Turbo Restrictors and Air Cavitation....

[SL666]

i remember (i post on PF as GSRman) thats right people F33R ME! 4 i have met the zilla!

[myne]

Below is a compilation of random thoughts based soley on uneducated genius.

Would, could variable pitch turbo vanes help?

On a helicopter, the pitch of the trailing blade is steeper so that it can get more "traction" with less air to grab. Wouldnt it be nice to have some sort of vacuum advance on the turbine blades?
Obviously there are huge engineering limitations there, but what if the housing inlet itself had fins that directed air in the opposite direction to the turbine. Would that reduce the chances of compressor surge by increasing the energy imposed on the fins?
What I mean is, under normal circumstances the turbo is getting 1 bar of pressure. This pressure has a mass, this mass is more or less just going straight into the blades. There's a deflection force being applied to the air, etc.
When the air pressure is actually a vacuum, could you reduce overspinning by directing the air opposite? The cause of the force is different, but with appropriately angled fins, the effective force might be the same. How that really helps power I'm not sure. I just like thinking.

The only other idea I can think of is to treat the tract between the turbo and the restrictor as a plenum - complete with velocity stacks at both ends.
If you think of it like that, is it possible that each fin creates a similar acoustic resonance to an engine valve? Could there be ways to tune that plenum?
An obvious thought is to make the plenum huge. It'll take a fair bit longer for the air to drop in pressure, which could add valuable extra power between shifts. The shifts, and any throttle lifts would give the plenum time to recover.
On tight tracks, I'd expect this to yeild pretty decent results since you'd constantly be topping up the plenum before it runs out.
Pressurising the air going into the restrictor would be nice too. A big scoop facing forward would add a decent amount at higher speeds I imagine. Probably way too obvious but hey, it's worth mentioning.

So the summary of my ideas : velocity stacks, plenums, variable vane, and resonance tuning.

All normal NA stuff I know, and probably well known in turbo, but a lot of discussion I see seems to treat FI as some totally different beast - like the laws of physics change just because of a fan.

[oldcorollas]

one thing that people get confused about.... just before surge line, the turbo is moving AS MUCH air as it possibly can from the amount of air coming in the inlet... it's not pulling vacuum as such (vacuum is very very different... ie, you count the air molecules instead of psi ) it's more to do with the pressure difference across the blades.... ie... the compressor can only support so much "push" and when it surges, it is still pushing actual air.. and shitloads of it.... just that the air in the manifold is pushing back...

with a restrictor, this still happens... but the manifold pressure gets lower as rpm increases, BUT the the limitation (i think) is the inlet pressure going too low, even tho air mass is still getting thru... and so even tho you have a flx of air, the low pressure in the inlet combined with the high pressure in the outlet, pushes air back thru the blades... BETWEEN the blades.... i imagine that the blades are still actually grabbing and pushing air.. but there is a finite space between them for air to flow back...

same happens if you ran a turbo with a blocked exit.. or into blocked container.... at some point, the pressue will push backwards.. by reducing the imlet pressure, the point at which this "surge" occurs will be at a lower outlet pressure..... or something like that

fwiw, turbomolecular pumps work when there is already low vacuum present.... (and a rough pump pulling away the molecules that go thru it) since the blades can't sustain a high pressure differential.....

[mic*]

(Pretty much answered/dispelled by Stu whilst writing this)

Excuse the ignorance, but what is the purpose of a restrictor in front of the compressor?

I cant see how it is anything but "restrictive" in terms of the MAF allowable/capable before the compressor wheel "loses grip".

I would have thought that the "lose grip" point would be at a fairly constant vacuum level (know idea what value that may be) at the inlet of the compressor, more or less regarless of the overall pressure differential.

If it is a means of trading off for more bottom end, or somehow critically changing flow dynamics, can someone please explain the theory of it.

Ta.

[RA23]

Yep, that makes sense there. I only had to read it twice to get it to make sense.
Keep this thread going, its a good explanation for restrictors atm.

Btw, i do know its just the compressor "over spinning" but didnt really connect it with "pulling through". And yes, cavitation is the wrong word, but a lot of the workshops i talk to sort of look at you with a huh when you talk about the turbo blades going supersonic etc. They know the word cavitation and so i generally use it.

As Oldcorollas already said, cavitation only occurs in fluids, so it is the wrong term. But there is some more confusion: "over spinning" and blades going super sonic and super sonic restrictors....

The highest possible flow rate through a restrictor is when the airspeed in the throat is sonic (i.e. M=1.0), the airspeed that goes with this is sqrt (1.4*287.05*T), airspeed in m/s, T in Kelvin. You need polished surfaces and a very good bellmouth, also the angle behind the restrictor must be gentle (<5 deg) and no sudden changes, large radiuses everywhere.

If the compressor would go super sonic at the tips you will not get a bang, but rather a lot of noise (continuous) as the shockwaves that are produced are being produced all the time and not just when "breaking the sound barrier".

Also compressor surge and choke are wildly misused terms, the following site has some more background:
http://www.cs.purdue.edu/research/cs...gineering.html

Just found a better explanation:
http://www.turbodriven.com/en/turbof...compressor.asp

The conclusion is that this is quite a complex matter.....

