AEM Water Injection Failsafe Device: Part 2 – Catching Our Mistakes and Upgrading Our Water Injection System

,

This was causing our water pump to suck air and cavitate when the pump motor was running at high speed under higher boost levels. The fittings didn’t leak water but they did allow air to enter under high suction conditions and it doesn’t take much air to cavitate a pump. The problem was easily fixed by going to longer lines!  This immediately fixed our problems and resulted in much higher water flow at higher boost levels and less scatter.  Now full throttle runs would lie on top of each other and our entire operating range plot flow distribution was narrower.

We made the line from the water tank to the filter just a few mm too short.  This little mistake allowed for air to leak in in-between the tank and the pump which was worse the faster the pump ran.  The fittings did not leak water unless they were wiggled.  The line from the filter to the pump was not seated correctly either.  These invisible problems would have been nearly impossible for us to catch without the injection monitor as we cannot stare at the flow meter for an entire gear at full throttle, we do have to do stuff like drive and watch the tach as well!  Small problems like this could damage an engine and make tuning difficult.  Even experienced people can still make mistakes and this one was so insidious, we had to share it with you.

This was a great example of how the AEM flow monitor and how it datalogs can help even fairly experienced tuners catch an elusive problem.  We had been driving around with a malfunctioning water injection system for several months and would have never suspected anything if it were not for the monitor.  This issue would have made tuning a problem and perhaps could have resulted in engine damage.  We are really believers in the water monitor and for the first time have confidence in water as a peripheral tuning device because of it.

The log with everything fixed looks much better.  We mounted our flow meter in the trunk by the pump for a cleaner simpler installation with shorter wire runs although AEM recommend placing it closer to the nozzle.  This makes for more scatter in the data which isn’t ideal but it still works.  For instance, the spike in flow at low flow percentages is caused by the momentum of water in the line continuing to flow when the throttle is shut suddenly even after the pump is shut off.  It’s not a problem but the data looks a bit odd.  The control limits are now tighter and the system does not false alarm anymore like it was occasionally doing previously.  Our system is working perfectly and the monitor is effectively watching the system.

While we were trouble shooting our system we took the opportunity to make some upgrades, when we were at the PRI trade show we noted that AEM had upgraded their water injection system.  We took a trip down to AEM to pick up the latest parts to upgrade our system to the latest spec.  The most notable change is the pump which has been changed to the pump previously used on their heavy duty diesel system or as an option for the gasoline system that we have.  Now the heavy duty pump is standard equipment across the board.