compressor maps.pdf

fast tech

/ TECH / COMPRESSOR MAPS /

fast tech

OFF THE MAP

BEFORE

STARTING…

The ﬁ rst rule of working on

cars and using tools of any

kind is don’t ever skimp on

decent protection. Goggles,

gloves, ear defenders,

masks and a set of overalls

should be in your garage.

Use them.

When using power tools,

protective gear is essential

— grinders and welders can

make a real mess of your

soft skin and bone if you get

it wrong.

Never work under a car

without supporting it using

axle stands. A car falling on

you is not something you’ll

be laughing about down

the pub.

How and why you should bone up on Compressor Maps before

choosing that expensive turbo for your big-power engine.

WHO IS STU?

A Level 5-trained fuel-injection

technician, in the past Ford nut Stu’s

worked for a Ford RS dealer, a well-

known fuel-injection specialist and

various tuning companies. Then six

years ago, he joined forces with Kenny

Walker and opened up Motorsport

Developments near Blackpool,

specialising in engine management

live remapping, as well as developing

a range of Evolution chips which are

now sold all over the world.

He’s also jointly responsible with

Webmaster, Petrucci for www.

passionford.com. Started in 2003, it’s

grown rapidly from a few friends

contributing, to one of the biggest

Ford communities on the web.

Stu’s enviable knowledge of the

workings of modern-day Ford

performance engines means

that he’s just the man to explain

how and why things work, and

most importantly, how they can

be improved!

Having worked as a tuner for over

16 years, Stewart ‘Stu’ Sanderson

is one of the most respected

names in the business.

Deciding on the spec of your engine — whether it be a blown CVH, Zetec or YB like above — is relatively

easy compared to choosing the right-spec turbo, as there are so many variables to consider

turbo, or indeed entire engine, will

be used for the ﬁ rst time and no

airﬂ ow data is available.

For the record, another method

of airﬂ ow measurement is also in

use today, this alternative method

being the ‘volumetric ﬂ ow’. This

method of calculation has its results

expressed as cubic feet per minute

(cfm). If you wish to convert this

volumetric ﬂ ow rate back to mass

airﬂ ow, you need to simply multiply

it by the air density. Air density at

sea level is 0.076 lb/ft.

A golden rule always worth

remembering, is that a reasonable

average consumption ﬁ gure for an

engine — using the mass airﬂ ow

rating of lb/min — is 1 lb of air will

be consumed for every 9.5-10.5 bhp

produced by the engine. So we can

safely calculate our estimates using

the ﬁ gure of 1 bhp = 10 lb of airﬂ ow

and normally be within 10 per cent

of accurate.

So, if we take an engine from the

standard 4wd Sierra Cosworth, we

can see it consumes approximately

22.4 lb of air every minute while it is

ﬂ at out, producing its maximum

power of 224 bhp.

Remember that these calculations

are subject to air temperature

variations since our air changes

density if it is heated or cooled. The

colder the air, the denser it becomes

and the more oxygen is available.

This temperature variation has an

obvious effect on engine power

output, too.

Back to our new engine. We have

built it to make 440 bhp and we

how a large turbo

makes more power than a small

one with the same boost pressure.

Plus where that power comes from

and how it affects the volumetric

efﬁ ciency (VE) of the engine.

I’m sure that by the end of the

article many of you ended up

thinking, ‘Why don’t we just ﬁ t the

biggest possible turbo and reap the

rewards?” That’s what we’re going

to look at this month: how to

choose the correct-size turbo for

your engine and avoid the common

mismatches that can lead, at best,

to a poor throttle response and

driving experience, and at worst,

terminal engine or turbo damage.

The ﬁ rst thing we must do is work

out what kind of airﬂ ow we’ll need the

turbo to process in order to achieve

our target horsepower ﬁ gure.

Let’s pretend we are building

ourselves a 440 bhp Cosworth YB

lump. We have already chosen the

intake manifolds, compression ratio,

headwork and cams to ﬂ ow the

required amount of air, so all we

need to do now is ﬁ nd a turbo

capable of supplying it without

exceeding its design parameters

and becoming unreliable. The way

to do this, as is quite often the case,

is with a little maths.

related to the amount of bhp the

engine develops. The VE of an engine

is almost always highest at the point

in its rpm range that it makes its

peak torque ﬁ gure. Above this rpm,

pumping efﬁ ciency per stroke starts

to fall away but work done by the

engine as a whole is still increasing

due to the ever-increasing speed it is

making torque at.

Generally, engine airﬂ ow is

measured using the ‘mass airﬂ ow’

principle and this method of

measurement usually has its results

stated as pounds of air per minute

(lb/min). It is both normal and

acceptable to simply estimate your

airﬂ ow requirements for a new

engine installation where the new

Words: Stewart Sanderson

Photos: Jon Hill, Honeywell

Turbo Technologies, Ford

ENGINE AIRFLOW

Engine airﬂ ow can be directly

0122 JULY 2006 FAST FORD

FAST FORD JULY 2006

0123

LAST month we looked at

fast tech

/ TECH / COMPRESSOR MAPS /

fast tech

we will have a compressor

efﬁ ciency of 74 per cent, meaning

we only lose 26 per cent of our

turbo’s effort to heat. This is good.

the low consumption area, right

into surge. For a more detailed

explanation of surge, have a look

at the boxout.

Surge will stop only when the

compressor speed slows enough

to reduce the boost pressure and

move the operating point back into

an area of stability. This situation is

commonly addressed by using

dump valves to deal with the closed

throttle issue. These valves vent

intake pressure to atmosphere so

that the mass ﬂ ow ramps down

smoothly, keeping the compressor

out of surge. In the case of a

recalculating bypass valve, the

airﬂ ow is recalculated back to the

compressor inlet.

COMPRESSOR

MAX FLOW

The maximum ﬂ ow capability of any

compressor is shown on the map.

Look for the point where the

compressor wheel speed reaches

maximum and passes through the

least efﬁ cient island. The horizontal

reading here is the maximum that the

compressor can ﬂ ow.

So, we already have our target

airﬂ ow ﬁ gure, and we have our

compressor map and are now

equipped to read it. Now what?

We now need to look at the map,

and see where we will be on it

when the engine is consuming 44

lb/min of air.

Sadly, a quick look on that T3

60 trim compressor map shows

us that the peak volume of air

available from that turbo is around

the 30 lb/min of air mark, so little

use at all for anything over 300 bhp,

and it’s not much good at that

either, showing a compressor

efﬁ ciency of 65 per cent at best. So

we need to look a little further a

ﬁ eld... Let’s try the newer style of

turbochargers and examine a

GT3076R compressor map.

We can see from this compressor

map that we can shift our peak of

44 lb/min of air with this compressor

at as little as 0.75 bar of boost

(1.75 PR).

Although the efﬁ ciency isn’t at

its best at this low pressure, it’s

unlikely our engine will make 440

bhp at such a low boost pressure

anyway, so this could well be a good

compromise turbo choice for us if it

doesn’t surge or cause any other

issues with our engine. But how do

we know what boost pressure is

required to make the power on our

target engine, so we can plot the

compressor map properly?

SPEED LINES

Crossing the circular efﬁ ciency

islands are the turbine speed lines.

The numbers next to these lines

represent the approximate turbine

speed required to make the

compressor generate that

particular airﬂ ow. Hopefully, this

is self explanatory.

COMPRESSOR

MAX PRESSURE

If you take a look at the pressure

column, and ﬁ nd the pressure that

corresponds with the highest point

on the efﬁ ciency islands, this is

the maximum pressure that the

compressor can produce. 2.9 PR

at sea level is 1.9 bar of boost, for

example. The compressor’s

maximum pressure is governed

by the compressor wheel’s

rotational speed.