[takai]

Excuse the ignorance, but what is the purpose of a restrictor in front of the compressor?

The purpose of a restrictor is exactly that, to restrict things, and in this case its horsepower output. A restrictor will go a long way to creating parity in a class (like IPRA) where there are many different cars with different power outputs. Or limit top end power in a class where there are large solid objects (trees)... like rally.

[myne]

same happens if you ran a turbo with a blocked exit.. or into blocked container.... at some point, the pressue will push backwards.. by reducing the imlet pressure, the point at which this "surge" occurs will be at a lower outlet pressure..... or something like that

Sounds similar to "vortex ring", a problem helicopters can suffer http://en.wikipedia.org/wiki/Vortex_ring

Or am I completely on the wrong track with what you're saying?

[oldcorollas]

ummm maybe ....


for trubo molecular pumps, they have a certain compresstion ratio for different gases before "back streaming" occurs.... which is effectively the same thing but opposite... follow?


a fan (like turbo, or turbo pump) seems to be limited by the pressure differential... or in other words, the compression ratio... between inlet and outlet. if you lower the inlet pressure, you lower the maximum outlet pressure... if you decrease outplet pressure, you also decrease minimum inlet pressur eneeded...

this works both in vacuum and pressure systems... which are effectively the same thing.. since pressure is relative regardless if it is 10^-8, or 10...

[tricky]

for trubo molecular pumps, they have a certain compresstion ratio for different gases before "back streaming" occurs.... which is effectively the same thing but opposite... follow?

OT, but have you ever seen one of these fail? It's pretty special... $15000 (for a small one) going down the drain in significantly less than a second... Well maybe it's more looking at the pump afterwards that's special.

Managing the 'overspinning' will mostly be trial and error. The maximum efficient speed of the turbine is dictated by flow characteristics (dictated in turn by restrictor profile, gas temperature, pressure due to localised variations). Norbie's suggestion of an optical shaft velocity sensor would work, but pressure sensors would probably allow you to run closer to the limit with more certainty, due to variables affecting the maximum efficient shaft speed. The trick would be to use pre/post compressor pressure sensors and get some good data, then model the wastegate function from this.

Of course, optimising the flow path and restrictor location is pretty important too, but it sounds like you have this covered... Gentle restrictor gradient and as far from the compressor as possible.

BTW, good discussion!

[oldcorollas]

but pressure sensors would probably allow you to run closer to the limit with more certainty, due to variables affecting the maximum efficient shaft speed. The trick would be to use pre/post compressor pressure sensors and get some good data, then model the wastegate function from this.

Motorola already make MPX series differential pressure sensors so you get a 0-5V output (i think) based on the difference in pressure... for this application.. you 'd need what? max pressure = 30psi? min pressure = -7? -10? say a 50psi differential sensor to be sure = 10psi per volt too...

search www.digikey.com for MPX (site was just down )

[kingmick]

dont they call it blade resession in a rotor blade? half the prob with choppers is the trailing blade on the verge of beaking the sound barrier which causes problems like the blade self destucting? hence the shape on the rotor tip to keep them under supersonic.
lol i have 10 hours in rotaries but forgotten most as its been a long time. used to sit captivated with my uncle teaching me about it! he has 23,000 hours in them, he is a ledgend.
now i keep reminding him why choppers fly," they vibrate so badly the earth rejects them" hahahahaha, cant beat a fixed wing for speed.
great topic keep it coming

[tricky]

Motorola already make MPX series differential pressure sensors so you get a 0-5V output (i think) based on the difference in pressure... for this application.. you 'd need what? max pressure = 30psi? min pressure = -7? -10? say a 50psi differential sensor to be sure = 10psi per volt too...

search www.digikey.com for MPX (site was just down )

Oooh! Very nice!

http://www.digikey.com/scripts/DkSea...S&Cat=34931716

There are a ridiculous number of sensors there. Yep, mostly 0-5V. It's a bit unfortunate that they don't quote resolution though.

[myne]

Could you use something like a BOV to re-route air from PAST the turbo to BEFORE it if the pressure differential got to a predetermined level?

Can a BOV work like that?
Would it actually do anything useful?

[brett_celicacoupe]

remember toyota's little intake restrictor stunt

http://homepage.virgin.net/shalco.com/tte_ban.htm

[mic*]

Could you use something like a BOV to re-route air from PAST the turbo to BEFORE it if the pressure differential got to a predetermined level?

Can a BOV work like that?
Would it actually do anything useful?

Interesting idea. Keeping in mind that it is the shaft speed you want to control, i would think that if you had the pressure differential/ratio gauged, twould be more precise to use the signal to bleed the wastegate and control things that way.

Thinkin about that further tho, without a restrictor inplace, i dont think a BOV could do anything but limit outlet pressure - ie inlet is at ATM pressure so can go no higher. With a restrictor inplace, there will be a pressure drop between it and the inlet of the compressor yeah? So if BOV air was plumbed between there, it could bring that airspace back to ATM (outlet would drop a bit too) -> reduce ratio -> spool more = more L/min to the motor or just more L/min goin round in a circle getting hotter???

Thoughts?