It's an unarguable proposition that all cars are different - even the same make and model. That statement applies even more so to modified cars, where each change hasn't been carried out with production line precision. So, while when you read a magazine article that deals with changes to "your" car, it's always more interesting that an article on a obscure vehicle that you haven't even heard of, the fact that it is the same model is no guarantee that you'll get exactly the same results.
This story's about Dan Hockin's R33 GTS 2.5 turbo Skyline - but its combination of starting mods and the execution of them means that Dan's car is probably not like your car - even if you too own a GTSt Skyline! That doesn't make the story any less interesting, though, because Dan's is a typical tale of real world tuning.
The Starting Point
As with all other R33 Nissan Skylines in Australia, Dan's car is a private Japanese import. The car was purchased with these mods already in place, not all of which were to Dan's liking:
- HKS airfilter pod (breathing hot air)
- HKS turbo timer
- Bleed boost control
- Ignition timing retarded by 5 degrees
Dan then added:
- 255/235/45 tires on 17s
- 3-inch exhaust from the cat back with a JunBL muffler
- Autometer boost gauge
Wanting to gain better boost control, Dan fitted the system detailed in the AutoSpeed article, "Brilliant Boost". This approach uses a valve to vary the flow available to the factory boost control solenoid - reduce that airflow and boost will increase. The guinea pig car for the development of the AutoSpeed system was an R32 Skyline GT-R, a car on which the control worked perfectly. However, on Dan's car, the Brilliant Boost system was anything but brilliant.
Prior to the Brilliant Boost control, the Skyline had run a basic bleed boost control. "When I bought the Skyline it was running 6 to 10 psi, via a dodgy homemade bleed pipe. This had the tendency to spike (not two-stage) and the car lost a lot of its inherent smoothness and drivability. Boost would not come on till around 3000 - 3500rpm. Bit of an all-or-nothing affair," recalled Dan.
"I then found the Brilliant Boost system described on AutoSpeed. I saw it had been used on a car similar to mine, minus one of the turbos, and on reading the article thought it sounded a cost-effective method of achieving accurate boost control. Electronic systems were retailing for $600-$1000 - way out of my budget. I bought the valve and associated fittings and set the system up as detailed.
"I found the system to be easy to fit and to accurately control max boost. There was a noticeable difference in drivability (much better than the bleed), providing the boost was kept below 9 psi. Above this it became a two-stage system, with boost rising to around 6-7 psi below 4500 rpm, and after 5000 rpm jumping to 10 psi. While this was nice in a straight line and gave you that bit extra (quite a lot actually!) up top, it wasn't very nice on your favourite section of twisty roads, as you had to be quite careful on the limit."
Other than the bleed and this system, Dan hadn't experienced any other boost control approaches, so he figured the two-stage boost curve was par for the course with a cheap DIY system. So while he wasn't rapt with the boost behaviour, it wasn't on his list of things to change. Not at that point, anyway.
The Apexi S-AFC
Instead, he managed to get his hands on a secondhand Apexi Super Air Flow Converter, an interceptor module that wires into the ECU and allows the modification of the fuel curve. But why change the fuelling when so far the mods were pretty minor?
"The main reason for fitting the S-AFC was more long term thinking, and cost considerations, rather than immediate gains. I hope to fit a large front-mount intercooler soon, and together with larger exhaust and pod filter I was under the belief that the flow characteristics would have changed. To counter this, the mixtures would need to be altered to get the best out of the system," he said. "As a cheap second-hand unit came up I grabbed it, and when a good priced IC comes up, I'll do the same. I suppose when money is an issue you grab what you can when you can, and not always in the preferable order."
Unlike some Japanese aftermarket electronics manufacturers, Apexi really have their act together. When we reviewed the Apexi Power FC ("The APEXi Power FC Engine Management ECU") we were very impressed, and while the Super AFC is not nearly as sophisticated (or expensive!), the documentation supplied and tuning capability are both extensive (If you want to download a PDF of the many-page instruction book, go to "APEXi USA - Electronics").
The S-AFC changes the air/fuel ratio by intercepting the signal from the airflow meter, and then altering it so that the ECU thinks that there is more airflow passing into the engine than there really is (ie results in a richer mixture) or less airflow than there really is (ie results in a lean mixture). This can be done for both small and large throttle openings (the throttle percentage is user-definable for both settings) and at 500-rpm intervals. For example, if at light throttle openings it is desired that the mixture be enriched at 2000 rpm, the 'up' arrow key can be pressed and a bar graph on the display grows in height as this mixture change is made. Alterations of as much as plus/minus 50 per cent are possible.
