% in this case.

Dimensionless constant

In turbo theory there is a number of formulae, used for describing compressors and turbines that are known as 'dimensionless'. This is because they correspond to constants in the theory, which can be used to compare and model different designs.

One of these is the 'supply coefficient', which defines the perforance of a compressor in delivering a flow of gas.

It's defined as the radial flow divided by the rotational speed and it's a constant for a given design. Turbocharger compressors typically have values from 0.26 to 0.30

Schreckling gives figures in his book which allow the value of the supply coefficient (Sc) for his compressor to be calculated. It comes out as 0.23.

Sc can be written as : volume flow/(rotational speed * wheel diameter³)

In Schreckling's case, that's 0.1/(1250 * 0.07³) = 0.23

I measured the volume of my plastic bag and it came out to 0.1 m³.
I also did several more runs at different speeds and came up with a consistent value of 60 seconds to fill the bag at 33 revs/sec [proportionally, at 132 rev/sec it was 15 seconds].
So my supply coefficient is 0.1/(60*33*0.06³) = 0.23

It's absolutely spot on!

Schreckling style compressor

One thing that isn't mentioned anywhere in the books I have is the question of blade length.

Although the diameter of the impeller is involved in calculation of the ultimate pressure and the blade angle also is taken into consideration, there is no mention of the importance of the blade length. (Even the number of blades is mentioned as a design factor)

Obviously though, the length of the blade is going to have an important effect on the airflow. (In the limit, a zero-length blade will not work at all)

Looking more carefully at the compressor design in Schrecklings book, I noticed that the blades are pretty well as long as possible. And looking at the design of turbocharger impellers, I can see that they are made with blades that extend forward along the axis, making them much longer than they would otherwise be.

The blades of this Mitsubishi wheel are at least twice as long as my own design attempt.

This photo was taken before I trimmed the wheel to size and opened out the inlet, reducing the blades even more.

Anyway, I made a new wheel, based as closely as I could on Schrecklings design but in Carbon Fibre. (Photo later, just now it's in the engine, cooling on the bench)

When I installed it and ran the engine up, there was a distinct improvement in response. Still not self-sustaining but about 20% higher ultimate rotational speed for the same fuel input.

Unfortunately, something is tightening as the engine heats up to operating point. It frees as soon as the engine begins to cool but it's stopping me from making any real observations. That, and a hot spot in the combustion chamber, clearly visible in the exhaust need to be addressed before I can go any further.

Later

Found the cause of the tightening. The rear bearing hasn't quite enough axial play. There's a temporary packing between the turbine wheel and rear bearing that shouldn't really be there but I was using it to adjust the position of the wheel with respect to the NGV. When I removed it, the tightness went away.

Some figures. When cold, the engine rotates at 15ms per rev with the hairdryer pushed against the intake (4000 rpm).
Running, with the NGV glowing bright red - estimated 700 degrees C, the speed is 6ms per rev (10,000rpm). So that's an increase of 4:1 in speed for an (absolute) temperature rise of 990/290 = about 3.4:1, suggesting that some at least of that speed is attributable to an increase in massflow as the compressor spins up.

I could be off in my estimate of the temperature but to account for a 4:1 speed increase on temperature alone (without an increase in massflow) would require a temperature close to 900 degrees.

New Compressor

Compressor test rig revisited

Completely remade the compressor wheel and diffuser assembly. Put the whole thing together... and the difference? None!

It's really wierd that whatever I do seems to make no difference.

Well, no, it isn't really. The problem is that faced with a lump of engine that doesn't work I don't have any feel for what might be the problem. If it were a piston engine, there wouldn't be a problem but the jet turbine is something where I have no experience to go on.

I keep coming back to the fact that I don't have any facts or figures that relate to the actual engine. Calculations show that at such-and-such a blade angle and such-and-such temperature, the turbine wheel should generate yeay watts of power, compressing so many kg of air per second to such-and-such pressure. But there's no way to tell if it actually does.

Anyway, I do know that there is something wrong with the combustion. So I've worked out a way to drive the compressor section from an electric motor and run the combustion chamber in the airstream.
This avoids having a blower just blast air into the engine which doesn't tell me anything.

I've been trying to do this before by driving the shaft directly from a motor, which had to be placed right in the exhaust (putting it at the front reverses the sense of rotation and runs the compressor backwards).

For obvious reasons that wasn't too good. The motor was in the way when looking up the pipe and had to be on a long shaft to get it out of the exhaust. The long drive vibrated and shook the engine/motor, affecting the alignment and causing speed variations. Igniting the engine caused the shaft to expand, creating friction that had me on a wild-goose-chase for a while.(It looked like the bearings were starting to lock up when hot.)

Then I hit on the idea of driving the engine from the side using pulleys. It works fine!
The shot above shows the compressor running with the engine lit. I get a beautiful clean blue burn with the flame 'locked' in the fore-part of the chamber even at full pressure from the gas bottle.

Now I have to think about what this allows me to measure.

Update

With the motor driving, I get a clean, blue burn. I reassembled the engine with the NGV and turbine in place, then tried again.

The motor drive spins the compressor fast enough to produce the same clean burn but the moment I take off the drive the combustion collapses into yellow.

Is it possible that I'm not getting enough power from the turbine wheel? When I shaped it, I left the blades rather thick - almost 3mm at the tips. Maybe that is affecting the turbine power.

Update - 12th Feb

Reshaped the turbine blades, making them much lighter and about 1mm thick. Also added extra vanes to the NGV assembly to smooth the airflow further.
Still no improvement.

A bit of foreign material got into the compressor so I had to strip it down. On checking measurements, I find that the wheel is rather thicker than I designed it for. Because it is broadly conical in shape, when I trimmed the diameter it widened out the blade tip height to almost 10mm instead of the 8 I was aimimg for. Since my calculations show an ideal height of 6.5mm, maybe I have gone too far?
I'll make a new one.

Much thought later...

I've reached the conclusion that my compressor is too 'thin'.
Although it reaches the required pressures, it doesn't deliver enough volume of air.

I've tried to calculate the dimensions but this is an area where the treatments presented in both Kamp's and Schreckling's books are very vague. Although there are several pages of math, when you read through them carefully they make little sense. The symbols used don't always correspond to the diagrams and formulae are often presented without proof or reference.
I suspect the translation in both cases, there's no excuse for swapping subscript symbols between 'th' and 'w' in mid-argument. But it's just the sort of typo you would get if the editor didn't understand the subject.

Whatever; Although it's possible (with difficulty) to calculate the blade height for the compressor wheel from the shaft power and other dimensions, the result relies so much on assumptions as to be almost useless.
The probable error due to 'unknowns' in the calculations is wider than the operating tolerance of the engine.

I will have to experiment.
The KKK compressor used in Kamp's design has a tip blade height of 5mm. whereas Schrecklings is 8mm.
Mine is 6mm. I will try making one about the same as Schreckling's.

Unfortunately, that means I will have to remake the entire diffuser assembly too.
Still, it's a chance to use that aluminium ingot I made.