13
$\begingroup$

Looking at the history of jet engines, I noticed that the British first made centrifugal-types, but the Germans went straight into axial-types.

Which type is easier, if any? By that I mean, which type is easier to design and produce? As far as I can tell, they appear to be very similar pieces of rotating machinery, just the compressor blades are shaped differently and the airflow follows a different path.

Are the equations of centrifugal flow somehow much more complex than axial flow, or vice versa? Are the shaft bearings of one much different than the other? Are there tougher geometric problems of lubrication with one but not the other?

Note that I am not asking which is more efficient or the better choice. That would have to take into account much more than the cost to design and build something. For now, I'm only asking about which type of engine is easier to build, if any.

$\endgroup$
1
  • 2
    $\begingroup$ +1 All my career, I have been told that the blades in a centrifugal compressor are much easier to manufacture. I too would love to see a detailed answer. $\endgroup$ Feb 1, 2019 at 17:16

2 Answers 2

21
$\begingroup$

Centrifugal is easier to make. It's just a big vacuum cleaner with a kerosene burning furnace behind it instead of an electric motor. You have a one piece impeller, as opposed to numerous little blades and discs, and can be cast or machined from a block of metal. It's not dependant on the aerodynamic lift of blades to force air from a large space into a small space. It just pulls it in and flings the air out into a small space, so it's more resistant to surge. To get surge in a centrifugal compressor engine you have to really obstruct the inlet, whereas with an axial engine you just have to disrupt the flow enough to induce blade stall.

So in the 30s it was WAY easier, cheaper, and faster, development wise, to get a nice running engine using Whittle's design with the centrifugal compressor. The big downside is that it is less efficient than axial. You can't get the compression ratio that you would get with an axial because it's difficult to incorporate multiple stages, and the engine is quite large because of the need to fling air out to compress it. The Germans went with legions of expense, grief and operational delicacy for the better efficiency and smaller profile of axial.

The most obvious example today is the fact that nobody uses axial compressors for turbochargers and most in fact use similar impellers as the turbine as well. You won't see axial compressors on any of the little micro turbojets used in RC. A lot of smaller turboprops also use centrifugal compressors, sometimes in combination with axial stages. That should tell you something about how much easier and cheaper it is to make.

$\endgroup$
2
  • 5
    $\begingroup$ I'd like to add that some low thrust turbofans use centrifugal compressors for compactness and weight too. $\endgroup$
    – user14897
    Feb 1, 2019 at 20:07
  • 1
    $\begingroup$ Good point.. shorter. $\endgroup$
    – John K
    Feb 1, 2019 at 21:48
3
$\begingroup$

Centrifugal compressors were common on WW2 piston engines, so jet engine designers could build on this experience. And those centrifugal compressors had heritage in all kinds of liquid and gas pumps.

Axial compressors have the advantage of a potentially higher compression ratio, at the cost of being more complex (more compressor stages).

speculation on my part:

A centrifugal compressor can be machined from a single billet, and the back plate serves to reinforce the individual blades, so it's easier to make a compressor strong enough.

It also means there's more surface area to cool the blades, so you have less problems with the blades melting (on the turbine side).

$\endgroup$
2
  • $\begingroup$ Regarding cooling, neither axial nor centrifugal compressors need any cooling, as far as I know. I've never heard of any problems/dangers of compressor stages melting except maybe for supersonic aircraft. $\endgroup$
    – DrZ214
    Feb 2, 2019 at 23:03
  • $\begingroup$ fixed that, thanks. $\endgroup$
    – Hobbes
    Feb 3, 2019 at 8:12

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .