Boost level and torque curve

[Kinkstaah]

Tuning the wastegate to do it. That is all. Most people want the boost to not fall off like the most recent example. Those also look like dyno runs with an Auto/Torque converter setup, which does fun things to the graph.

The boost tapers down like that because the turbo cannot supply the same amount of air at 7000rpm that it can at 3000 in terms of PSI. That, or the tuner has decided that it tapering off like that is what someone chose to do.

IF you have a wastegate that can't bleed enough air to slow the turbine, and IF that turbo can flow enough air to feed the engine at high RPM, you get 'boost creep' which is a rise of boost pressure beyond what you are capable of controlling and/or want.

None of these show symptoms of that, but if you had a run that was 20psi at 3000rpm, and 27psi at 7000rpm, it could be an example of that. Or simply that the person wanted boost later for their own reasons...

The dyno graphs don't always show the full context.

[Dose Pipe Sutututu]

You can manipulate the torque delivery by ramping in boost gently, then throwing it all in after peak torque to keep the torque flat.

It's nothing magical.

[Duncan]

right, but fundamentally, for a given mechanical setup, you are either using all the torque (and therefore power) it will give, or you are choosing to run it less efficiently.

Many tuners will have a practice of identifying peak available torque and then winding it back a couple of % for safety, but unless you are working around a very specific issue like a weak gearbox, there is nothing to be gained by making 20 or 30% less than the engine can

[thx78]

therefore on the first examples, as we see, changing cams (graph 2) influences the quantity of torque at high revs its OK for me. so a tuner can act on the wastegate via the boost controller to increase the boost at high revs? on the last example, the boost does not decrease ok, but the torque does. this can come from cams etc etc ok. but on the other curves the boost is not constant, it increases, this is what I find strange to my mind. even more so if it comes from the relief valve. sorry I'm very new don't blame me. in my mind I couldn't imagine how the boost could be higher after the spool

 

[Lithium]

There are a few variables here, some are relevant but not critical (IMHO) to help answer your question.

The two major things:

1) Ignoring anything to do with forced induction - all engines have their own natural torque curve, and it will ALWAYS roll over higher in the rpm.  There is a fixed relationship between power and torque.  When dealing with kw and nm, the relationship between them is roughly:

kw = (rpm * nm) / 9549

nm = (kw / rpm) * 9549

The peak torque of an engine (without boost) will typically climb until somewhere nearish the middle of it's operating rev range, give or take a bit - then start dropping again.   The nearer the minimum and maximum rpm of the engine the steeper that drop off tends to be.

2) Boost simply increases the density of the air going into the engine, which inflates the torque at that point.  The ramp up in the torque curve you see on a turbo engine is due to the boost rising, but it's essentially just multiplying the torque you'd see if it was naturally aspirated.  The roll over you see at the end will typically be what would have always happened with the engine, whether it was naturally aspirated or turbocharged.   If the torque never started dropping then power would climb infinitely.

The cool thing about this is you absolutely can tune the power delivery to suit the needs of the owner and/or the limitations of the car, and I regularly do this.    With modern turbos we've got to the point where a setup that someone may run well over 20psi of boost with could actually reach target boost well under 4000rpm if the tuner/owner WANTED to - and a lot of people seem to do this when there is actually no realistic benefit, generally it just adds a massive amount of strain to the engine and drivetrain and often actually makes the car harder to drive.

As a general rule I tend to tune the boost curves for cars I tune to reach a "useful" torque level through the rev range and will often end up with a curve that ramps hard to a point, then creeps for the rest of the rev range - not to make the boost curve "soft" as such, but more to make sure its neither laggy nor pointlessly violent in it's delivery.   There have been cars I've tuned to be almost like a centrifugal supercharger (or naturally-aspirated-ish) where they actually only hit like 8psi of boost before opening the gate, then ramp up the next 10psi over the rev range... if the car is "loose enough" to drive.

On the flip side I've tuned a car that had stock cams and the engine's natural torque curve fell over HARD in the higher rpm and resulted in a slightly awkward power curve to work with, in that case I actually started ramping up boost to boost torque in a way to offset the engines "NA" torque drop off... at peak rpm actually running a good 5psi+ more boost that what the "flat curve" would have defined.  This gave the owner an extra 500rpm or so of useable rev range, and had a fairly solid impact on times he was running at motorsport events due to being able to hold gears a bit longer and also falling into a more useful part of the rev range in the following gears.

Here's an example of an RB in a GTSt I've done the "softened" boost curve to not pointlessly ramp straight to the max boost target early in the rpm, but still made sure it builds useful boost.  If you went in the car you'd not guess at all that the boost curve was doing anything "weird", it feels like it spools immediately and accelerates relentlessly (traction dependent) and holds to max rpm.   I don't know if you'd guess what the boost curve was doing by driving the car, or even looking at the dyno plot... but imho it suits the combination.

 

[thx78]

so on the first graph the boost and the torque during spooling at 4000 are deliberately limited? and then there is a progressive increase in boost with the revs

[Lithium]

Here's the torque curves from the car I ramped boost up later in the rpm to allow a slightly wider useful power curve - the power curve is a bit weird shaped also thanks to the TVIS (or whatever they call it with the 4EFTE in this Starlet) which changes the volume of the intake manifold throughout the rpm range, but you can see that the green power curve actually holds later on with the extra boost... but looks almost more like the kind of thing you'd expect from a cam or exhaust change

[Lithium]

3 minutes ago, thx78 said:

so on the first graph the boost and the torque during spooling at 4000 are deliberately limited? and then there is a progressive increase in boost with the revs

That's torque and power, it's all from a single run.  The boost curve is "held back" from it's peak target in the 3500rpm to 5000rpm range from memory, so it ramps hard to something like 18psi then climbs more progressively to 23psi nearer 5000rpm.   It makes the torque (and power) ramp more "natural" and less hard on parts and traction, it doesn't feel artificially held back.   

Edited Sunday at 11:11 PM by Lithium

[thx78]

3 minutes ago, Lithium said:

C'est du couple et de la puissance, tout cela à partir d'un seul tour. La courbe de suralimentation est retenue de 3 500 à 5 000 tr/min de mémoire, donc elle monte fortement jusqu'à quelque chose comme 18 psi, puis monte plus progressivement jusqu'à 23 psi plus près de 5 000 tr/min.

so you can decrease or increase the boost depending on the diet as you wish?  

 

by acting on the wastegate?

[Lithium]

6 minutes ago, thx78 said:

so you can decrease or increase the boost depending on the diet as you wish?  

 

by acting on the wastegate?

Correct.  In the case of the 500kw dyno plot I showed you the car actually runs two boost control solenoids for boost control and a 5psi wastegate spring.  It allows me to control how much boost pressure is applied to both sides of the wastegate valve at any point and fairly accurately control boost target as a result.

I've tuned it so that it's able to target anywhere from 5psi to 25psi depending on what's needed.  The target tables I've set up in that car are Gear vs RPM, so every gear has potential for a different boost (and torque) curve.   First and second gear have quite low boost targets, third gear actually has different target boost all the way through the rpm range as it's a stock RB25 gearbox - the boost targets have been chosen to maintain a peak of 600nm (what the owner has set as the maximum torque he's happy with putting through the stock 3rd gear) but it carries that to the rev limiter.   The boost curve to achieve that is something of a ramp up, then hold, then ramp up again and the power curve looks more like a flat line haha.

 

Edited Sunday at 11:22 PM by Lithium

[thx78]

I am impressed with all this level of adjustment. I didn't expect all this possibility

[Lithium]

1 hour ago, thx78 said:

I am impressed with all this level of adjustment. I didn't expect all this possibility

There is a LOT of stuff that can be done, it all depends on how much time and money you want to spend on doing in.  Not all ECUs will be able to do it, and the more control you need the more time and knowledge needs to be put into making it work. 

