WET AND WILD
Our journey into the world of water injection begun with the CRE Datsun 1600, back in 2011. We were exploring ways of extracting more power from the FJ20, without any major hardware upgrade, when we stumbled across an article on the benefits of water injection.
After conducting more research and speaking to a number of experts in the field, it was decided that this was the path we’d take. Our research also points to the fact that Aquamist were, and we firmly believe still are, the world leaders in aftermarket water injection systems. A HFS3 system was acquired and the install and testing begun.
Installing the Aquamist unit involved mounting the tank and pump in the boot, the control solenoid, flow sensor and atomising jet(s) in the engine bay and finding a home for the control unit inside the cabin. Once the hardware install was completed, the control unit was wired in and set up of the system and testing commenced. As our experience with water injection was very limited at this stage, it was decided that a single jet mounted in the post intercooler pipe work would suit our requirements. The jet size chosen was 1.0mm, rated at 520cc @160psi.
Once the install was completed, testing got underway, with our good friends from Unigroup Engineering tasked to recalibrate the Autronic ECU. On the dyno, the car made 290rwkW with the WMI system activated. This was an increase of 30kW, 11 per cent over our previous best. This was achieved by increasing the boost pressure from 22psi to 28psi. Timing was left untouched as we felt we had reached the tuning limit of the single jet.
The upgrades to the 1600 continued over the next months including a twin-scroll turbo charging system, CDI ignition system and a much larger gearbox. The water injection was recalibrated also and while peak power remained the same, power in the lower and mid-range were vastly improved. Although the 1600 was performing better than ever, we knew that there was more we could achieve. As the engine was knock limited to 290kW, it was decided that the most cost effective way to overcome this was by upgrading the flow of the water injection system. The original idea was to simply added an extra jet but that was scrapped after some new information came to light.
We found out that almost all WRC rally cars ran water injection and exclusively Aquamist systems… prior to it being banned. They all used direct port injection, which utilises one jet per inlet runner. The idea behind direct port injection is that it allows for equal distribution to each cylinder, where maximum cooling and knock suppression is achieved. So we implemented this setup along with a small post intercooler jet. In addition to this, we also added a small pre-turbo jet to improve the efficiency of our turbocharger, as the boost we intended on running would push past the turbo’s peak efficiency range.
With install completed, it was time to hit the dyno again. We had managed to get our hands on some methanol, so we mixed up a batch of 50:50 water methanol to see how it would stack up against straight water. The results were as follows:
268kW @22psi – without water or water/methanol injection
291kW @25psi – water injection
312kW @30psi – water/methanol injection, with 512Nm of torque.
We were now beginning to see the true potential of the water injection system. For this setup, we followed the Aquamist guide for selecting jetting requirements. This proved to be a little conservative for our specific application, as once again the tune was limited by knock.
The catalyst for the next round of upgrades came from an electrical failure, causing the water pump to stop working, overheating the engine and damaging the cylinder head. With the head removed, the opportunity presented itself to conduct some major reworking of the electrical system. This included upgrading the ECU as we had simply ran out of aux input and output on the Autronic. After securing a great deal on a Link G4+ Extreme, it was wire in place, along with a new 24-tooth trigger wheel and LS2 coils.
The head was rebuilt and replaced, swapping out the standard head bolts for ARP studs. After damaging one of the direct port jets the morning of the dyno, a solution had to be found. We decided to remove the remaining direct port jets, blank the holes, and fit to large jets in the post intercooler piping. These along with the pre turbo jet, gave a total max flow of 850cc, a 260cc increase over what was previously installed. The dyno result was 331kW @32psi.
With some extra time up our sleeve during this dyno session, we plumbed an extra jet into the post intercooler pipe, bringing the total water meth flow to 1150cc. For the first time since we began using water injection, we observed a decrease in power, recording 320kW at the same 32psi. This decrease was attributed to one of two factors, injecting too much fluid into a localised area, causing saturation of the air and rendering the extra fluid ineffective or we were simply using too much fluid overall. Either way, the net result was the same, excessive cooling inside the combustion chambers. The extra jet was removed and power restored to 331kW.
You may be asking why we didn’t increase the boost pressure and timing to compensate for all the extra cooling. At the stage, we felt that the turbocharger and fuel system were at their limits. Time to upgrade!
As previously mentioned, we had reached the limits of the turbocharger and fuel system. Resisting the temptation to go silly with the turbo size, we opted for the GEN2 Garrett GTX3071R, a direct replacement for our GT3071R unit. And in twin scroll of course! For the fuel system, the Carter lift pump and 044 were replaced with an E85 capable in tank Walbro 460.
We modified an XR6 fuel pump cradle to suit the 1600 fuel tank and finished it off with some speed flow fittings and braid Teflon lines. Injectors were upgraded to Bosch Motorsport 1550cc items. We also conducted a major overhaul of the water meth injection. This involved fitting 4×0.6mm direct port jets, 2×0.5mm post intercooler jets and 1×0.4mm pre turbo jet. In addition, we were no longer using injector duty cycle to control the rate of injection. Instead, we wired the Aquamist controller input to a PWM output on the in the Link ECU.
From there, we set up a custom load table in the ECU to drive the PWM output. This gave us greater control over the water injection rate, especially as we were now jetted to 1600cc total flow. Back on the dyno and the new turbo setup was showing great improvements. 307kW was achieved @ 24psi without water meth injection.
With water meth injection, 30psi and 5 degrees more timing, power increased to 361kW. It should be noted that the water injection system was operating at below 50% duty cycle for this test. As we were having a few miss fire issues, we decided to end the session, leaving things as is until the issue was rectified.
This round of modifications is currently underway and will focus around minor defect rectification and fine tuning of the water injection system. It is also our intention to conduct some testing of different ratios of water meth, starting with a 50:50 mix, then 80:20 methanol/water mix and finishing off with 100% methanol. Thank you for your support and stay tuned.