Category: Uncategorised

When your KTM 525XC Quad bike isn’t already bonkers enough and you’re looking for the versatility of EFI over carbureted then you obviously swap in the engine from a KTM 690. They’re both KTM, should be simple right?

For engine control, an ECU Master EMU Black was selected.

A housing was used to ensure that the ECU was protected from the harsh elements encountered in the desert and during racing. This enclosure houses the ECU, a fuse box, and a map switch. The map switch is used to switch the startup fuel, fuel target map, and ignition table within the ECU.

Most components are fairly plug and play, especially with a custom wiring harness being constructed, however finding an accelerator position sensor (APS) which was appropriate proved difficult.

We initially started with a Volvo APS, but it used a rather unique output system. As the accelerator position changes the voltage goes from 0 volts, up to 5 volts and then back down to 5 volts. Unfortunately due to this rather unique approach we couldn’t set it up to work with the ECU Master.

Next we looked at some OEM thumb throttles. These generally run a cable to the throttle body and use a single potentiometer to generate their voltage output. As the cable will physically pull the throttle shut in the event that the potentiometer fails this has an effective failsafe built-in. Unfortunately for us there weren’t any off the shelf alternative OEM solutions which would work with the packaging constraints that we had to work with.

In the end we turned to modification, using the original KTM ATV thumb throttle and creating a custom solution to allow it to accept the APS from a Yamaha YZF-1. The Yamaha APS was chosen due to its dual potentiometer design, meaning it has a failsafe built in. As this quad is going to be used for desert racing, the chances of dust ingress into the APS is far higher than on a road-going vehicle.

In order to fit the Yamaha APS a “key” was designed which converts the key on top of the axle within the KTM throttle assembly into the keyed hole in Yamaha APS.

The Yamaha APS uses a “following” type of signal, with both potentiometers going from 0-5V as the accelerator travel is increased. To use this style within the EMU Black we selected the Honda FN-2 check type for the failsafe check type.

To secure the “key” a 3D printed part was created to house the APS, and form a cover for the throttle. To mount this to the throttle M5 heat inserts were used. The part itself was designed in Fusion 360 with the base of the part created using a flat-bed scanner and the rest being added manually with caliper measurements.

The original throttle plate also got a 3D printed insert to stop dust and dirt ingress.

The 3D printed components were printed our new BambuLabs X1 C printer with the PT Motorsport logo being added using the AMS system.

The Haltech CAN Wideband Controller Kit is an all in one wideband solution with plug and play functionality with Haltech ECUs. After seeing online that you could add heater control functionality while using them with Link ECUs I decided to delve deeper into it and check whether this was really the case.

To give some back story, some other ECU brands also have cross-compatibility with the Haltech Wideband Controllers. Two of these are:

Link (G4+, G4X, G5)

ECU Master (Black)

At this point I had planned to have the instructions for how to set up the Haltech WBOs on each of the above ECUs. Instead of blindly trusting the internet, I decided that I hated having spare time and jumped into confirming whether the CAN heater controls actually worked.

Cue an absolutely giant rabbit hole of CAN sniffing, wideband sensors, and ECUs.

The original premise of this article was to let people use a Haltech WBO with other ECUs and retain proper heater functionality. I believed this to be a fairly simple endeavour involving a little bit of CAN setup. As I understood it, the Haltech WBO uses the engine RPM signal being sent via CAN to determine what the engine state is and to therefore control the wideband sensor.

With that in mind I tried the following setups:

The ECU Master Black software allows you to select the Haltech CAN Wideband controller straight from the Oxygen Sensor Menu

Once connected if you output a Haltech V2 CAN-bus dashboard stream the engine rpm will be broadcast on the correct CAN ID for the WB1.

The Link is slightly more involved. In the CAN setup menu, the Mode must be set to user defined. A channel must then be set to “Link CAN-Lambda” with the CAN ID set to 950.

With this done, the Link will receive the Lambda data from the Haltech, but the wideband heater will always be on when the ECU is on.