It’s worth noting that the pressure

the engine actually receives is

further inﬂ uenced by various things

such as intercooler efﬁ ciency and

air temperature, so the pressure

read at the engine itself may be

somewhat lower than is generated at

the compressor outlet of the turbo-

charger. Always bear that in mind.

SURGE LINE

To the far left of our chart is the

surge line. Operating to the left of

this line represents a region of

severe ﬂ ow instability. Surge is

recognized by the driver as anything

from a small ﬂ uttering noise to a

wildly ﬂ uctuating boost with a harsh

chattering noise from the turbo.

Continued operation within this

region can lead to premature turbo

failure due to huge thrust loading.

Surge is most commonly

experienced when one of two

situations exist. The ﬁ rst and most

damaging is surge under load. It

can be an indication that your

compressor is too large and is often

experienced at low rpm where the

engine’s air consumption is still low.

Surge is also experienced

when the throttle is quickly closed

after boosting. A lot of people

incorrectly refer to the sound as

‘wastegate chatter’ — in fact it’s the

compressor going into surge and

stalling. This occurs because the air

consumption by the engine is

drastically reduced to almost nil as

the throttle is closed, but the turbo

is still spinning and generating the

airﬂ ow. This immediately drives the

operating point to the far left of

the compressor map,

All turbos come with a compressor map like this. They compare

the compressor’s rotational speed to its output airﬂ ow

Here’s the map from a T3. It shows that the peak volume of air

available is 30 lb/min — little use for anything over 300 bhp

CHOKE AREA

The area to the right of the outer

most elliptical circle is the least-

efﬁ cient area of a compressor —

known as the choke area. When the

compressor reaches a certain rpm,

the air moved by the compressor

wheel in the diffuser area of the

compressor housing is moving at or

past the speed of sound. When the

air speed reaches sonic speed, the

amount of airﬂ ow increase is very

small, as compressor wheel rpm

increases as the compressor has

effectively now reached its limit.

The choke area is rarely noted

on a compressor map, but can

usually be found by dropping a

vertical line down from where the

fastest wheel speed curve ends on

the right hand side of the map. This

vertical line is the approximate max

airﬂ ow the compressor is capable

of, regardless of efﬁ ciency or

pressure ratio.

now know, using the calculation

above, that it is consuming

approximately 44 lb/min of air. How

do we ﬁ nd out what turbo we need

to purchase and ﬁ t using this info?

Let’s take a look at the

compressor map (above right) from

a Garrett T3 60 trim compressor, as

ﬁ tted to the 4wd Sierra Cosworth,

and see how all the information on

the map is interpreted.

Immediately we can see that it

has two axes:

The vertical axis is what’s

known as a pressure ratio axis.

The pressure ratio is the air

pressure out of the turbo plus the

atmospheric pressure. Translated,

this indicates that a pressure ratio

of 1.5 means the air leaving the

turbocharger

has 1.5

times

the

pressure it entered with.

As you can see, the pressure ratio

depends totally on the ambient air

pressure. For example, at sea level,

if a turbo boosts at 14.7 psi, the

ambient pressure is also 14.7 psi so

that’s 2.00 on the compressor map.

Take that turbo to a higher elevation

where the ambient pressure is less

than 14.7 psi and still have the turbo

boosting at 14.7 psi and you will ﬁ nd

the pressure ratio would be higher

(this method of working is called

Absolute Pressure).

Incidentally, turbo efﬁ ciency

almost always decreases as the

elevation increases (the pressure

ratio increases). In other words,

turbochargers lose performance

and become less efﬁ cient as

elevation gets higher.

The horizontal axis is the

airﬂ ow axis. This shows us airﬂ ow

out of the compressor in pounds of

air per minute. We can see by

following these lines vertically, that

we move through little islands with

percentage ratios tagged to them.

