Sensors (Wolf3D)

[RTz]

This is the first of several articles aimed at aiding the installation of a Wolf3D EMS. Its not intended to be ‘all encompassing’. Rather, a simple guide to get things jump-started. I will add/update as time permits.

 

What are the Minimum Sensor Requirements?

 

For a fuel only system, you will need a minimum of...

 

RPM sensor (trigger)

Load sensor

Water temp. sensor (WT)

 

 

 

With those three sensors, you can successfully run any typical gas engine. Later I’ll cover adding supplemental sensors to achieve a better overall package.

 

Summarizing the Three Sensors...

 

 

WT (Water Temp.)... The primary function of this sensor is to aid in cold running. When the engine is ‘cold started’, a richer than normal mixture is needed for best operation. As the engine warms up, fuel can (and should be) reduced. Its a bit like an electronic version of the choke.

 

RPM... A common name for this sensor is ‘trigger’. Trigger’s can be very simple, or very complex. This is an entire topic on its own and outside this articles scope (I’ll cover some basic triggers in another installment). A fuel only system is simple enough to require only a very rudimentary trigger. For example, the stock Datsun electronic ignition distributor contains a VR sensor (Variable Reluctor), with six teeth. This trigger can only tell the ECU when a cylinder is approaching, but not which cylinder. This type of trigger would only allow batch-fired injection (and distributor based ignition).

 

Load... There are a number of ways to determine load. The three most common are...

 

MAP

AlphaN

MAF

 

Briefly....

 

MAP (Manifold Absolute Pressure)... is most commonly used and is sometimes referred to as ‘speed density’. It measures the pressure in the manifold to determine load... high pressure=high load. Its advantages are simplicity and accuracy under most conditions, with a natural tendency toward altitude and barometric pressure compensation. Its biggest drawback is its inability to compensate for modifications or engine wear. In summary, anything that changes the airflow of an engine under any given condition without a corresponding change in manifold pressure causes an inaccuracy. Note: a 3-bar MAP sensor is built into Wolf's ECU.

 

AlphaN... uses a throttle position sensor (TPS) to determine load. Like MAP, it has no ability to compensate for modifications or engine wear. Additionally, there is no barometric pressure compensation. You may want to add a barometric pressure sensor to compensate for altitude and barometric changes. Another pitfall is that pure AlphaN will not work with a turbocharged engine. A TPS is unable to distinguish the difference between 2 psi or 20. AlphaN’s advantage is with aggressively cammed NA engines (making around 10” vac. or less). In short, most hot cammed engines vacuum signal is unstable at idle. With a MAP based system, this causes erratic cell movement at low rpm/light load (i.e. idle), within the fuel map (see footnote for a sample fuel map).

 

MAF (Mass Air Flow)... uses a thin piece of heated wire. As air moves across the wire, the wire cools. This in turn, lowers the resistance of the wire. The ECU uses the resistance to calculate the mass of the air drawn into the engine. MAF is one of the most accurate methods currently available. Its primary drawbacks are cost and installation complexity.

 

 

 

Footnotes:

 

Highly Recommended Sensors... I mentioned earlier you can add sensors to enhance the overall performance. The two that should be highest on your list should be a TPS and IAT.

 

A TPS, optional on MAP/MAF based systems, is capable of doing a number of things, but its most valuable function is acceleration enrichment. The MAP/MAF sensor can serve the same purpose, but its a little slower responding, so achieving crisp throttle response is a bit more challenging. There are a number of other TPS benefits such as fuel cut, flood clear, data acquisition, and auxiliary activation... we’ll get to those later.

 

The IAT (Intake Air Temp.)... As the temperature of air increases, its density decreases. We want to match the volume of fuel to the mass of air, not its volume. The IAT is part of the means to compensate for air density changes.

 

Ignition Timing Control... For a system that includes ignition timing control, the above mentioned list meets the minimum requirements for a distributor based system. If you want to run a multi-coil ignition, a more advanced trigger will be necessary.

 

Fuel Map... The fuel map is a 3 dimensional graph expressed as numbers (as is the timing map). Across the top, you have RPM (in 125 rpm increments). On the left are load bands (in approx. 7% increments). The numbers in the cell’s are injector pulse-widths (‘on time’ in milliseconds)....

 

 

 

 

Same map, in graphical form....

 

 

 

[RTz]

It seemed wise to start with the least understood sensor... the trigger.

 

Wolf is very versatile in this respect. The advantage is that you have the latitude to be very creative with the type of trigger you use. The downside is that making sense of all the choices can be a little daunting.

 

It would be impractical for me to list all the possible permutations, so I'll try to teach you how to 'fish' instead.

 

There are three primary means of triggering...

 

1) Crank Trigger... Allows for very precise timing. Its main shortcoming is that while it knows when a piston is at TDC, it doesn't know whether its TDC on the compression stroke or exhaust. This means you can use a multi coil ignition, but only in in a wasted spark configuration, and only batch-fired injection.

 

2) Cam Trigger... Can be used to run fully sequential injection and ignition. Its drawbacks are usually design complexity and the potential for timing shift as the belt/chain stretches or displays instability. However, the stock Datsun distributor (electronic ignition) contains a VR sensor and a trigger wheel with 6 teeth on it. This qualifies as a cam sensor because it is geared to the crank 2:1, just like the cam. In other words, the distributor drive 'mirrors' the cam.

 

3) Crank+Cam (Reference + Sync)... The accuracy of a crank trigger with the power of a cam sensor. In this scenario, the cam sensor is usually only one tooth. Its only job is to tell the ecu which event the cylinder is on and has no impact on actual timing.

 

Each of those three can have a variety of tooth counts, sensors, and 'types'...

 

Tooth counts.... Commonly 4,6,8,12,24,36,60. The simplest of these is to run a count equal to the number of cylinders for a cam sensor, or half the number of cylinders on the crank, i.e. 3 teeth for the L6.

 

Types...

 

Single pulse... A single pulse trigger is the most rudimentary. An example is a 3 toothed crank sensor, equally spaced (120 degrees), on the L6. This gives the ECU just enough information to establish RPM and when a piston is approaching, but not WHICH piston.

 

Dual pulse... An example of a dual pulse sensor would be to add one more tooth immediately after one of teeth of the single pulse sensor, making it a '3+1' trigger wheel. Now the ECU can determine WHICH cylinder pair is approaching TDC. I say 'pair' because, for example, cylinder one and cylinder six reach TDC at the same time. If you utilize a dual-pulse CAM sensor, there would be 6 primary teeth with one 'dual pulse tooth' (6+1). With a 6+1 Cam sensor, the ECU can tell the difference between one and six. This is the key to having fully sequential ignition and injection.

 

Example Dual Pulse trigger, rotating counter-clockwise. There are 6 primary teeth with a 'sync' tooth just after a primary...

 

 

Missing Tooth... Same potential as the dual pulse trigger, but there are normally a larger number of teeth... Common counts would be 36 (making it a 36-1 wheel) and 60-2. I probably wouldn't recommend anything less than a 12-1.

 

Example missing tooth, 36-1 crank trigger....

 

 

Pick Up Sensors....

 

VR.... Variable Reluctor's are generally inexpensive and readily available. They are 'passive', meaning unpowered. They have only two wires. They can be used with very small teeth and separation between teeth, so they work well in compact applications. They produce a sine-wave, increasing in voltage with increasing RPM.

 

 

Hall Effect... Generally more expensive than VR's. They are powered sensors, with three wires and normally operate at 5 or 12 volts. They produce a square-wave (on-off). The fixed voltage makes things easy, but their biggest drawback is that they need room. Hall sensors don't discern small, closely spaced teeth very well. I would normally reserve the use of a Hall effect to a crank trigger (because diameter is usually generous) or a single tooth cam sync.

 

 

Optical... Most often the most expensive, it still remains my personal favorite. Normally produces a 12 or 5 volt square wave ('on or off'), and can be quite compact. In my experience, its nearly faultless.

 

 

As you can see, there are pro's and con's to each.... so what's the best choice for your application? Depends on your needs, but in terms of the best L6 trigger for the least amount of effort, I prefer to use an optical trigger from a late 280ZXT. Its nearly a bolt-in. It requires only that you also use the quill shaft that drives it (its splined instead of keyed like the NA L28). It has six slots, one for each cylinder. This puts in in the 'single pulse cam sensor' category. If you use it in unmodified form, you can run batch injection and distributed ignition. If you drill a .100" hole, .060" to .080" AFTER one of the slots, it now becomes a 'dual-pulse cam sensor'. This means fully sequential injection AND ignition are available. It takes all of 30 minutes.... if you take your time. Here's a picture of the modified disc....

 

 

Note: that disk is actually from an Z32. The only difference is that all 6 slots are the same width on the 280ZXT.

 

 

The exact location and shape of the hole is not terribly critical. Its only there to provide a 'sync' so the ECU knows which slot is which cylinder. I've personally tested this sensor to over 7500rpm, without trigger errors. I know others have run them over 10,000 rpm, without error. What's wrong with it? Outside of being a little bulky, you'll probably want to modify the distributor cap to get rid of the unused HT outlets. One way is to simply cut them off and fiberglass over the top. Here's a picture of one with a billet aluminum 'cap'....

 

 

 

 

If you're wondering about backlash in the drive gears... I installed a cam and crank trigger on an L4 (same drive mechanism as an L6), attached a dual channel oscilloscope, and ran the engine, while recording the traces. Comparing the traces showed +/- one degrees 'scatter'' about 90% of the time, with a maximum of 2 degrees, occasionally. Not Formula 1 consistency, but quite good for most of us.

 

Note: One 'can't do' with a trigger is use tooth counts that are not evenly divisible with the cylinder count. For example, a 6 tooth wheel on a 4 cylinder... 6 divided by 4 = nonsense.... and that's how the ecu will see it. On the other hand, 12 divided by 4 would be 'logical' to the ECU. "But Ron, you said I could use a 3 tooth trigger on the crank?". I sure did. My reference is to cam timing in this case. Since the crank makes two complete revolutions for every cam revolution, a 3 tooth crank wheel is equivalent to a 6 tooth cam wheel. 6 divided by 6 = 'logical'.

 

By now, a few things should be evident...

 

1) No single pulse trigger will run multi-coil ignition or sequential injection. Batch injection and distributor based ignition is the so called limit.

 

2) A crank mounted dual pulse or missing tooth trigger, opens the doors to wasted spark multi-coil ignition but still must retain batch fired injectors.

 

3) A cam mounted (or distributor) dual-pulse (or missing tooth), provides a means for full sequential injection and ignition.

 

4) A Reference+Sync. trigger will also deliver fully sequential ignition/injection, with the accuracy of a crank trigger.