To change to a setup with an RPM lockout so that the heater is only on once the engine is started the following CAN channel needs to be added.

On a separate Channel, change the mode to “Transmit User Stream 1” with a transmit rate of 50 Hz and a CAN ID of 864.

Once this is done, a custom stream must be set up.

Add a Frame, make sure the Frame setup is as follows
Frame Size: 8
ID Position: None

Then add a Parameter and select “Engine Speed”

What next?

After doing the above on both a Link and an ECU Master Black ECU with a Haltech WB1 plugged in we found that there was no difference in how the controller behaved.

Upon turning the car on (without starting the engine) the wideband heater was immediately on (drawing 1.4-1.6 amps). This isn’t what we expected at all, so we went back to basics and did the same test with the WB1 plugged into a Haltech Elite 2500. Expecting a different result, the wideband behaved in exactly the same way, heating up immediately irrespective of the engine state.

In doing all of this, the CAN stream from the WB1 was sniffed as well to try and determine what state the wideband was in. For a more detailed write up on that, see our “Haltech Wideband CAN sniffing and DBC files“ article.

In summary, the Haltech Wideband controllers are an excellent option for use with other ECUs, but their behaviour didn’t quite match what we were expecting and as a result, setting them up with another ECU is even easier as the engine output CAN messages made no difference to how they behave. This also means that they can be run as a standalone unit without any ECU involvement if you just need to add a wideband sensor/controller and would like to be able to upgrade at a later date. If that is the case then you can add a standalone Arduino powered gauge to monitor your wideband readings, to find out more about that jump over to our “Making an Arduino Nano powered gauge for the Haltech WB1” article.

The CAN sniffing and testing was performed on a Haltech WB1 with Firmware Version 1.9 and 2.3

In some motorsport disciplines, the use of data loggers is prohibited, but the use of live gauges is not. That’s how this project came into existence.

The goal was to have a standalone gauge that worked with a Haltech WB1 without needing a Haltech ECU.

The basis of this board was a PT CAN IO mini, but the same result can be achieved using an Arduino Nano and an MCP2515 module.

Using a 1.3in OLED screen a display layout was decided upon which allowed for all the information required to be easily viewable. The main data we are interested in is the Lambda itself, so it is the largest text and in the centre of the screen. We are also interested in the Wideband sensor state as this is useful for diagnostic purposes. As the sensor state is only needed when the Lambda value is zero, the Lambda text is replaced with the sensor state.

To program the arduino we used the information in the “Haltech Wideband CAN sniffing and DBC files” article to know which CAN IDs we needed to watch, as well as which bytes we needed for each piece of information.

Once the gauge was working correctly it was time to add a startup screen (because who doesn’t love a bit of flair). Using the image to cpp converter found here:

https://javl.github.io/image2cpp/

A bitmap was generated for the PT logo. The logo was resized to correctly fit the 128×64 pixel screen.

With the bitmap generated the startup image should be simple enough to implement, but due to some idiosyncrasies of the Nano the image always appeared incorrectly. To solve this, the image needed to be stored in the SRAM rather than the flash memory, this meant that the logo had to be as small as possible to ensure there was enough memory left for the program.

To also help with diagnostics, when the correct CAN message is not received the display shows that there is a CAN error.

If you are interested in making your own standalone WB1 gauge, or in making a similar project, the project files can be found on GitHub

https://github.com/ptmotorsport/HaltechWB1_CANGauge

A 3D printable 52mm gauge mount is currently being designed and will be available as a purchasable enclosure or a printable file once complete.

A lot of OEM vehicle engine management systems lack proper forms of engine protection, with nothing more than a “low oil pressure” light. Usually in cases where the light comes on, it’s too late for the engine. While engine protection isn’t as exciting as big turbos and bulk power, it is a crucial part of the longevity of your engine. Whether you have a stock engine or a fully built performance motor replacing an engine uses time and money.

There are multiple types of engine protections which will be detailed below.

Knock

Knock sensors are crucial to ensuring an engine is running safely. Knock sensors are used to measure noise coming from the engine, and are used to detect detonation or “ping”. Detonation is when fuel is ignited by the combustion and heat from the cylinder rather than the spark event. This ignition is not only at the wrong time during the engine cycle, but it also has a less smooth explosion. As detonation can very rapidly cause damage to pistons, valves, cylinders, and heads it is important that it is kept under control.

Oil Pressure

Oil pressure based protection is one of the most common types of engine protection. A sensor is used to monitor oil pressure and the ECU monitors this pressure.

Low oil pressure can result in catastrophic engine failure through spun bearings, damaged camshafts and thrown rods. It can also cause damage to attached components like turbochargers. For this reason, shutting off an engine which has low oil pressure can save vital components.

When racing the risk of oil starvation is increased due to high sustained load while engaged in high g-force cornering, braking, and accelerating. This can cause oil to surge away from the pickup and starve the engine of oil.

As oil pressure changes with engine rpm and temperature, these factors are often added as tables which change the engine cut activation. The following shows an oil pressure limit table based on rpm and pressure. If the engine is over the set RPM and sees an oil pressure lower than the one in the corresponding cell the limit will kick in, either limiting the engine to a low RPM or shutting off the engine entirely depending on how the limit has been set up.

Engine Temperature

Engine temperature can be based off either coolant or oil temperature, and both are important to keep an eye on.

Oil temperature being too high will usually cause pressure problems before there are issues with the lubrication properties.

Coolant temperature is more commonly monitored, and is included in a basic form in most OEM cars. A high coolant temperature can cause blown head gaskets, warped and softened cylinder heads, and even seized pistons. Even at a lower level, a coolant temperature which is too high can lead to detonation.

For these reasons having engine temperature sensors which allow the ECU to limit RPM, or cut the engine if temperatures get too high can be the difference between an engine which will live to race another day, or an expensive lawn ornament.

Coolant Pressure

Coolant pressure is often related to coolant temperature, but it can also give an insight into other parts of the engine. . A coolant hose which has popped off will cause a drop in coolant system pressure, this can allow the driver to stop and fix the problem before the car overheats. Coolant pressure that is high can be a sign of high engine temperatures which make it more likely for a head gasket to blow due to warping of the mating surface. Irregular spikes in coolant pressures can also be a way to diagnose a blown head gasket as the escaping combustion gases raise the pressure in the coolant system. With the increasing prevalence of combined pressure and temperature sensors adding a coolant pressure sensor and corresponding cutoffs in the ECU is an easy way to monitor leaks within the coolant system.

Fuel Pressure

Fuel pressure is another important factor as a drop in fuel pressure will usually lead to less fuel entering the cylinder than expected. This will cause the engine to be lean, which can cause damage. At lower RPM an engine running lean can just result in stalling, and poor performance, but under higher loads it can lead to detonation which may cause engine damage. Running lean will also cause higher cylinder temperatures which can also contribute to engine damage.

Fuel pressure can drop due to a failed fuel pump, or fuel pressure regulator. If this happens while the engine is under hard acceleration it can cause catastrophic engine failure. Fuel pressure isn’t always monitored as the base pressure is often set with a gauge, but most fuel pressure regulators have a bung which can be used for a sensor which can then be wired into the ECU.

EGT

High Exhaust Gas Temperatures (EGT) can cause damage to engine components, turbochargers, and the engine itself. In a diesel engine high EGTs are generally caused by the engine running too rich and is burning too much fuel compared to the amount of air it is getting.

Generally an EGT safety limiter is in place for diesel engines. The location of the EGT sensor(s) is important in ensuring your limiter numbers are set correctly.

Injector Duty Cycle

Injector duty cycle follows the same logic as fuel pressure with the same problems and dangers. If the injector duty cycle is at 100% then the injectors are effectively wide open. If the injectors are at 100% flow, the maximum amount of fuel is flowing through the injectors, it is then possible for more fuel to be necessary for combustion, but not enough is able to be provided. This will lead to a lean condition, and may cause engine damage. As the injectors are directly driven by the ECU it is usually fairly simple to add a form of injector duty cycle based protection to your ECU.

Lambda

Lambda (or AFR) protections also look at fueling, and can be another method to protect an engine from damage due to running lean, but Lambda protections can also be used to protect an engine from running too rich. As mentioned above, an engine which is running too rich can have EGTs which are too high. Using Lamdba based protections allows for a large scope of protection and can be used to prevent other issues. However, due to wideband sensors being positioned in the exhaust it is inherently a reactive system and while it will catch problems it may be too slow to effectively protect the engine from things like a sudden drop in fuel pressure. With that being said, as Lambda is generally monitored anyway it is not a particularly large extra step to set up when configuring your ECU.

Boost/MAP

While a properly tuned car is unlikely to overboost, especially to a level that causes engine or component damage, it is still possible for mechanical failures to occur. For example, a wastegate solenoid line can break, causing the turbo to continue building boost beyond where the wastegate should have opened. This can lead to increased cylinder pressures, causing bent rods, broken pistons, and lifted heads. For this reason a separate boost based engine limiter can be set up to limit the engine’s RPM if boost exceeds the set amount.

RPM

A general RPM limit is a great idea, as while an engine revving to the moon will sound incredible, it will likely only do that once before it starts to make some less incredible sounds. There is no hard and fast rule about what your RPM limit should be, but your tuner or engine builder should have a good idea of where to set the limit.

It is possible to have engine protection which is set up in an incorrect fashion. An ignition cut at high RPM can cause valve float due to high exhaust pressure from unburned fuel igniting in the exhaust manifold, on cars with shim-over-bucket style cam followers this can cause the shim to come out of its seat and cause engine damage. It is also important that the limits are set correctly. The location of a sensor can effect the figures which are read by the ECU. Or the numbers in the ECU might not be appropriate.

While not every aftermarket ECU will have the option to turn on all of these protections from a single tuning panel, many are able to be added through the use of User Defined tables. Engine protections also aren’t a magic bullet, and failures can still occur, but having as many safeties as feasibly possible will help ensure that smaller failures don’t turn into major ones. It is always important to talk to your wiring harness manufacturer, tuner, and engine builder about which forms of engine protection are best for you, and how best to implement them.

When adding sensors, or configuring an ECU you will often need to input calibration data so that the ECU knows which input values represent what data.

Often calibration tables will be available in ohms like this example from a Bosch Motorsport coolant temperature sensor

For some ECUs this is fine as you can select between units when inputting a calibration table

If this isn’t an option you can manually calculate the voltage using ohms law

Unfortunately this takes a lot of time to do manually, and if using a pullup resistor adds more complexity.

Fortunately, with a little bit of spread-sheeting know-how you can create an automatic voltage calculator with the ability to change the sensor input voltage and pullup resistor value on the fly. The below link is to the PT Motorsport sensor calibration converter.

PT Motorsport Sensor Calibration Converter

We have included conversion tables for some common sensors, as well as the pull-up resistor values used by some common aftermarket ECUs.

If you need to convert your own sensor calibration you will need to first download the document so that you can edit it.

The main section of interest is the grey table. This has the sections for altering the pullup resistor and sensor supply voltage values as well as the sensor calibration data.

For the pullup resistor value you can either input the value you intend to use, or if you are using an ECUs internal pullup resistor you can input the value. The table includes some of the common ECU pullup resistor values. If you are unsure which pullup resistor to use, or want to know more about why we use them, read this article.

You can also enter the sensor supply voltage. In most automotive applications this will be 5V.

Finally, enter the sensor calibration data.

Using the example Bosch sensor above and using it with the internal pullup resistor in a Haltech ECU we can input the following data. Where there isn’t enough data to fully complete the table you can simply remove those rows.

As the data is entered the table will update, giving the correct output voltage values. Along with the table, a graph is also generated. This graph is useful to determine whether your pullup resistor is suitable for your application. For more information on why we use a pullup resistor, and how to choose the correct one for your application see our article.