Dan had the S-AFC fitted and tuned by John Keen of DAT Racing in Adelaide.
The first step was to read the fitting and tuning instructions. The S-AFC is connected to the ECU harness and so the pin-outs of the individual car to which it is being fitted are required. Obviously, get even one of the seven connections wrong and the device won't work.
Next the battery was disconnected. This reduces to zero the chances of making an accidental short circuit, but it also clears the ECU's fuel trim memory - not necessarily a good thing.
The trim panel was pulled away and the ECU loom accessed. The correct wires were picked (using the ECU pin-out table) and then....
...the right wiring connections were made. John used crimped-on connectors that take a male spade terminal.
In addition to displaying the fuel changes, the S-AFC has an informative display that can be used for short data-logging, to display sensor inputs, and show configuration values. It's therefore easy to check that the throttle input, airflow meter input and tach inputs are all working as they should be.
Dyno tuning commenced with John holding some light loads while he used the S-AFC to alter the fuelling. The air/fuel ratio meter that was being used was the in-built (old model) Dyno Dynamics on-screen meter, with the probe mounted at the tailpipe. The display showed that the air/fuel ratio could be changed at will, with the S-AFC showing good tuning sensitivity (without being too sensitive!)
At small throttle angles John used the S-AFC to increase the fuel flow by about 12 per cent. This was needed because the Skyline (as with most cars) maintained an air/fuel ratio of around 14.7:1 even with moderate throttle. (Manufacturers set the air/fuel ratio to 14.7:1 to allow the cat converter to work most efficiently. It's good for emissions but it dulls throttle response.) However, the fact that the Skyline was using this air/fuel ratio at these throttle openings also shows that it was staying in closed loop. That is, the feedback of the factory oxygen sensor was being used to trim mixtures to this ratio. You'd therefore expect that the feedback loop would be able to tune out the changes being made by the S-AFC at low throttle angles... At high throttle angles, the air/fuel ratio of the Skyline was pretty well on the money; John trimmed it back by 3-4 per cent through 5000, 6000 and 7000 rpm.
A dyno run showed that 132kW was now available at the wheels. However, more importantly, the dyno curve showed that there was a dramatic step in mid-range power - from 96kW to 117kW in just a few hundred rpm! This was the result of that boost jump from 6-7 psi to 10 psi.
And what did Dan think of the changes resulting from the fitment of the S-AFC? "After driving the car for a few days I've decided that outright power has not increased noticeably but pick-up definitely has. You don't need to floor it to get good acceleration, and therefore can feather the throttle more." But while he was happy with the S-AFC, having seen the appalling power and torque curve jumps as the boost changed so rapidly, Dan decided that a new boost control system was now a priority. "The boost controller needs to be sorted to bring required boost on earlier and in a constant form, instead of the 2-stage format it has now," he said.
John Keen had suggested to Dan that to get the most accurate tuning, Dan should have an oxygen sensor fitting welded to the exhaust close to the turbo. John could then use his very accurate Autronic air/fuel ratio meter to do some final trimming of the mixtures.
The New Boost Control
Unhappy with the poor results being provided by the Brilliant Boost system, Dan decided to fit the third AutoSpeed boost control, as detailed in "The Audi's DIY Boost Control - Part 1". This uses a pressure relief valve and a pressure regulator mounted in series between the boost source and the wastegate actuator. The pressure relief valve setting determines when the actuator sees boost (allowing wastegate creep to be dialled out) and the pressure reg sets how far the wastegate opens, allowing the peak boost value to be controlled.
Two systems using different pressure regs were covered in the AutoSpeed series - Dan decided to use the cheaper (and so less accurate) of the two pressure regulators as it was more easily bought in the short time available before the next tuning session was booked. I agreed to help him tune the new boost control, a process that is best carried out on the road.
With the system plumbed into place, the pressure reg and pressure relief valve were both set to have no affect. Then, step-by-step, the pressure regulator was opened, until a peak boost pressure of 10 psi was realised. Testing was being carried out in second gear; with a peak of 10 psi the regulator allowed boost to rise a little to 11 psi in higher gears. However, given that the airflow through the intercooler is greater the faster that the car is travelling, this small boost rise in the upper gears was regarded as being of no consequence. With peak boost set, the pressure relief valve was then closed down (again, step by step) to give faster boosting through a lack of wastegate creep. Again, this performed as expected, with more boost available on smaller throttle openings, and at full throttle, boost rising more quickly.