NEW ’CHARGER

COMPRESSOR

MAPS

All turbochargers have a chart

known as a compressor map

available for them. These are small

charts with a mine of incredibly

useful information within them. They

relate the compressor’s rotational

speed to its output airﬂ ow, and also

its actual efﬁ ciency. Choosing a

turbocharger without these

charts is impossible. We

know already from our

previous estimations that we

want a turbocharger that

can ﬂ ow 44 lb of air

per minute in a

reliable

fashion

to give

our

EFFICIENCY

ISLANDS

We now have the chart content

itself, and this content is presented

to us as a series of elliptical circles.

These circles are known as

Efﬁ ciency Islands and relate to the

turbo’s Adiabatic Efﬁ ciency (see

boxout). If we look at the island in

the centre, this has a rating of 74

per cent. This means, that if we

keep our airﬂ ow within this island,

WHAT’S ADIABATIC

EFFICIENCY?

When a compressor compresses the air, it heats it up. This is a fact with all

air compressors, even the simple bicycle tyre pump — remember how it

used to burn your hands if you pumped too fast? Adiabatic

efﬁ ciency is a measure of just how much we heat

the air when we compress it. The more

efﬁ cient the compressor, the

less heat is created by the

compression process.

WHAT’S COMPRESSOR SURGE?

When a compressor is processing air and our engine is consuming it, all is

ﬁ ne. We have torque on our turbine wheel generated by the airﬂ ow from

the compressor in the ﬁ rst stage. However, when we generate more

airﬂ ow than we can consume, things begin to get messy. The air pressure

is backing up at the compressor, slowing it somewhat and the resistance

at the compressor wheel starts to exceed the energy available at the

turbine wheel, so our compressor starts to stall (a rapid decrease in

rotational speed).

This puts the turbo into a vicious circle of events, as the stalling

compressor drops our airﬂ ow back into an area that the engine can

consume, and the turbine starts once again to spin back up, until it hits the

ﬂ ow limit and again starts to stall... so we have a surge of sudden speed

and boost, followed again by a sudden stall. This is repeated many times a

second and is known as compressor surge. This isn’t the only form —

a full technical feature could be written on surge alone, but this basic

description serves our purpose for this month.

target

bhp

ﬁ gure,

and of course we need it

to easily ﬂ ow less than this for all

other engine operational modes

below wide open throttle.

So, how do we read a compressor

map and determine that a

turbocharger is indeed suitable for

our needs?

This is a brand-new GT3076R

turbo. It’s shown minus the

turbine housing as this is how

they’re supplied by Garrett

Thanks to the large turbine

housing needed, the T4-shod

RS500 was a nightmare to

drive on the road, with little

performance below 4000 rpm

0124 JULY 2006 FAST FORD

FAST FORD JULY 2006

0125

fast tech / TECH / COMPRESSOR MAPS /

really making more like 100 bhp.

That means, if we take our

average consumption figure of

1 lb of air for each 10 bhp, we

can also assume we will be

consuming 10 lb/min of air at

3000 rpm.

This means though, that if we

only make 100 bhp at 3000 rpm,

with 0.75 bar of boost pressure,

we will actually place ourselves

to the far left of the map and into

the surge area of the GT3076R,

where as on our T3, we were

operating efficiently with no

problems at all. This surge is very

dangerous to both the turbo and

the engine and must be avoided

at all costs.

This is where a change of

turbine housing can come in very

useful. Increasing the turbine

housing’s Area Radius will result

in less turbine speed for any given

gasflow, thus slowing down our

compressor and hopefully putting

us into a more efficient area of

the compressor map when our

boost is made at a slightly more

efficient engine speed.

A good example on this

compressor map is 55,000 rpm

and only 0.35 bar of boost — this

will just about keep us on the

efficiency map and hopefully

solve the surge issue we had with

higher turbine speeds and more

consequential boost pressure, but

of course there is a trade off. Our

engine response will have

suffered somewhat and feel a

little less responsive due to the

slower boost climb. Naturally

though, a slight lack of response

is far more acceptable than

deadly compressor surge.

In a nutshell, we need to

customize the turbo’s response

range to keep the compressor

away from its surge area until our

engine gets to a speed where it

can consume enough air for a

compressor of that size and keep

it away from the choke area once

our engine is consuming lots of air.

