Sale ends todayGet 30% off any course (excluding packages)

Ends in --- --- ---

Small turbo lacks gut-punch in the low/mid RPM's

General Tuning Discussion

Forum Posts



Tech Articles

Discuss all things tuning in this section. News, products, problems and results. 

= Resolved threads


Hi all,

this could potentially have a mechanical solution instead of tuning, but I wanted to get some input before I go and change things.

I have a gen 3 3sgte with a gtx2860r turbo, internal wastegate w/ 7 psi spring. Car's running an elite 1500 and I'm using the MAC 3 port boost control solenoid that Haltech sells on their website. Turbine housing is .63 ar and I've already ported the wastegate.

I had the car street tuned earlier this year as there are zero Haltech tuners in my area, and when my out-of-town tuner was available, we had scheduling issues with the only load-bearing awd dyno in town.

While dialing in the boost, he told me that the boost was coming on too strong for the controller to react quickly enough, and the car kept overboosting. The target we were shooting for was 20 PSI (right towards the end of the efficiency map for this turbo) but the car would just blow right by it. So he moved the target down to 17 and modified the duty cycle to slowly open the iwg door over the rev range. The car pulls nicely at 6-7k, but it really lacks that midrange hit that I was expecting when selecting a smaller turbo for this street car.

I've been reading that a conservative timing can cause the turbo to spool much faster, so I'm considering booking some steady-state time myself to dial in the table. Since the car was street tuned, it's impossible to know how accurate the existing timing config is.

Alternatively, I've been wondering if moving to a 14 psi wastegate spring would be the answer. It's a fair bit of effort to take the turbo back off again though, so I wanted to see what the others who have been around longer than me thought.

I know that getting the timing map nailed would have horsepower gains, and I'll probably do it myself someday - but I don't want to take a car to the dyno that has a mechanical limitation.

All things being equal, that will only make it worse. I would back the DC at low boost off so it let's the gate start to open earlier and confirm both ignition timing and cam timing are in the right ballpark.

I would agree with Micheal - you seem to be getting a shed load of boost and, normally, that would mean a shed load of torque. That you aren't suggests the camshaft timing may be out, compromising the gas flow through the engine when it's needed, and timing which can both reduce power and increase boost by imparting more energy into the exhaust.

While the camshaft(s) are less likely, IMO, to be the main problem and can be a pain to check, unless you KNOW the timing is correct you may otherwise be wasting your time.

With the ignition timing, many people who should know better will run it excessively retarded, to 'be safe', and this has two significant, undesireable results...

1/ The engine is not using the charge in the cylinder effectively for power, and is passing more energy through the exhaust gases. This means what is in the cylinder isn't being used fully, costing power/torque, and...

2/ The exhaust gas temperature, and hence pressure, is higher because, normally, the gases' expansion within the cylinder volume as the piston was being pushed down was also losing energy (cooling) from the expansion, and they also had more time to lose heat energy into the head/cylinder bore/piston crown. The energy used for heating in the charge was the same, the heat energy loss within the engine was less, so there would be (significantly?) more exhaust gas pressure driving the turbine and hence providing boost.

I would suggest doing what you could to minimise the boost pressure, and get the timing where it should be - even if it further reduces 'boost' - and work up from there with increasing boost levels.

It may seem strange that I'm not that concerned about maximising "boost", but that is JUST a measure of the resistance to the gas flow. It should really be considered a consequence rather than an aim, unlike, as most people seem to think, the end-all and be-all. As a general rule, the lower the boost for the desired power level, the more efficient the whole process will be. For example, it is common for both power to increase and boost to drop with better flowing cylinder heads, with NO other changes.

Thank you both for the reply! It seems that I did not accurately communicate, sorry about that. It was a good idea to check though, and both camshafts are lined up with the crank. All hail Gritty, btw.

When the tuner was trimming the duty cycle (open loop), he set it to function just like a manual boost controller, albeit very slow. So I'm getting like 7 psi at 3k rpm, 10 psi at 4k, 13 at 5k, and so on. It's a smooth power band, but not as much fun on the street as 17 psi at 4k rpm, which I'm sure this setup can do.

I thought about what Slides said though, and that makes total sense with the wg spring, I was thinking about it wrong. If I need to change anything mechanical, I guess it'd be going from .63 to .72 a/r turbine housings to get more flow and less boost creep. I don't want to lose the responsiveness of a smaller housing if I can manage it though, so that's why I'm asking about too much ignition retard. The other mechanical change would be moving to an external waste gate (obv not easy to just do) or porting the internal waste gate even more; I'm afraid the latter would involve taking out too much material though.

Ah, I did mis-understand what was happening.

It would seem the problem is the opposite of what I thought, with the tuner pulling all the boost right where you wanted it?

Again, it would be wise to check with those much smarter than I, but the .72 would be expected to worsen the issues, as it would be less effective at lower rpm where the exhaust gas flow is low, and more effective at high rpm where it would choke the exhaust less?