However, did the new system get rid of the two-step boost curve? The answer to that is 'not completely'. If held in second gear from a low speed, the boost rose firstly to 8 psi, then jumped to 10 psi. But if the car was driven hard through the gears, the boost rocketed around to 10 psi (11 in higher gears) following each gear change, holding absolutely steady at the designated value. Dan suggested that the change in boost value might be the result of the variable camshaft timing, which alters in one step. In any case, the boost control was markedly improved over the previous two systems (the Brilliant Boost flow control valve and the basic bleed) and so Dan was happy.
It was time to go back to the dyno.
Second Dyno Session
While on arriving for the second dyno session Dan suggested that he could still feel the changes in response made by the S-AFC, examining both the air/fuel ratios and the S-AFC screen showed that the alterations that had been made the week before were no longer present. In fact, the S-AFC appeared to have lost completely the revisions to the low throttle fuel curve; perhaps this happened when Dan re-wired the S-AFC into the glovebox. Whatever - but it was interesting that the driver could still perceive improvements that were no longer present....
The more accurate Autronic air/fuel ratio meter, together with the revised sensor location, meant that far closer scrutiny could be made of the mixtures. And at light loads the self-learning behaviour of the Nissan ECU could be clearly seen. Even if the S-AFC was used to make huge mixtures changes (eg +20 per cent), the Autronic air/fuel ratio meter display very quickly showed the mixtures reverting to their factory standard 14.7:1. Score one for the Nissan ECU and zero for the S-AFC... (This re-learning behaviour is typical at light loads when interceptors are used - see "The Fueltronics AMFC")
However, the change in boost behaviour with the new control system (and also perhaps the fact that a better air/fuel ratio meter was being used) meant that the air/fuel ratios at high loads were read off as being much too rich for max power. In fact, John decreased the fuelling at full throttle by a massive 17 per cent at 4000 rpm, 4 percent at 5000 and 7 per cent at 6000 rpm. He also put in some intermediate 500-rpm sites, dropping fuel at each of these by about the same amount.
Time for a dyno power run - and what a difference was revealed! Not only was peak power up (now reaching 148kW) but the midrange hole in power was gone almost completely. In fact the mid-range was up a staggering 16 per cent and the two-stage boost curve was all but gone. But there was even more to come. Returning the ignition timing from its 5-degree retarded position back to standard reaped an exhilarating top-end of 156kW at the wheels! Over the starting point, the new boost control, S-AFC and return to standard ignition timing boosted mid-range power by a staggering 33 per cent! After a test drive Dan was grinning from ear to ear...
After a few days we asked for his driving impressions.
"The main differences I noted in the new boost control was the accuracy of max boost pressure, which is held steady at 10 psi. The turbo now comes on boost from as little as 2200 rpm, with half throttle. In the higher gears boost comes on almost instantly, especially when going up through the gears from 1st. This is most noticeable in the midrange where some 30-40 kW were made as seen on the graph.
"The only problem I have encountered is in second gear from low rpm, where boost has the tendency to swing round straight to 10 psi, then back off to 7 psi before rising again. This will hopefully be improved on with fine-tuning, as it seem as if there's a little too much 'overboost' initially. With the two valves working more in unison I hope this will be sorted out. Initially the boost was set up at around 24 degrees ambient temp, and recently the weather has warmed up to 31. I haven't noticed any measurable change in the behaviour of the system, apart from the small general loss associated with higher air temps and power losses. Basically, boost comes on hard from as little as 2200 rpm, and holds dead steady at the desired maximum. There's no spiking or '2 stage' behaviour, apart from the 2nd gear problem which I think can be ironed out with time."
Conclusion
In summary the changes made were:
- Fitting and tuning of Apexi S-AFC using a dyno and an Autronic air/fuel ratio meter with front-mounted sensor
- Removal of Brilliant Boost system and fitting of AutoSpeed Audi-type boost control
- Returning of ignition timing back to standard
And while it's easy to say in retrospect that some of the power and driveability 'gains' were made simply as a result of removing bad modifications rather than fitting good ones, the fact remains that Dan is a very happy man - his car has 18 per cent more top-end power and 33 per cent more midrange grunt than when he started the process!
Contacts:
DAT Racing
08 8277 4222
|
The Dyno Curves - Power and Tractive Effort
Red - as the car arrived Blue - Audi-type AutoSpeed Boost Control and S-AFC fitted and tuned Green - as above plus ignition timing returned to standard
|