If you're willing to spend the time and money and have the right hardware and skills involved there's a lot that can be done. 

Edited Monday at 12:35 AM by Lithium

[thx78]

to have this type of adjustment possible with a curve as I wish, what budget (engine) should be planned?

 

[Duncan]

 

I'm still not sure what you are trying to acheive, but I wanted to clarify something about Lithium's response to your question...

20 hours ago, thx78 said:

so you can decrease or increase the boost ......as you wish?  

 

20 hours ago, Lithium said:

Correct.....

 

Only within certain limits. All of the examples Lithium gave were of detuning for a particular reason.

Before the engine/turbo/wastegate combination has hit full boost, you can't increase it....the wastegate is shut and the turbo is pushing as hard as it can

Once the combination has hit the target (controlled) boost, you generally have full control over whether it makes more or less boost, because it is being limited/reduced from its potential. This is generally done by controlling the wastegate

In the top end, with an undersized turbo, it is possible you won't get to your target boost anymore if the turbo is choking up. In this case you cannot increase boost any more without changing the setup.

2 hours ago, thx78 said:

to have this type of adjustment possible with a curve as I wish, what budget (engine) should be planned?

No change in cost compared to any decent build, other than tuning time to get the engine to make less power than it is able to at a certain point

[thx78]

one thing escapes me, if the turbo spool provides the maximum boost and we can't control anything before that, how can the boost be higher by it alone.

 

here around 27 psi at 4200 then up to 32 psi at 6500 rpm

 

how does the tuner do this

[thx78]

they choose the number of psi (27) around 4200? Could it produce more at this engine speed with the same turbo?

Edited Monday at 08:42 PM by thx78

[Lithium]

59 minutes ago, Duncan said:

Only within certain limits. All of the examples Lithium gave were of detuning for a particular reason.

 

Good points!  Took those for granted - though I *did* actually give an example of winding boost up more than it had been before where I wasn't actually specifically looking for more peak power.  The Toyota Starlet dyno plot that I shared and mentioned showed an overlay from the previous tune it had and the one I'd redone, I left the boost targetting the same as the old tune but then after peak power I ramped boost up by a good 5psi or so over what the old tune had at the same rpm.

The reason I felt comfortable with this (though the owner of the car had a "are you sure?" moment when I suggested it) is that the setup wasn't turbo limited, it was largely head sealing limited and the owner was a bit concerned as for the last few seasons he'd had issues with head lifting - sometimes not completing an event without having some headgasket issue, so he didn't want to run any more boost than it was.

The reasons I was ok with raising the boost a good 5+psi MORE than that was that I left it where it was in the middle, and only increased it where VE (and therefore cylinder pressure vs psi) were dropping hard and I didn't stop the torque from dropping, just reduced the drop.  Well, there is that and the fact that the previous tuner had it overtimed by near 7degrees at peak torque - but that's another story haha.   

I didn't really go into detail about that "turning it up more" thing but now you've mentioned the "not detuning", sometimes the boost *can* be turned up higher than you'd expect if the setup allows for it and you do it smartly.  I've tuned things to run 30psi on BP98 "safely" that a few years ago (or still?) people would cry that it was a stupid idea - but given they were well intercooled, low EMAP turbos and only doing that kind of boost where VE is dropping etc I was pretty confident it wasn't as cowboy as it sounded at face value and we never ended up with issues as a result of it.   

 

Edited Monday at 08:43 PM by Lithium

[Duncan]

24 minutes ago, thx78 said:

one thing escapes me, if the turbo spool provides the maximum boost and we can't control anything before that, how can the boost be higher by it alone.

 

here around 27 psi at 4200 then up to 32 psi at 6500 rpm

 

how does the tuner do this

we can only guess exactly what happened from our keyboards, but to me it looks like they were targeting 27 but the wastegate was too small or poorly controlled and the boost crept up, that is quite common

[thx78]

so this is an anomaly??? I thought it was a voluntary increase

[thx78]

I don't understand anything anymore.