The GT3076R compressor map shows us that it can supply the

required 44 lb/min easily

PROBLEM AREA

This very tuning is the part of turbo

selection that causes us the most

problems, as we almost always

hurt the turbo’s response time by

increasing the rear housing size.

But it’s a necessity with large

turbochargers and high power

outputs if we want to avoid the

dreaded surge (turbine inlet

pressure problems aside — see

last month).

Anyone who’s driven both the

Garrett T3-equipped Sierra

Cosworth as well as the later

Garrett T4-equipped RS500 will

know what that larger turbocharger

feels like on the road — pretty

damn awful in most respects.

But without the large turbine

housing, the T4 compressor would

run almost immediately into surge,

as the standard 2-litre YB engine

simply cannot consume enough

air in its lower operating range to

support a compressor of that

immense size. So the turbine

housing had to be sized ccordingly

to keep the turbine and ompressor

speeds down to a sensible speed

until the engine was capable of

consuming the volume of air from

the compressor.

The result was virtually no

performance below 4000 rpm,

but once that compressor span up

to speed, all hell broke loose and

the top end performance of the

motor with its large compressor

and big turbine housing was far

sweeter revving than its little T3

sister that was running out of

puff around the same time this

monster got into its stride.

There you go, you can’t have

your cake and eat it just yet,

although, not too far away are

turbochargers with large

compressors and a turbine housing

that adjusts itself according to how

much airﬂ ow we want to ﬂ ow

through it at the time, called

variable geometry turbochargers.

At the moment, the mechanism

isn’t quite capable of dealing with

the intense heat we experience at

huge power levels, but it’s working

ﬁ ne on thousands of turbo diesels

and I’m conﬁ dent they will be with

us on big-power petrol Fords very

soon indeed.

Sadly, without access to some

very serious equipment, we can’t

easily, so it will be an exercise in

trial and error. All you can really

do as enthusiasts is undertake

some research and see how much

power a comparable engine spec

makes on an engine dyno at the

points you’re interested in and

work from there.

So, although we know the

GT3076R is almost certainly going

to provide our 44 lb/min of air with

ease, we still don’t know if it will

surge at any point. How do we

estimate this?

some educated guesswork if we are

to avoid surge.

Here’s an example: if our

engine, at wide open throttle,

actually moved enough air at

3000 rpm to spin our turbine and

compressor up to 100,000 rpm

and as a result, we generated the

expected 0.75 bar of boost, and

the compressor tried to start

processing 44 lb of air per min,

we would have to be generating

440 bhp at 3000 rpm to consume

it. Clearly this won’t be the case,

so what will actually happen?

To answer this we need to

know how much air we will really

be consuming at that sort of

plenum pressure and engine

rpm... So, let’s presume we are

WHAT RPM?

Well, we know that our chosen

turbo requires approximately

100,000 rpm to ﬂ ow 44 lb/min of

air with approx 60-65 per cent

efﬁ ciency at 0.75 bar of boost. We

can also see from the compressor

map that if we actually need 2 bar

of boost to make our 440 bhp, the

turbo will be spinning at over

120,000 rpm yet actually be

working more efﬁ ciently than if it

were generating only 0.75 bar. So

how do we now ﬁ gure out what

rpm the turbo will be rotating at on

our engine? Is it not best to have it

spinning up at maximum speed as

soon as possible? Can that create

further headaches?

Well, this is where the problems

really do begin, as it is far more

complex to plot how much air the

engine will use at a given rpm point

before it reaches peak power and

we really do need to start using

The future: this is the adjustable turbine housing from a variable-

geometry turbo as found on Ford’s 2-litre TDCi engine. These

turbos can’t cope with huge power levels, but it won’t be long

NEXT MONTH

Fuel pressure: why it’s

important, and why turning

it up is sometimes worse

than turning it down?

0126

JULY 2006 FAST FORD

fast tech

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: