All about MINI - An Introduction to Water/Alcohol Injection Systems

An Introduction to Water/Alcohol Injection Systems

by Jeff Howerton
All images courtesy of the Author

Introduction

There is no question that Water/Alcohol Injection (WAI) is a valuable tool in the tuner’s arsenal for obtaining more power from an engine. The main benefit of these systems is knock suppression and reduced intake air temperatures (IATs). The combination of these items can result in more power through the capability of increased ignition advance and reduced exhaust gas temperatures (EGTs).

There are many types of these WAI systems on the markets by over a dozen different manufacturers. Some of these manufacturers cater to certain types of cars; big V8s, sport compacts, diesel trucks, etc. Others make universal systems adaptable to many makes of cars. With all of the systems available, which system is best for a certain application, or which vendors are better than others? In the following pages we will look into the various types of systems and explain how they work and how these variations may be applicable to the MINI drivetrains.

Volumes have been written about this subject in the past. Much of the following information has been taken from technical papers, discussions with industry leaders, and my own experience testing and evaluating systems. A special thanks to Richard Lamb of Aquamist Systems for allowing the use of technical diagrams and the paraphrasing of some of his writings.

Application to MINIs
How Does All This Apply to the MINIs?

Testing has shown the R53 responds well to WAI. With amazingly high IATs, despite what intercooler is used, and its appetite for octane in the timing maps, a properly applied system can give significant gains across the entire RPM range. If one wishes to use a simple single stage system, the midrange can benefit greatly and the top end will receive some IAT cooling. If set up to inject the proper quantity in the midrange, the top end may not get as much as it would like, but some is better than nothing. One could add a second activation point and solenoid/nozzle to help out the top end. If one wishes to move onto a progressive system, the difficulty comes in injecting the correct amount of water/alcohol across the entire RPM range. We have seen the PPS system has a limited dynamic range which limits its effectiveness across the entire RPM band. Many cars with superchargers have a MAF sensor which can be tapped for a relatively good signal to use for progressive injection. The R53 does not have a MAF sensor. Being supercharged, the boost profile is not linear with RPM either. Therefore, the boost signal is compromised for an accurate injection signal as well. This leaves few options for finding a signal respective to engine load for accurate injection. Combine this difficulty with some of the shortcomings of a PPS system, and you could be chasing many variables trying to tune a system correctly. That’s not to say a properly set up simple system cannot be effective, but one should know the limitations you are working with in formulating your expectations. For the R53, IDC is the most likely candidate to give an accurate signal for injection parameters.

After much testing and data logging, I have some observations that may be helpful in considering the final package. Tuning plays a pivotal role in how the car will respond to a WAI. If the car has pinging issues, as many stock R53s had from the factory (or the addition of a smaller pulley can cause), a simple system could provide good results. The WAI can suppress detonation, cool the IATs, and provide some additional octane which will advance the timing maps providing more power. If the car has been tuned properly, and does not have detonation issues even under hard use, the WAI system will have a smaller effect. The car will already be running optimum timing for its setup and will be mostly used to combat high IATs. If the engine is re-tuned to take advantage of the strengths of the WAI, the sky is the limit.

The R56, with its super-lean AFRs and hot EGTs seems like a prime candidate for WAI. As for the application of the hardware, this will also be a difficult issue. The U.S. R56 has a MAF sensor, but it is not providing information like a “typical unit” due to the characteristics of the direct injection engine. This engine can run with excess air, similar to a diesel engine, and can leave the throttle plate 100% open during various loads on the engine. Therefore, the MAF sensor signal is not useful. The engine is turbocharged, but being a smaller, quick spool unit, the boost profile does not match true engine load over the entire RPM range, same problem as the R53. Again, this leaves few choices for an accurate engine load signal for the WAI to use. Even the fuel injector duty cycle is drastically different from typical fuel injected engines making it harder to capture than normal. I know of one manufacturer that has captured the IDC signal, so there is hope for the R56. These systems should be coming to market as you read this.

