Flow vs psi?

[Lwilliams]

Not sure on where to post this, as most diesels have turbos I thought here was the best place...

Something I think about now and again I can't understand it 100% hoping some one could help?

The reason why I am asking is people say check the flow of the turbo, you need a good flow etc...

Please explain if my thinking is not accurate etc.. but surely the best setup would be to aim to get optimum efficiency out of the turbo with the highest psi you could as this will force air in to the cylinder quicker and more compact?

Would the above be the best way to match a compressor to an engine? obviously you use a compressor map to estimate too?

I understand there is a lot more than just this to factor in, so could some one explain?

[Ruan]

This is a pretty confusing topic - you'll hear people harp on about making 140hp whilst using the original Garrett T2 making 32psi boost.Yet for example my GT2052V made 175hp at 20psi of boost....

Yet you'll be shocked to hear when I also say - more pressure at the inlet will make more power...

HOWEVER - that's in an ideal world - unfortunately turbochargers are not efficient - efficiency is the key to unlocking the turbos potential - don't utilise the turbo hard enough, efficiency drops, run the turbo too hard, efficiency drops and you'll overspeed and destroy the thing - there's a "sweet spot" for every compressor. This is where your compressor map comes in - this is what shows you how much air you can move, at what shaft speed, at what pressure ratio and what efficiency the compressor would be working at.



Here's a compressor map for a GT4508 - stonking great turbo. First thing you'll probably think "where's the boost axis" - well you'll probably see something called "pressure ratio" - essentially this is boost, however as you can see, it starts at a value of 1 - in this case we'll take 1 to be barometric pressure (around 14.7psi, you've probably noticed if you work in new money that's bang on 1bar) - this takes into account what happens when you go up the side of a mountain or "compound" charge the engine - whole other kettle of fish.

Along the bottom - you have airflow - at this point everyone is probably shouting "see, flow isn't the same as pressure" - well, unfortunately it is, since your engine isn't changing in size, the airflow is a constant at a given engine speed. Next thing to take into account is the curved lines with numbers at the end of them. These are the various different shaft speeds that the compressor would be spinning at for that given pressure and airflow. Finally there's the big circles or "islands" - these are the "efficiency islands" that the compressor would run at - the closer to the central island, the more efficiently the turbo is running.

These "efficiency islands" are key to understanding what's going on here - notice on this turbo, your peak efficiency is about 78%!! Not great at all, and if you look at the compressor map for a GT1549:



Anything past 1.6 bar of boost and you're off the scale!!! The turbo shaft speed is off the chart, too fast for the poor thing! Efficiency through the floor etc - you can see here now that it's very easy to run a turbo RIGHT out of it's efficiency and totally overspeed it.

TL;DR: Essentially my point is that no, it's quite complex to understand what turbo will be best for an application, high pressure is good, but not at the expense of poor efficiency and overspeeding. It's all to do with efficiency. If you're running the turbo outside of it's efficiency range, you're generating a lot of heat rather than moving air - heat is bad, it reduces your air density and therefore the amount of oxygen atoms you have to burn your fuel with.

I keep saying I'll do a big post on understanding compressor maps fully one day - maybe this will push me to do it.

[Dum-Dum]

This is all I took from that.

[bashbarnard]

Thats genius. Where on earth can you get comppresor maps? Then your able to compare. I think a thread should be put up about this. I have what I would call a fairly good understanding of this but not as good as ruan.

[Dave]

No one has an understanding as good as Ruan

Search for compressor maps, at some point the manufacturer would have made them for customers to pick the appropriate size.

[bashbarnard]

Fair enough ill have to take a look when I get chance. Always willing to learn some more especially as it will benifit me at some point.

---

His like turbo yoda from mcm

[Lwilliams]

Thanks ruan was hoping you'd comment...

I understand most of that, but (please tell me where I am going wrong here) when I think about it as long as your turbo is in its efficiency range and the turbo flows enough for the engine. The higher the psi the better (within the island)?

The other thing that confuses me is on the graph for the gt 15.. how does it flow the same at 2.2 and 1.8? surely it should flow more at 2.2? If not why would you run it at 2.2, when you can save some of the turbo life?

And is the flow a positive correlation between that and the psi? as when you fulfil what the engine requires (e.g the cylinder) and there is a restriction, only then will the psi go up. Obviously the more you can flow the higher the psi can get to (Given its the same engine)? or am I talking out of my arse?

[Eeyore]

Someone needs to sit down with ruan and document all of his knowledge otherwise when he dies we will lose all of the knowledge.

[bashbarnard]

Someone needs to sit down with ruan and document all of his knowledge otherwise when he dies we will lose all of the knowledge.

This the man knows everything . what happens if he get Ebola.

[Eeyore]

Someone needs to sit down with ruan and document all of his knowledge otherwise when he dies we will lose all of the knowledge.

This the man knows everything . what happens if he get Ebola.

Or if he get hostage by petrol owners in order to hold the diesel tuning world at ransom D:

[ADV_93]

My brains hurts

[Ruan]

Engine air consumption in CFM is a formula.

CFM = ( L * RPM * VE * PR) / 5660

Cubic Feet per Minute = ( Engine size in litres * Engine speed * Volumetric Efficiency * Pressure Ratio) / 5660 (constant to correct to CFM).

This is how you calculate the increase in airflow since you're multiplying by the pressure ratio So say on your 1.9 litre engine you want to run 2.0bar boost - plug that into the equation...

1.9 * 4200 * 85 * 3.0 = 2034900 / 5660 = 359.522 CFM = (roughly) 25.1 lb/min

Say you want to calculate that for 1.0bar boost -

1.9 * 4200 * 85 * 2.0 = 239.68 CFM = (roughly) 16.7lb/min

Yes - more boost equals more air... However look at that graph - if you ran a GT15 at 2.0bar, it'd be moving 25.1 lb/min @ 2.0bar -



So that's a fail - I also know for a fact that that top shaft speed line corresponds to about 180,500rpm - so you can imagine the sorts of speeds you'd be talking about making a GT15 try and run at that.

However it will turn your turbo into a TOTAL hairdryer and be completely pointless and provide absolutely no more power - only succeeding in damaging components.

I understand most of that, but (please tell me where I am going wrong here) when I think about it as long as your turbo is in its efficiency range and the turbo flows enough for the engine. The higher the psi the better (within the island)?

Yep - absolutely.

The other thing that confuses me is on the graph for the gt 15.. how does it flow the same at 2.2 and 1.8? surely it should flow more at 2.2? If not why would you run it at 2.2, when you can save some of the turbo life?

Airflow increases with increases of pressure with the same airflow demand. All things being equal - you can't move the same amount of air with a lower pressure unless you increased your volumetric efficiency - and once you're to 100% VE, you're moving as much air as the engine can move, unless you make the engine bigger! Your only way forwards is more pressure. You'll notice that the shaft speed would be lower for the lower pressure at the same airflow rate, that's what happens when you have a smaller engine

And is the flow a positive correlation between that and the psi? as when you fulfil what the engine requires (e.g the cylinder) and there is a restriction, only then will the psi go up. Obviously the more you can flow the higher the psi can get to (Given its the same engine)? or am I talking out of my arse?

Absolutely - regardless of the turbocharger, if you keep applying more pressure (at the same temperature/density, i.e. not from a turbo running at 1000000*C) to the inlet, your engine moves more air which equals more power.

So basically to all those who say "There's no replacement for displacement" - well judging by the CFM calculator, there's quite a nice replacement, positive inlet pressure

[Lwilliams]

Thanks Ruan the amount of knowledge you have and how you deliver it is amazing repped

[Ruan]

There is one problem with all of this - we're working off the standard density of air here. We're assuming the air into the turbocharger is an average temperature and an average humidity - however this is almost never the case, turbochargers are sucking red hot air most of the time! You also could be up the side of a mountain or at sea level - there's SO many factors to consider.

This is why Garrett measure in lb/min airflow which is a *MASS* airflow not in CFM which is *VOLUMETRIC* airflow. Converting between the two is not an easy process, lots of variables. That's why I say "roughly" x lb/min.

You could read for hours and hours on the technicalities behind all this. But I think what I'm trying to say is, before you instantly wang on a turbo, consider the air feed. It has to be as cool and dense as possible! If the high density wasn't there in the first place, you're pissing in the wind trying to make more power with a pissy little turbo feed.

You also have to consider that the compressor map is starting at a PR of 1 - 1 being the inlet pressure of the turbo - the outlet pressure is a ratio of the inlet to outlet pressure. This means that you could have a massive restriction on the inlet to the turbo, you'll never make much power since the turbo is still working like it always does - the less restriction you have going into the turbo, the better - it's a fairly critical part IMHO.

Sorry for my essays - I could talk all day about this as I'm sure you can tell.

Dave will vouch for me doing this in real life also.

[bashbarnard]

Go ahead man talk all day. There is a time and a place to be a car bore this forum is one im learning new things from it and im sure others are. Im in the trade and im always learning new stuff

[Dave]

There is one problem with all of this - we're working off the standard density of air here. We're assuming the air into the turbocharger is an average temperature and an average humidity - however this is almost never the case, turbochargers are sucking red hot air most of the time! You also could be up the side of a mountain or at sea level - there's SO many factors to consider.

This is why Garrett measure in lb/min airflow which is a *MASS* airflow not in CFM which is *VOLUMETRIC* airflow. Converting between the two is not an easy process, lots of variables. That's why I say "roughly" x lb/min.

You could read for hours and hours on the technicalities behind all this. But I think what I'm trying to say is, before you instantly wang on a turbo, consider the air feed. It has to be as cool and dense as possible! If the high density wasn't there in the first place, you're pissing in the wind trying to make more power with a pissy little turbo feed.

You also have to consider that the compressor map is starting at a PR of 1 - 1 being the inlet pressure of the turbo - the outlet pressure is a ratio of the inlet to outlet pressure. This means that you could have a massive restriction on the inlet to the turbo, you'll never make much power since the turbo is still working like it always does - the less restriction you have going into the turbo, the better - it's a fairly critical part IMHO.

Sorry for my essays - I could talk all day about this as I'm sure you can tell.

Dave will vouch for me doing this in real life also.

Many conversations/discussions/fistycuffs had at the pub about pre-det and injection timing/duration/compressors and general diesel bible spewing haha

[Ruan]

I think there's been good week long periods where we won't speak to each other on the basis one thinks the other is wrong and won't back down

"What do you mean increasing advance won't increase PCPs!?! Are you f*cking insane?!"

[Rippthrough]

Red hot air can make your boost gauge peg silly figures but it never makes many ponies

[Dave]

I think there's been good week long periods where we won't speak to each other on the basis one thinks the other is wrong and won't back down

"What do you mean increasing advance won't increase PCPs!?! Are you f*cking insane?!"

Hahahaha ok ok .....


#notevenpartofthesmashedxudclub

[ekjdm14]

Ooh bit of thread necromancy FTW today methinks... Why this isn't slathered across the sticky bit at the top I don't know, because if folk want to tune their DT and not try to learn about turbos then they may as well just keep a stock of K14's and keep running 30psi through to 5k till the turbine pops... (Hint-: with a big Xantia FMIC and cool-ish air intake, the stock K14 will shit it's guts out the downpipe after approximately 2.5 pulls at 25-27psi not even going over 4500rpm!)

I've looked briefly over compressor maps years ago when I had my Supra but never really took the time to understand the meaning of them since the standard CT26 could comfortably run 14psi and that meant about 340bhp which was fine by me at the time... Now though, having oversped one royally and had a nice little warning (the type without oil ingestion and broken rods), it's definitely time to look into the whys and wherefores of turbos...

Currently thinking a GT2052V is going to be the answer, I could've guessed from other big power projects but having looked, it's probably the closest to not whizzing it's guts out while still making decent numbers & it should be versatile being a VNT too. (currently going for more top end power than torque as it's a completely stock engine, but the map suggests it should be able to make good low end torque if needed too).

Sorry, rambled on long enough now but at least this thread is back at the top again...