It might be time to have a good re-think on your boost control strategy, especially the way you have the hoses conected to the controller - might be worth looking at a 4 way? Oh, don't forget to go over the hose and fittings' inner diameters and check for restrictions that may slow the pressure transitions to the controller and waste gate? It would also ge worth checking the waste-gate is completely closing as a leak will delay it coming on boost, too.

Ahh so you could very well be right on that. I would hope that the waste gate door would be slightly bigger with a bigger a/r housing , but who is to say for sure without having experience with both ?

Hoses are new silicon and the run is probably 18 inches, halfway in the middle is the Mac 3 port. I don't think I can use a 4 port on the internal waste gate, but I'll keep that in mind if I go to an ewg in the future.

Just as a test, I unplugged the boost solenoid and the car holds 7ish psi until like 6k, then it creeps up to 10. I suppose that means that more porting would be needed to the wg door to hold 7 psi steady, but I really want to be in the 17 to 20 range, so I'm hoping I can compensate for that creep in the duty cycle table.

I then plugged the solenoid back in and set duty cycle from 0 to 3500 rpm to 100% and I hit 17 psi at 3250 rpm. This thing builds boost like crazy.

I'm inclined to agree the existing boost control strategy isnt optimized. Maybe switching to closed loop and trying to work out the PID functions is what I need to do next. The duty cycle table here isn't set up like I expected it to be though, so I'm going to need to do some learning here. At least I can mess around with this on the street before booking dyno time. The only reservation I have is that the tuner disabled knock control, so I don't/won't know if the engine is knocking as I hit different cells in the map, further left than what was getting hit before.

I've been in a similar situation with an sr20 albeit much laggier combo, targeting 20 psi with boost control on a 7 pound spring would always overshoot. A 14 pound spring solved all issues, this was with a greddy profec at the time though so no total control, mapping duty cycle with the Haltech on less than ideal gate springs it might be possible to get somewhat decent results if you creep up on it, it's just not worth the extra time and effort for most tuners to dial in. Tuning my own 13b with a 7 pound spring that creeps to 11psi I can say it's a lot of work trying to get stable boost that comes on strong and holds where I want it (12-16psi), each gear needs its own special attention

So that's an interesting thought that I hadn't considered. It does look like the elite will be able to detect what gear the car is when comparing the ratio of vehicle speed to rpm, so that's really cool. I'll add that axis to the boost control table and start messing with boost per gear.

In relation to that topic, would anyone be able to please explain why the closed loop base duty cycle default channel for the x axis in the haltech is "boost control target pressure (corrected)" ? Why wouldn't I just use manifold pressure instead ?

I was able to answer my own question re: closed loop base duty cycle by a good round of RTFM. Despite searching via the normal means, I wasn't able to find the below article until I went to open a support ticket and was auto-directed to a number of topics matching my query.

Now I'm working on tuning boost by gear. I have first and second already done, with varying levels of success (target in first is 17 psi, but I'll settle for 15.5 if the boost curve is much steadier and predictable than a bipolar duty cycle map that peaks and then valleys).

For ease of searching in the future, I copied the content down below. I hope this doesn't break any forum rules as this is free and available on Haltech's website.

source: https://www.haltech.com/support/309315000033092435

For this example I am using one of the Haltech Trim Modules to allow for more than one boost target setting. This example shows the purpose of the Base Duty table.

For Setting 1 the target will be 10psi. For Setting 2 it will be 16psi. For Setting 3 the Target will be 18psi but I want 20psi from 5000 rpm and higher.

My Target Boost level for this example

Because there is more than one target value the boost control solenoid needs to be pulsed at a different duty for different boost levels. This is achieved by giving the ECU a starting point for the different boost levels. The Closed Loop Base Duty table does this for us. As can be seen in the below image, the duty cycle is mapped over Boost Control Target Pressure so that as we change the Target the ECU knows which duty cycle to use.

Base Duty table for this example

Based on this table, when I am using Setting 1 (Target is 10psi) the system will use 15% duty to reach this target boost amount. If this is not correct I will be only adjusting the column only for 10psi until it does reach 10psi. The other values in the other columns have no impact at all on the controller at this time.

When I am using Setting 2 (Target is 16psi) the system will use 38% duty to try to reach 16psi. If this is not correct I will only adjust the 16psi column until I get 16psi. The other values in the other columns have no impact at all on the controller at this time.

When I am using Setting 3 (Target varies between 18psi and 20psi) the system will use 46% duty until 5000 rpm to try to reach the 18psi target boost. From 5000 rpm it will then use 54% duty to try and reach 20psi boost In this case I would be tuning only the 18psi column under 5000 rpm and the 20psi column above 5000rpm.

It must be noted that it is very common as a Haltech Tech Support employee to see users who for some reason change the horizontal axis from Boost Control Target Pressure to simply Manifold Pressure. As can be seen by the above description of how the system works, doing this does not give the system a way to know a starting point to get to the Target. It is also common to see users who set the entire table to be one value, which means the system will only reach a single target accurately and all other targets can never be reached correctly.