Conclusion
Conclusion

Specific tuning of these systems can range from moderately easy to extremely difficult depending upon the nature of the WAI system, the general build of the engine (stock to extreme) and the overall goals for the system. To add more complexity to the issue, you can inject an infinitely variable mix of water and alcohol, and different types of alcohol to boot.

The single and two stage systems inject at fixed rates, are very predictable and relatively simple in operation. With a good failsafe system they can be very effective and reliable. PPS systems can be simple as well, but certain design limitations must be considered before implementing a system. If components are run outside their original specification, one must consider reliability issues as well. An extensive failsafe arrangement may need to be considered for this type of system. PWM systems may be considered to be the “upper-end” of WAI systems. They can precisely meter flow, react quickly to changing demands, and can provide good atomization at low flow rates. With a top-notch failsafe arrangement, the PWM system can provide the ultimate in flexibility and safety.

The cost of these systems is usually proportional to complexity and capabilities and, as is the case with most things, you generally get what you pay for. That being said, some of the more simple systems can do an admirable job of providing good results.

Components
Components

Pumps

Currently, Shurflo is the main supplier for the majority of water/alcohol injection system makers, so it is appropriate to take a good look into their pump. You will be amazed to know how many pump variations Shurflo offers over the standard "off the shelf" configurations. It has been said they have over 300+ "custom" configurations on file.

Currently, there are three frame sizes for the pump-motor: short, medium and long stack, covering three power ratings of 60W, 100W and 150W. Most suppliers use the short stack, or 60w pump, though a few manufacturers use the 100w, and even 150w. Below is a general arrangement drawing for these pumps.

Shurflo’s literature states the maximum suitable pressure for the 8000 series pump is 150psi. Shurflo, to my knowledge, has never approved running pumps at a higher operational pressure. As we will see later, many manufacturers run these pumps as high as 250psi to compensate for the design shortcomings of their system. Running these pumps at higher pressures can have damaging effects on the pumps and shorten their lives, sometimes quite quickly.

The pump cam angle is important in that it dictates the final specifications of the pump's flow rate and pressure. Shurflo offers 2, 2.5, 3.0, 3.5 and possibly more profiles of cam angles. Depending on the application, WAI makers can select the most suitable cam for their application but most will not publicly state what cam angles are supplied with their systems.

At first glance, using the highest lift cam will produce the most flow and pressure from the pump, but if this cam is matched with a small motor, it will cause undue stress on the motor windings. This is very similar to going up a steep incline in high gear where the car's engine is being stressed.

On the other hand, a low cam angle will produce less pressure and flow. Some Progressive Pump Speed system (PPS) manufacturers prefer using the low angle cam because of the following:

1. Less stress to the diaphragm - long term mechanical reliability.
2. Less stress to the motor - lower running temperature.
3. With a flow range up to a 600-800cc/min or so, there is ample pressure generated to hold the system at 150psi.
4. A much smoother control range from a PPS controller. High cam lobe tends to ramp up too much pressure with the same duty cycle applied.
5. The pressure spike for each pulse is also much smaller; this allows the peak pressure closer to the "150psi demand switch." Overshooting due to larger cam angles will cause the infamous "pulsing" often associated with PPS systems.

Note: A system advertised as 150psi@3 liter/min will NOT out-perform a 150psi@1 liter/min system. Often, the latter is a much better system. As far as the raw material cost difference is concerned, there is none. These days, hype in numbers rules the market.
There are two basic ways of running the pump in the system. First is the Bypass Pump, and second is the On-Demand Switch.

The Bypass Pump is just as its name implies. The pump has internal bypass valves built into its pump head and can run continuously at full pressure with no output. Excess pressure is being "by-passed" internally by a set of spring loaded poppet valves (x3). This type of pump configuration is better suited to Pulse Width Modulation (PWM) valve systems rather than PPS systems. The pump is switched on just prior to an injection event and maintains a "steady" line pressure continuously through the entire delivery cycle. Only a very small number of PWM valve WAI manufacturers use this set up.

The second type of pump activation is the On-Demand switch. Rather than employing internal bypass valves, the water pressure can be limited by using an on-demand switch. This method is simple. Every time the pump hits the "set pressure" of the on-demand (pressure) switch, the 12V power feed to the pump is interrupted. Shurflo recommends this method should only be used in an application where the usage is intermittent and not cyclical.

Against Shurflo’s recommendation, there are WAI systems on the market which are employing this technique to control water pressure. Most PPS systems only hit this pressure at peak injection pressure, and intermittently, so long term damage to the pump is not severe, but pressure spikes of 20+psi exist.

If this on-demand method is being used to regulate water pressure on a PWM valve system, it is a completely different story. The "intermittent" use now becomes "cyclic" use. This will create long term problems on the switch as well as the pumping mechanism. Water pressure will also suffer from pressure spikes, sometimes as much as 20+psi spike. Using a low hysteresis switch may reduce the ripple but the other half of the problem still remains.

On the electrical side of this arrangement, using an on-demand switch in a PWM system can cause significant stress on the electrical system feeding the pump motor as every time the pump starts, an electrical surge much greater than the full load amps of the motor rating is pulled down the wire. This electrical cycling can occur up to 3-4 times per second causing significant high load spikes down the wire and connections.

Nozzles

Nozzles vary somewhat from manufacturer to manufacturer, with some making their own and others using off the shelf pieces. Nozzles can be rated in many different ways, usually at some flow rate for a given pressure. There is a general convention for sizing nozzles that will be used throughout this paper. Nozzles are usually denoted by their “M” number, or gallons per hour of flow at 100psi pressure. So an M1 nozzle would flow 1 gallon of fluid in one hour, an M3 nozzle flows 3 gallons of fluid in an hour and so on. Even though “M” numbers are used for identifying nozzles, it is convention to discuss them in cc per minute of flow to equate them better to fuel injectors. Below is a chart published by Hago, a US oil heater nozzle manufacturer (fluid is fluid, it all follows the same laws). From the chart, one can see how the nozzles spray across a wide pressure range. We will be revisiting this chart later as it will become very important in understanding the performance of a popular type of system.

Below you will find some pictures of nozzles from various WAI system manufacturers. The first is a nozzle from Coolingmist, the next is a nozzle from DevilsOwn, and the last nozzle is from Aquamist.

The most important aspect of a nozzle is that it sprays as fine a mist as possible to atomize the fluid into the air stream. As a general rule, any nozzle will require at least 40psi of pressure to turn from a dribble to a good spray. Some nozzles come with an integrated screen to help prevent clogging.

Nozzles can be installed into the pipe, or injection point, in various ways. Some from the inside (pain in the a**), some from the outside, and others from the outside with a jam-nut on the inside (better use lots of Loctite on that one).

Fittings, Seals and Tubing

These components are pretty standardized for most of the systems available, but there are some differences and other things to keep in mind when sourcing a system. Most systems use nylon or some type of plastic tubing to carry the fluid. There are generally two types of fittings for this tubing, push-to-connect fittings (or One-Touch fittings) and compression fittings. Push-to-connect fittings are cheap, simple and relatively reliable. Compression fittings are inexpensive as well, and usually provide a more positive sealing mechanism for the hose. Barbed fittings are also available for these plastic lines, but these are generally reserved for low pressure applications.

Another type of tubing, or line, is plastic line with a stainless outer reinforcing braid. Generally known as “braided line,” this can be used with many different types of plumbing fittings, one of the most common being AN fittings. This type of line system is regarded as one of the more safe and reliable ways of transporting high pressure fluid in non-rigid line.

It is important to keep in mind that different types of alcohols react differently with various seals. Please make sure whatever medium you decide to use in your system is compatible with the lines and fitting seals in your system.

Signals & Activation
Signals and Activation

Various methods can be used to activate and meter a WAI system. One of the most common is the “boost” signal from a forced induction engine. This can be done with a simple boost switch (an on/off signal) or with a progressive controller. Another popular method is to use the mass air flow (MAF) signal from MAF sensor if the car is equipped with one. This sensor measures the amount of air entering the engine and can provide a relatively good way of metering W/A into an engine. If an engine does not have a MAF sensor or, if you wish to inject precise quantities of W/A across a wide RPM band, you can use the injector duty cycle (IDC). This signal is tapped from the fuel injector signal, and can be extremely accurate from idle all the way to redline.

Functionality of Different System Designs
Functionality of Different Systems

The following section will explain 4 types of WAI systems available today.
1) Single stage
2) Two stage
3) Progressive Pump Speed system (PPS)
4) Pulse Width Modulation (PWM) Valve controlled system
There are also derivatives of each of these systems, including systems controlled by the engine management computer and direct port systems. Generally only the PWM system is controlled by the engine management system, usually as an extra injector. Any of these could be used as a direct port system, but usually only the more sophisticated systems are taken to that complexity.

The Single Stage System

The single stage WAI system, contrary to what the heading implies, is not as basic as most people expect. In some cases, it will outperform a two-dimensional progressive system. It should not be underestimated in its capabilities in both performance and ease of tuning.

The single stage system, having a single trigger point and a fixed flow rate, is probably the simplest of systems available. Due to its consistency and repeatability, it is very easy to tune. This type of system is normally set to start spray in the peak torque region, usually mid-RPM range, where the engine is most likely to knock.

As the RPMs climb in an engine, with a single stage system, the mass of W/A to air will decrease. This means a greater percentage of W/A is being injected in the midrange and will tend to lean-out as the airflow increases at higher RPM. This is not always a bad situation depending upon the application. The tendency to knock is generally the greatest in the midrange where peak torque occurs. There is still a fair amount of W/A being injected at higher RPM, where the tendency to knock is less. On supercharged and turbocharged engines, IATs can climb in the upper RPM range, and this injection up top can help mitigate this. The exception can be some heavily modified engines where greater amounts may be needed in the upper RPM range due to increasing boost or modified timing maps.

FOR:
1) Low cost, simple and dependable.
2) Easy to tune due to its consistency.
3) Very effective on a stock factory set-up or moderate modifications.

AGAINST:
1) Dynamic operating range is narrow, may not be as effective on high RPM knock suppression.
2) For high power or high percentage alcohol applications, considerable fuel has to be taken out of the tuning maps to make the AFR tolerable. Some sort of failsafe mechanism is necessary to prevent engine destruction when the WAI fails to deliver the correct flow.

The 2-Stage System

The definition of a 2-stage system is simply adding a second manifold pressure switch (or secondary triggering mechanism) and second solenoid valve. This system gives greater flexibility than the single stage as well as extending the flow range in the upper RPM range. This solves a problem that can occur on the single stage system of too much injection in the midrange or not enough when the RPMs climb.

As this type of system is usually triggered by a boost mechanism, it still won’t address any RPM related flow issues. For a supercharged engine, the boost curve is non-linear with RPM, but heat generation somewhat is, so certain dynamic parameters are difficult to meet but better addressed with a 2-stage system. For a turbocharged engine, the most significant regions of the RPM band to address are the boost ramping stage and the engine’s maximum torque range. A 2-stage system can fit these two regions nicely, allowing some form of cooling demand during the ramp-up stage; the second stage provides the in-cylinder cooling and knock suppression as the engine is under the most stress during peak torque generation.

The 2-stage system, like the single stage, is very predictable in its operation and easier to tune than some progressive systems.

FOR:
1) Relatively low cost to give a significant improvement over the single stage system.
2) Provides well defined triggering points during the boost cycle.
3) Minimizes overflow in the midrange and underflow in the upper RPM region.

AGAINST:
1) Trigger points require some time to set up.
2) Triggering points may differ on each gear if you have a fast spool up turbo or quick boost generation.
3) Requires a bit more care during tuning than the single stage system.

Comparison of Predicted Performance of Single and Two Stage Systems

Progressive Pump Speed System (PPS)

Many of the systems on the market today are pump speed based designs. Due to their popularity and some operational complexity, we will spend a little more time looking into their operation. The PPS systems are popular due to simplicity and cost. But, does the pump speed system perform better than a 2-stage system? We will lay out the facts of the system and you can make your own decision.

Progressive Pump Speed systems are designed around, as the name implies, changing the speed of the pump to change the injection amount of the W/A. This design is simple and inexpensive to manufacture, as all that is needed is a pump, electronic controller and the usual fitting needed by all systems. The simplicity of the design is also one of the system’s biggest drawbacks. This is due to the laws of fluid dynamics which dictate the relationship of pressure and flow. Looking at the nozzle chart we introduced earlier, the pressure needs to be increased 4X to increase the flow by 2X. That is, for an M5 nozzle in the chart, the pump would need to span 40-160psi to achieve a change in flow from 200cc to 400cc of flow. One would be led to believe from the marketing of system manufacturers that the pressure-flow relationship would be 1:1, but it is not. This is a best case scenario for the PPS system, and is not quite as progressive as the name would suggest. We will look at other parameters that will show how this best case scenario is difficult to achieve for the PPS systems.

There is one major point about these systems to keep in mind. As mentioned earlier, almost all systems on the market today use a Shurflo manufactured pump. Shurflo pumps are rated only to operate to 150psi. Any system manufacturer that is claiming to operate the pumps up to 220psi or even 250psi are using these pumps outside their design and operational parameters, with sometimes minor or significant consequences. These manufacturers are running the pumps at elevated levels to try to gain more dynamic range out of their systems. The increased pressure is not needed functionally as the nozzles will provide the same atomization at 150psi as they will at 220psi. Obviously, boosting the range from 40-160psi to 40-220psi gives their systems greater range, but at what cost? We will come back to this subject later.

The heart of PPS systems is an electronic motor speed controller, varying the speed according to an input. The input could be a MAP sensor, a MAF or any other input. It is normally a 2-dimensional system. A manifold-pressure (boost) triggered system does not take into account where the motor is in the RPM change, it merely reads boost and injects accordingly. Boost, especially in supercharged cars, and smaller, quick spool turbos, is far from linear with RPM. In fact, the boost profile can be very flat compared to RPM. This can lead to over injection in the mid-range and under injection on the top end despite the “progressive” nature of the system. If the controller is set to compensate the other way, that is, injecting the right amount in the upper RPM range, the injection will drown out the power in the midrange. Both of these scenarios are less than desirable for making power across the entire RPM range. Sometimes with turbo systems, boost can peak earlier in the rev range and taper off slightly as the motor reaches redline. In this case you would be tapering off the injection just when you may need it most. This may have been where the term “kaboom” originated.

Let’s look at some of these factors in play and how they affect a real world setup.

Example: A 150psi pump, a turbo spooling to 20psi, and a 30psi checkvalve.
Pump Maximum Pressure = 150psi
Manifold Pressure Start = 10psi
Manifold Pressure Ends = 20psi
Inline checkvalve back pressure = 30psi
Minimum pressure to atomize at nozzle = 30psi

This, and every checkvalve equipped system, will always start out handicapped by the checkvalve back pressure. This is in addition to the required nozzle atomization pressure, which is the pressure required to actually produce a spray instead of a dribble. This equals 30psi (checkvalve) + 30psi (required to atomize at nozzle). So the beginning of the dynamic range is 60psi. The top end of the dynamic range is limited by the checkvalve choking the pump and the additional 10psi of boost in the system. This limits the top of the dynamic range to 110psi. This leads to a mere 35% change in flow from the PPS system. From the beginning ideal of 200cc to 400cc range on an M5 nozzle, we are now limited to approximately 240cc to 325 cc, far from what one is lead to believe from a “progressive” system. From some manufacturers’ literature, one would be led to believe that twisting the knobs on the controller would handle the range needed from, let’s say, 5-20 pounds of boost. This is definitely not necessarily so. This limited dynamic range is illustrated in the chart below.

There are also many other factors that could affect the performance of the 2-D progressive pump system; in fact, too many to look at in depth. Let’s look at the predicted performance chart from before, with the PPS system added. At first glance, it doesn't appear there is a distinct advantage for adding a progressive controller. Adding a bigger nozzle doesn't expand the dynamic range, it just shifts the entire curve higher.

As you can see in the graphic above, the PPS system has not given the user any significant dynamic range increase.

Many PPS systems use a checkvalve for stopping dribble and siphoning at the injector. A few systems do use a solenoid for this, but we should look at the effects of the checkvalve on the system. This is shown below in a graphical example.

As can be seen from this graphic, the system starts to inject at 12psi and reaches it max value at 24psi. The dynamic range appears to be from 114cc to 325cc, but the nozzle doesn’t produce a decent spray until 30psi positive pressure at the nozzle. This means the actual dynamic range is from 176cc to 325cc’s. This is only an 84% change in flow, less than half of the ideal we looked at earlier.

As we have shown, a checkvalve can be somewhat detrimental to the operation of a PPS system. Let’s look at all of the positives and negatives:

Positive effects of a checkvalve:
1. Retains some pressure in the line to compensate the next injection event. A 20psi loaded checkvalve will keep 20psi of pressure in the line after injection.
2. Stops W/A from being siphoned into the engine if the water jet is installed in the vacuum side of the manifold.
3. Prevents emptying the entire tank into the inlet tract if the tank location is higher than the jet (gravity fed) or the car is parked on an incline.
4. Stops some dribble after an injection event. Even when the power of the pump is switched off, the inertia of the rotating mass keeps the pump running for a second or so.

Negative effects of a checkvalve (not always published):
1. The presence of a checkvalve has a very significant impact on the dynamic range of a 150psi progressive pump speed system. A 20psi inline checkvalve will instantly drop the 150psi PPS system down to a 130psi span.
2. A normal nozzle requires approximately 30psi to produce a decent atomized spray. An inline 20psi checkvalve means the pump has to produce 50psi to produce a decent spray. Add to this the boost pressure, say 10psi start and 25psi end, and the dynamic range is now from 60psi to 105psi.

Some PPS system makers are now offering an inline solenoid upgrade so that the dynamic range is improved by a good margin. This probably should have been engineered into the design and included from the beginning. As noted above, many PPS systems run their pumps up to 200-250psi in an attempt to gain more dynamic range. Shurflo’s literature clearly states the 8000 series pumps are only rated to 150psi, and in fact should only have a 2psi or less checkvalve installed in the line. But, ideal vacuum is -14.7gauge psi, so what is a WAI manufacturer to do? The apparent answer is to go outside the manufactures specs and install at minimum a 16psi checkvalve. As they say, your mileage may vary.

An issue with the PPS systems is response time of the pump. When asked for an immediate spray, there can be a delay. If running this system on the street or at the track where there are quick throttle and load changes, the pump’s inertia will hinder its response. This may cause it to have a lag in spray response time when quick throttle changes are applied. If using this system for top speed runs or drag racing this issue is much less of a problem.

If using a demand switch with the PPS system one must also be aware that as soon as the pump reaches full speed, near max spray, the spray will begin to pulse as the pump runs up against the demand switch pulsing power to the pump. This can lead to irregularities in the spray patterns at heavy loads and high in the RPM range.

In summary, the dynamic range of the PPS system is quite limited in most cases. Even with the ideal case of 200cc to 400cc, this only accounts for say a doubling of RPM from 3000RPM to 6000RPM at a single throttle load. Start changing the throttle position or run outside the RPM range, which we know is much more limited than the 3000 to 6000 mentioned above, and you can have tuning and consistency issues. This combined with the pump inertia issues and pulsing at high RPM limits the effectiveness of this type of system.

FOR:
1) Easy to set the start and end point.
2) Some correlation between manifold pressure and flow.
3) Cost effective.

AGAINST:
1) Limited dynamic range.
2) Extra cost can easily be spent on a higher performance two-stage system with greater dynamic range.
3) If the PPS system is used to replace high % fuel with alcohol, re-mapping the EMS 3-D fuel map will be very difficult due to the wide dynamic flow range demanded by the engine and limited dynamic range offered by the WAI system.
4) Pulsing due to demand switch is approximately a 20psi ripple (some systems bypass this switch, but doing this risks running system pressure beyond design limits).
5) Slow response time due to inertia of a rotating motor mass - lagging (start) and over-run (stop).

