Math for axial compressors

[ddthj]

Hey guys pretty much a newbie although I've been watching diy turbine builds for a long time now, I've always wanted to make an engine with an axial compressor but I really don't know where to start when it comes to the math.

I don't have any performance requirements other than to self-sustain, and only one design constraint:
Maximum outer diameter ~220mm

I'd also like to have as low of an operating temperature as possible, because that may allow me to 3D print parts for the compressor side, although I assume that would create issues of its own since lower temps would require a higher pressure, which would just mean more heat generated by the compressor.

If you have some sources to point me towards that'd be great, the information I'm looking for includes:

number of stages
inner diameter for every stage (since they increase to account for continuity)
blade airfoil shape, number of blades per rotor/stator 
blade angle of incidence
blade height
etc.

I understand some of the velocity triangles for the blades, but I don't know a way to tie that into the rest of the design, any help would be appreciated.

[azwood]

Mar 17, 2018 22:35:50 GMT -5 ddthj said:

Hey guys pretty much a newbie although I've been watching diy turbine builds for a long time now, I've always wanted to make an engine with an axial compressor but I really don't know where to start when it comes to the math.

I don't have any performance requirements other than to self-sustain, and only one design constraint:
Maximum outer diameter ~220mm

I'd also like to have as low of an operating temperature as possible, because that may allow me to 3D print parts for the compressor side, although I assume that would create issues of its own since lower temps would require a higher pressure, which would just mean more heat generated by the compressor.

If you have some sources to point me towards that'd be great, the information I'm looking for includes:

number of stages
inner diameter for every stage (since they increase to account for continuity)
blade airfoil shape, number of blades per rotor/stator 
blade angle of incidence
blade height
etc.

I understand some of the velocity triangles for the blades, but I don't know a way to tie that into the rest of the design, any help would be appreciated.

Talk to anders or John here think anders 3d prints then sand coast his parts not that hard to diy cast from alloy.

[racket]

Hi

I think you need to read , read and read some more on axial compressor design , its extremely complex getting all the stages to balance out .

Low temperatures mean low power :-(

Unless you're a very competent constructor don't even attempt it .

The GTBA is the place to enquire www.gtba.co.uk/ , there are very bright guys on that Site who do know the answers to your questions , but will probably try to convince you not to waste your time .

Making the compressor is only the first of your problems, then you need to have a combustor that can cope with the high airflow rates that an axial comp produces , then theres the turbine wheel to drive things ................unless you've built a number of normal turbine engines using centrif comps to fully understand whats required , an axial build is just fantasy .

Sorry for sounding so negative about this.

Cheers
John

[smithy1]

What John said....Do I understand you correctly, you wish to 3D print plastic compressor parts..?? What material are you planning to make these parts out of?? These parts may need to spin at ~60-80,000rpm....it just wont end well, add to that the rpm, temperature and pressure the compressor is required to produce just for it to run and you'll end up with a dripping mess.

Even the best of aluminium compressor wheels can't stand those types of forces for to long, we're talking many hundreds of kgs of tension stress on each blade at tip speeds approaching the speed of sound. Even at a lower speeds down near self sustain ~20-25,000rpm, we're still talking huge stresses, a 3D printed plastic compressor will come apart well before getting there, let alone last long enough to reach self sustain.

It's probably best to experiment with an appropriate turbocharger type turbine to start with...this will give you experience to know what to look for and the pressures & temperatures involved etc...

Those of us who have been playing with DIY turbines for many years have a healthy respect for the damage they can do to the unsuspecting. They can be a lot of fun and building one yourself brings a lot of enjoyment...but they can be very dangerous if not treated with respect.

Smithy.

[ddthj]

In response to John and Smithy:
Yup, I understand the monumental task of doing something like this, or at the very least I understand my slim chances of finishing it.

smithy Yes I believe there are several components that I should be able to print. Material would probably be ABS or Nylon. I'm unsure if I could print the rotors themselves, if not I could print a press mold to press aluminum sheets to shape the blades. the rotor itself could potentially be printed as well, and the stators for sure.

Again I just want to see if I can make something self-sustain, if by chance the materials and design I use are able to stay together at whatever stresses sustain rpm would produce, that's why I came here asking about the math, so that I can see for myself if I can design something that looks good on paper.

I'll check out gtba too it looks like a great source,

Thanks for your comments,
-Mike

[racket]

Hi Mike

The biggest hurdle you'll mount is one of efficiency , if the comp isn't efficient then the turbine wheel will need to supply extra power and that generally means high turb temps , which exacerbates its construction as you'll be needing some sort of Inconel rather than stainless .

Then theres bearings and rotor dynamics to consider .

There are examples of axial comps on GTBA , so thats your best bet :-)

Cheers
John

[turboron]

A good place for you to start on small axial compressors in the highly successful Allison/Rolls-Royce C20B Series II (military version is the T63) gas turbine helicopter engine. It has a six stage axial followed by a centrifugal compressor with a pressure ratio of 6/1 and a mass flow of 3.0 pounds of air per second. Used compressor wheels may be available on eBay from time to time. The wheels are cast integral with blades. The vanes are cantilevered from the compressor case. Millions of $ were spent by General Motors, who owned Allison during the development, to solve the field problems with blade failures. 


