We've covered the concepts and we've covered the spray hardware - so what's this gee-whiz controller all about, anyway? Shown above during testing, we reckon that the Labtronics/AutoSpeed intercooler water spray controller is - ahem! - the best DIY controller in the world. It's easy to wire in to place, simple to calibrate - and in bare bones form, nearly as cheap as some pressure switches! However, it's a helluva lot more sophisticated than a switch, that's for sure.
The control system that AutoSpeed and Labtronics have developed has a number of unique features. It:
- adaptively monitors how hard the car is being driven;
- senses the temperature of the day;
- senses the temperature of the intercooler core;
- and measures how effective the water spray actually is at reducing intercooler core temp.
The result is an intercooler water spray controller that copes intelligently with almost every condition we can think of. In fact, during testing, at times its operation was almost uncannily good. For example, at high speeds the spray duration is typically much shorter than at low speeds - this is because when you are going faster, the intercooler stays cooler because of the greater airflows. If it's raining, the splashes of water on the intercooler keep it cooler - and so the spray operates far less frequently. Another example - by watching a LED on the electronic module, you can actually see the intercooler heat-soak building up at traffic lights, and also the rise in intercooler temp that often occurs after a boost event. (For more on this intercooler behaviour, see Part 1 of this series).
You soon realise that the times when the intercooler is hot (and so needs the spray to operate when you start driving hard) and when it is cold (no spray needed) are not at all obvious to the driver behind the wheel... let alone to something as primitive as a boost switch.
The Brain
In order to have effective adaptive learn characteristics, the AutoSpeed / Labtronics intercooler water spray controller has to be able to realise when the driver is going for a hard drive. Then, when it figures that the driver is really going for it, the controller keeps the spray on during high loads and during trailing throttle and gear changes. In this hard-driving situation it also switches the spray on very early.
The controller decides that you're going for a Fang by monitoring injector duty cycle - the higher the duty cycle, the greater the engine load.
Don't understand what injector duty cycle is? See the first breakout box.
So why take the approach of monitoring injector duty cycle? There're a couple of pretty good reasons:
- no expensive boost pressure sensor is needed, so the hardware is cheaper;
- multiple inputs from the airflow meter, throttle position sensor and so on are already incorporated into this signal;
- if required, the spray can be triggered well before boost even starts to occur;
- the system is as happy working on cars with programmable management as well as all types of factory EFI.
So why haven't other people used injector duty cycle as the load input on aftermarket controllers? One major reason is that digitally monitoring injector duty cycle in real time is incredibly demanding - Miroslav Kostecki (Labtronics' Chief Engineer) had to do some pretty trick programming of the PIC chip indeed....
While injector duty cycle is used as the major input in deciding when the driver is going for a fang, as you'll see, the controller does a lot more than just switch on the pump at a certain duty cycle. So how does the Fang Factor actually work?
- The Fang Factor is initially enabled when a user-adjustable injector duty cycle is exceeded. For the sake of this example, let's say that the owner has set the pot that controls this trip point to a sensitivity that corresponds to 25 per cent injector duty cycle. (Note that you don't actually have to measure duty cycle to set the sensitivity - you just twiddle a pot until a LED comes on whenever you're driving hard!)
- For every second that the measured duty cycle exceeds 25 per cent, the Fang Factor trip point is reduced by 2 percentage points (to a minimum of 0). This means that the longer that high loads are used, the more sensitive the system becomes to being tripped.
- Okay, but how does the system ever get back to its 25 per cent switch-on point? What happens is that every second that the measured duty cycle is below 25 per cent, 2 percent points are added to the Fang Factor trip point (up to a maximum of 25 in this example). So, at low engine loads, the system is constantly trying to get back to its original setting.
What this all does is make the Fang Factor trip more and more easily as you drive harder and harder. Sound good? It is!
However, there's no point in operating the water spray if in fact the intercooler is not even hot. And the core - acting as a heatsink - won't significantly rise in temperature if boost is being used only for a short burst, or if it's a cold day and the intercooler heat exchanger is so good that it's getting rid of heat as fast as it's being added. To monitor the actual intercooler behaviour, there are two temperature inputs to the controller - one monitoring ambient (day) temperature, and the other, intercooler core temperature. When the intercooler core temp exceeds the day temp by a user-definable amount, the Temp Factor trips.
Only when both the Fang Factor and the Temp Factor are tripped, will the pump switch on.
So that you can see easily what's happening, the status of both the Fang Factor and Temperature Factor can be monitored by watching two LEDs on the control module's PCB. By looking at these LEDs, you can actually see when the controller has decided that you're driving hard, and also when the intercooler temp is starting to rise significantly above the day temp.
Checking the status of these LEDs also lets you easily set the two sensitivity controls - one for Fang Factor and the other for the Temperature Factor.
And there's one other tricky function that also beavers away in the background. During testing it was found that while the spray switched on with great accuracy (ie only when it was really needed), when the boost event was over, it tended to stay on for periods that didn't accurately reflect how hard the car had been driven. To overcome this, we decided to use the temperature sensors to tell the controller how much above ambient temp the intercooler core was at the end of the boost event, and then set the delayed spray on-time from this input. In other words, if the intercooler is still hot after the spray has been operating, it stays on a bit longer. If it's cold, it switches off straight away. How long it keeps running depends on how hot or cold the intercooler is and also on the position of the Temp Sensitivity pot.
All Too Hard!
All sound really complicated? Totally confused? Sure, the internal logic may be complex - but fitting the controller to a car and setting it up is child's play!
There are just two control pots - Fang Sensitivity and Temperature Sensitivity. With the system wired in, you go for a drive. An assistant turns the Fang Sensitivity knob until the Fang LED lights up only when you're driving hard. You get out and feel the temp of the intercooler. If it's at the temp that you want the spray to come on, turn the Temperature Sensitivity pot until the Temp Factor LED just lights up. There - you've finished doing the calibration...
As for the wiring - there're just the two temp sensors (which connect straight to the module), one wiring connection to an injector, power, earth and the relay for the pump. That's it. There are even green LEDs on the board that light up to show you when the right connections have been made - easy!
Next week we take you through the step-by-step fitting and calibration procedures.
Intelligent Intercooler Water Spray - Part 1
Intelligent Intercooler Water Spray - Part 2
Intelligent Intercooler Water Spray - Part 4
Intelligent Intercooler Water Spray - Part 5
|
Injector Duty Cycle? WTF?
While it sounds really complicated, injector duty cycle isn't too hard to understand. The duty cycle is the proportion of time that the injector is open, squirting fuel. So at idle, the injector duty cycle is small - often only around 2 per cent. That means that the injector is open for only 2 per cent of the time - or, if you like, that it's shut for 98 per cent of the time. At moderate loads, the injectors need to provide more fuel, so they're open for longer duty cycles - say, 15 per cent.
But there's one trick - as engine revs go up, so the available time for the injector to be open decreases. (ie you still need at least one squirt per intake stroke - higher revs give you less time to get in that squirt.) As a result, as revs and load rise, injector duty cycle really starts to get a move on. At peak power and revs in a typical unmodified engine, the duty cycle of the standard injectors will be somewhere around 80 - 90 per cent. In modified engines still running standard injectors, hitting 100 per cent (ie they're open all of the time!) is not uncommon.
|
|
Control Strategies - Really Nerdy Stuff!
Once injector duty cycle was picked as the primary control input, a great deal of thought was given to how this information should be used. Here's a summary of some of the other approaches considered but then rejected.
Control Approach |
Control Details |
Advantages |
Disadvantages |
Threshold Switching |
Pump switches on when duty cycle is greater than a preset threshold (eg 30 per cent). Pump switches off when duty cycle is less than a preset threshold (eg 25 per cent). In this example, there is a hysteresis of 5 per cent. |
- Simple
- Early switch on
- Hysteresis means no problems of rapid pump cycling
|
- Pump switches off on gear changes and trailing throttle, even when driving hard
- Will operate even with a single burst of power
|
Acceleration Rate and Threshold Switching |
Pump switches on when rate of change of duty cycle exceeds a pre-set threshold (eg when duty cycle rises by more than 10 percentage points per second). This would occur during hard acceleration. Pump switches off when duty cycle is less than a simple threshold (eg 25 per cent). |
- Very early switch on - picks a fang happening almost immediately you leave the line!
|
- If set to be sensitive in lower gears, does not trigger in other gears
- If set to be sensitive in high gears, triggers when not needed in lower gears.
- Will operate even with a single burst of power
- Pump switches off on gear changes and trailing throttle, even when driving hard
|
Added Simple Delay |
Once pump is triggered, it continues to work for a period (eg 10 seconds) after it would normally be turned off. |
- Pump stays on during gear changes and trailing throttle
|
- Wasteful of water - delay always occurs
|
Moving Time Window Average |
The duty cycle is sampled rapidly. An ongoing average is calculated based on the most recent sample and the average of all of the previous samples. When this average is greater than a preset threshold (eg 30 per cent), the pump switches on. When it is below this threshold, the pump switches off. |
- Early switch-on
- Pump stays on during gear changes and trailing throttle, if the duty cycle average of the power squirt has been high
- Conservative use of water
|
- A long sampling time improves 'intelligence' of system but delays the initial switch-on
|
|
|