Pulse Width Modulation (PWM) valve water injection system

Pulse width modulated systems are based upon using variably actuated inline solenoid valves to inject W/A into the intake tract. The actuation of these valves can be triggered the same as in the previously discussed systems, such as boost or MAF, but also adds another option. The PWM system can be set up to act nearly exactly like a fuel injector, metering a precise amount of W/A into the engine. This mirroring of the fuel injector means they can deliver precise amounts from just off idle all the way to redline. These systems require a stable system pressure, normally held between 100-125psi. For this reason, they usually, but not always, use a pump with internal bypass valves instead of the on-demand switch. Before getting too deeply into the subject, note that there are two types of inline solenoid valves available and they have drastically different effects on performance.

Type #1
Pulse Width Modulation Type (Optimum operating frequency range: 30-80Hz)

This type of system resembles the modern automotive fuel injection system. The system can also be controlled by a third party engine management system (EMS) with a spare PWM channel. Delivery rate can either be mapped by an EMS or be set to mirror the fuel injector duty cycle. The latter makes tuning very simple and effective system.

This type of valve behaves similar to an on/off switch on a high pressure garden hose. Think of squeezing the handle very quickly on a hose nozzle. You get an immediate and strong stream. The longer the handle is squeezed, the more the flow (duration). Alternatively, rapid opening and closing (or pulsing) of the handle, per second (frequency), also controls the flow. The common EMS uses duration for load change, and frequency for RPM change. The dynamic flow ranges of these systems are extremely wide; 100:1 is normal. This is magnitude better than the 35% increase in flow we saw in one of the PPS examples.

A PWM WAI valve should closely match the opening and closing characteristics of a fuel injector. This is important for fuel flow mirroring algorithms since the modern EMS has a correction factor to compensate for the opening delay and closing delays inherent in mechanical systems.

Below you will find a graphical example of how this system injects.

Type #2
Proportional Lift Type (Chopped DC (~400Hz) or DC current)

The proportional lift type of valve resembles the action of turning the spigot on and off with the handle pulled on your nozzle 100% of the time. As more current is applied to the valve coil, the valve, or spigot, opens more. It is a very nice way to control flow, just not ideal against a nozzle in a dynamic system. There are a few minor problems associated with these types of valves: Atomization at low flow, and lift variations (hysteresis of the magnetic circuit), which can lead to approximately +/- 10-15% flow deviations. As a result the repeatability of the system is usually no better than 10%, sometimes worse, which can make trying to tune it to an engine very difficult.

All in all, the proportional valve will deliver liquid well compared to the PPS system. There are some similarities between the two. The nozzle tip pressure is directly proportional to the flow. This is because the proportional valve acts like a variable restrictor upstream of the nozzle tip. This has a significant result; restricted flow = low pressure, low pressure = low atomization.

Below you will find a graphical example of this system.

It is important to know some basic differences between the proportional valve and PWM valve systems before choosing a system with either. The proportional type of valve has a typical open time of 4ms and closing time of 4ms at its designed voltage. It is too slow to be used as a true PWM valve. The open/close speed can be increased by over-voltage pulses, but again running components outside their designed range can have consequences.

Here is an illustration of the difference in construction of the two valves, made by Clippard, a U.S. manufacturer of valves. Notice the proportional lift has a stiffer spring rate than the PWM valve, enabling the PWM valve to perform full on/full off a great deal easier.

Due to the way the PWM system meters its flow, based on a simple pulse width, it is very accurate. Further precision can be increased by introducing a suitable rising rate fuel pressure regulator (RRFPR), similar to units in common fuel injection systems, to maintain manifold pressure against water pressure. I t is also possible to factor in a small duty cycle increase to the valve relative to a boost increase.

Final consideration: If you are planning in the future to create your own injection map via a third party system, ONLY the "PWM-valve" can be driven directly by the ECU, matching the principle of a modern "fuel injection system" in every respect.

[NOTE: If you have questions or comments regarding this article, please visit this discussion thread in the Experiments :: Prototypes :: Ideas forum.]
Failsafes
Failsafes

Having discussed most of the core functionality of the systems, it is a good time to consider the safeties, and essentially how much risk you could be putting your engine in depending upon the tuning. In order for a WAI failsafe system to be effective, it has to be a standalone system covering multiple scenarios. Otherwise, it should not be advertised as a failsafe but merely an added feature.