For your reading, consider "Principles of Turbomachinery" by D. G. Shepherd published by The Macmillan Company, NY . My copy is the 1965 5th printing. You should be able to get a used copy on Amazon.


Thanks, Ron

[smithy1]

I have "most" of one these C20B compressor assemblies at home...

I still squirm at the thought of 3D printed compressor parts....I've seen ones made of the correct material come apart (with spectacular results), let alone ones made of ABS or nylon...I'm sorry but I believe there's no chance of ABS or nylon working at all...

Even sheet aluminium rotor parts would last 2 seconds, 99% of them are either cast, forged or billet, the others, like the Rolls-Royce C30 impeller are billet titanium...they need to be plenty strong.

Smithy.

[turboron]

If you are serious about this project you could 3D scan the C20B wheels and case/vanes to create files for 3D machining of billet copys. I would use a C20B tiebolt for the assembly.

Thanks, Turboron

[ddthj]

Unfortunately gtba requires paid membership...

I am fairly confident in the strength of printed materials, there are several videos online of printed parts running at 30k+rpm, not even ABS. Nylon would probably be best since it should flex a little more and be less prone to shattering. If a PLA part can run at 40k rpm just fine, nylon should have no issues at all.

youtu.be/logxPDmEUiI

I'm fairly sure this engine is just sustaining on the pressure of the butane (or maybe it actually is self sustaining?), but it at least demonstrates the parts can operate at those speeds even when made with weaker plastics.
Maybe it would be worth testing? It may be possible to made a F3D compressor from printed parts. Just throwing ideas at the wall to see what sticks haha. My printer is my most valuable tool so I'm just trying to see how much of an engine I could build with it.

Thanks again for your comments,
-Mike

[smithy1]

Had a look at the video...quite impressive actually and it was indeed running by itself, and well past self sustain, (which looked to be ~14-16k rpm area), a bit warm during the first run... but running by itself nonetheless...I would never have thought a plastic (PLA) compressor would be able to spin at 40k rpm+...I stand corrected..!


However, it is an F3D type compressor wheel and the blades are supported both front and back...it's not a "free blade" type wheel like the C20B or a centrifugal impeller like a turbocharger...then I think the plastic would indeed struggle...but kudos and well done to him just for getting it running.

Smithy.

[racket]

Hi Mike

LOL...............yep , GTBA isn't free , but we only gets what we pay for in life , no free lunches , if you want the best advice then the GTBA is the place to be .

The comp wheel in your video has "discs" either side of the wheel , discs are strong , unlike an axial comp blade that you want to construct , it has a very highly stressed root section , they're different animals .

The FD3 comp wheel was designed to be home built , thats why its shrouded for strength , remove the shroud and those comp blades won't last long .

Also 40,000 rpm is barely idling speed .

Some of the "plastic" industrial axial fans with fibre reinforcing can be run up to ~500 ft/sec tip speeds , so I'm not discounting "plastic" , there have been turbine engines with plastic parts , but be extremely careful , have good containment barriers , a chunk of plastic hitting your at those speeds can do damage :-(

You mentioned 220 mm as max dia , is this the engine casing size or the first stage axial wheel diameter , a low pressure axial comp will need a fairly large diameter combustor , we simply can't "scale down" combustors compared to full sized engines with their high compression ratios, so if 220 mm is the can diameter surrounding the flametube then your comp wheel/s might only be half that at ~110 mm ...............heh heh , isn't that the same size as an Allison C20 first stage .............lotsa airflow .

Cheers
John

[smithy1]

Mar 20, 2018 19:43:04 GMT -5 racket said:

Hi Mike

You mentioned 220 mm as max dia , is this the engine casing size or the first stage axial wheel diameter , a low pressure axial comp will need a fairly large diameter combustor , we simply can't "scale down" combustors compared to full sized engines with their high compression ratios, so if 220 mm is the can diameter surrounding the flametube then your comp wheel/s might only be half that at ~110 mm ...............heh heh , isn't that the same size as an Allison C20 first stage .............lotsa airflow .

Cheers
John

Just measured one John...~112mm dia on the first stage...actually all 6 stages are a similar outer diameter but the hub tapers out allowing compression until the air enters the centrifugal impeller...then the air really starts to get squeezed. I believe a total ratio of ~7:1...! Compared to the GT6041's ratio of ~3.8-ish.

Smithy.

[racket]

Hi Smithy

Pretty close :-) .....I think the C20's 6 axial stage only produces ~2.5-3:1 at most , the centrif stage with its very backswept blading multiplies that by maybe ~2.5:1

Cheers
John

[ddthj]

The max size cube I can print is 190x190mm, so provided that I print the outer case in halves and bolt them together I can actually get the outer diameter up to ~220mm, so yes the rotors themselves would have to be around 210mm in diameter if I decided to go with that size. The combustor would have to be all metal, and I could probably make it any size as needed. I think I'll have to scrap the idea of an axial compressor for now because I hadn't realized how much different their combustion chambers have to be, maybe someday

Until then I'll see about designing my own version of the FD3's compressor for 3D printing, would a larger diameter engine require less rpm to sustain? I remember hearing about that somewhere.

Thanks again,
-Mike