With this data we can input the values into our ECU calibration table

This will look different across ECUs, the below is from the ECU Master EMU Black software, but using the same values (for this example that would mean wiring in a 1000ohm resistor rather than using the internal 2200ohm)

You may have seen references in articles to pullup resistors, or seen their use in wiring documentation. A pullup resistor diagram will generally look like this:

But what is a pullup resistor, and why do we even need one?

A pullup resistor is a regular resistor which we use to provide a known resistance. Why do we need a known resistance though?

The simple answer is that ECUs aren’t multimeters and generally can’t directly measure resistance. Instead using voltage inputs for receiving information. In a sensor with only two wires, there is no voltage being introduced and therefore no voltage for the ECU to measure. To provide a voltage we use a resistor from the ECUs 5V output to the signal wire of the sensor.

This gives the ECU a variable voltage output from our sensor. This also means that we need a pullup resistor for every input.

For ensure the correct variable voltage we need a resistor before and after the ECU sensor voltage input. Our pullup resistor is the resistor that’s before the input, and our sensor acts as the resistor after the input.

To show why a resistor is needed on each side, lets take a look at what our circuit looks like without each resistor.

Without the pullup resistor, the 5V output would travel directly to the ECU input, giving a signal of 5V all the time, regardless of what the sensor is doing.

Without the sensor the ECU would simply see 0V all the time.

What about pullup resistors for switches?

You may also see pullup resistors used in switch diagrams or referenced in ECU guides.

Why do we need a pullup resistor for a switch when it should just be on or off?

Once again, we will look at what happens without the resistor.

In this case when the switch is closed we end up with a short directly from the 5V output to ground.

With the resistor in place, the ECU will see 5V when the switch is open, and 0V (Ground) when the switch is closed.

How Do I know what resistance I need?

Generally for switched applications a 1kohm resistor is appropriate, but for sensor inputs it can be slightly more complicated.

If you use the PT Motorsport Sensor Calibration Converter (full instructions here) you will end up with a graph of the values across the voltage range. If you change the pullup resistor value you will notice that the scaling of the graph changes.

In this instance, as the resistance increases the resolution at various points changes. We want our most used ranges of data to be on the sloped part of the graph. As shown below, the different resistor values lead top different “ideal” ranges.

Depending on your application, you would select a resistor that gives good resolution in the area you expect to see.

For example, for a pre-turbo IAT sensor a 2200ohm resistor would be appropriate as a temperature range of -20 to 80 degrees is within our expected range.

For an oil temperature sensor, a 330ohm resistor would be appropriate as the higher resolutions are in the higher temperature range.

Where should I install my pullup resistor?

If your ECU doesn’t have an internal pullup resistor, or if it isn’t an appropriate resistance, where should you install your own pullup resistor?

Depending on the type of sensor and the available connectors you can either add your pullup resistor in-line in the harness or at the sensor connector. If you are running a 2 wire sensor having the pullup resistor near the ECU makes sense as you can add the resistor before you branch your wiring harness meaning that less wire needs to be run. That being said, a pullup resistor can be located anywhere practical using the method outlined below.

To wire a pullup resistor like this we use open barrel splices. Simply strip a short section of wire from your 5V signal wire and your sensor wire, and crimp the resistor across the wires. Our example is un-covered, but in your own harness this would then be covered by 4mm ATUM heat-shrink.

If you are running a Syltech Combined Pressure and Temperature sensor then you can use a bootable Deutsch connector to protect and hide the resistor at the sensor connector end of the loom. In this application having the pullup resistor at the sensor end works as the sensor needs to have a 5V power wire to drive the pressure sensor, so there are no additional wires run.

In this application, insert the wire into the Deutsch pin, then insert the resistor leg before performing your crimp.

Once crimped, insert the pins into the correct locations. The wires joined by the resistor will need to be inserted at the same time. The end of the bootable Deutsch connector can then be covered with heat-shrink or an appropriate boot.