The minimum requirements for a failsafe to be effective are detection of:
1) Blocked jets, either complete or partial.
2) Pressure drop in the system.
3) Any flow discrepancy in the system in real time.
4) Any over-flow condition (due to leaky connections or a broken line).
There are two methods of detection in the system, direct type and indirect type. The direct type may be able to have a safety effect in real time. The indirect type will usually detect the fault some time after it has occurred. Some of the direct detect methods of faults are listed below.

1. Fluid tank level sensor:
Advance warning only, but little use for system power failure or blocked nozzle.

2. Loss of system power:
If the system power is interrupted to the WAI system, it will de-energize a relay and perform "Plan B." This will not detect a blocked jet, cut pipe or lack of fluid flow.

3. Inline pressure gauge:
This is very cost effective but requires user attention.

4. Inline single pressure switch:
If the system loses fluid pressure during an injection event, an inline preset pressure switch can activate a boost dump, map change, etc.

5. Two inline pressure switches:
This method can detect a blocked nozzle (over pressure) and "loss of fluid pressure" (under pressure).

6. Flow switch:
An inline flow switch, comprised of a spring-loaded magnetic plunger inside a tube the water flows through. The rate of spring determines the proximity of a reed switch against the plunger deflection. The output of this is an on/off signal and the cost is low.

7. Turbine flow sensor:
Good for a progressive system but requires complex electronic circuitry to report a difference between the "actual" and "planned" flow in real time.

Next we will list some of the indirect detection methods.

1. Knock detection:
This is often covered by the engine management; upon onset of knock, the ECU will retard timing to reduce peak cylinder pressure. However, if your engine is tuned beyond the scope on the engine's ability to retard timing range, it will not help. Effectiveness of this failsafe is confined to a mildly tuned engine.

2. AFR tracking:
AFR tracking is not well covered by the stock engine management under WOT conditions. This is because most cars are equipped with a narrow band lambda sensor. You need to consider installing a Wide Band lambda sensor with a settable trigger output. Also, the more water you are running in your mixture, the harder it will be to detect a discrepancy in flow through the AFRs.

3. EGT tracking:
The engine's combustion temperature can normally be estimated by exhaust gas temperatures. Unfortunately there are other factors that will increase the exhaust gas temperature:
a. Retarded timing by the ECU .
b. Retarded burn-rate caused by slower frame speed caused by additional injectant such as water, methanol and fuel.
c. Ideal AFR can also cause increase in EGT due to efficient burn rate.
d. Over lean AFR also promotes EGT rise.
This method of failsafe can be good, but overall it can be difficult to pinpoint.

4. Inlet temperature tracking:
If your car is equipped with a third party ECU, you can place another air-temperature sensor downstream of the water jet to warn of the loss of W/A injection. You need to create a "high" air-temp trigger failsafe output.

To summarize, it important to know how "safe" is a "failsafe." Ask many questions, it’s your engine. It is my opinion that many failsafes are called so and really are not. I will leave this topic with a few thoughts for you:

• Many systems today have a failsafe that will give a +12v signal upon a fault. If the system loses power (blows a fuse) at the same time, how are you going to get a signal from it?
• A failsafe is only as good as what it does to protect your engine from damage if it fails. Annunciation (an LED light) or reading a gauge, is only half the battle. A true failsafe will open a relay contact upon fault, so even a power failure is caught.
• Another scenario: What happens if the system begins to leak under pressure and only supplies a portion of the required flow to the engine? Only a couple of the options listed above have a chance to catch that.
• Some systems on the market claim to monitor flow through the system, even with a digital gauge to show the flow. They are actually measuring applied pump voltage (most PPS systems) which equals pump speed. The system could be running dry or have a nozzle blockage and it would still say you were injecting liquid! Even if the controller could monitor pump amps at the same time, if it is an on-demand system there’s no hope of catching anything due to the amp spikes.