Rae-San Electronics

Modern Electronics for Vintage Motorcycles

November Tinkerings

Posted by on Nov 6, 2013

I’ve been working on several aspects of the Rae-San CX ignition system at once.

  • Optimization of pickup noise
  • Modification of the software and hardware to use Coil on Plugs(COP)
  • Refinement and development of the Rae-San Hall Effect Sensor.

While these are essentially separate activities they are somewhat interrelated and hence why I’ve held off on the updates until now.

The idea is to end up with a single version that supports all variants with only a code change –

This has resulted in a slight delay to a few orders as I’m trying to finalise the version prior to shipping to these customers.

What’s the Result – I don’t want to read all that technical stuff ..

The end result is I have finalized the component values and connection scheme for the CX variants using the CDI pickups and driving 12V coils.

 I’ve also finalized the same design to be ready to take 1.6 ohm Coil On Plugs with only a software version difference and minimized the noise pickup.

 This also has a couple of flow on optimizations for noise pickup back into all other versions.

 The Technical Stuff …

 A bit of History

Prior to the CX500, Honda’s bikes used magneto or points based ignition – the CX500 bought an new player to the field – Capacitor Discharge Ignition – But like many things on the CX – it wasn’t plain CDI but a system that was very clever –that performed the timing advance but without the use of any mechanical advance or microprocessor. Instead it uses an analog mixing of the signals from 3 sets of coils to determine the firing time of the spark.

 One coil is used at low RPM – this provides the minimum advance and fallback timing of 15 deg BTC. Added to this is a signal from the “Advance Coil” at the rear of the engine – this is a broader signal and the amplitude is very dependant on RPM. As the RPM increases this signal increases and the spark is triggered progressively earlier.

 The final Coil is at 37 deg BTDC and is used to stop the Advancer coil generating too much advance by being subtracted from the Advance signal.

 So three coils to determine the timing –

In the 1982 onwards, Honda moved away from this system back to a mechanical advance and 12V coils – less clever but simpler –

 Also of interest is that many of Honda’s other bikes at the time – e.g. the VF750s and C were moving to a system using 12V coils, and dwell ignition – like the old points system – but triggered from a single pickup coil. This triggering is caused by a metal protrusion on the flywheel or crank. This protrusion acted through a single pickup – one edge setting the maximum advance, the other the base minimum advance.

 This is very similar to the 1982+ CX version except it uses a magnet and a coil rather than a self biased magnetic coil.

 We should look at these coils for a moment.

In the CX, two of the coils sets are located on the inside of the rear cover – in curved pickups – these are the max and min advance pickups. These are simple pickup coils – they don’t contain any magnets themselves but are triggered by the flywheel rotor passing.

 The flywheel rotor contains strong magnets held inside a “CUP” – but there is a brass “BUTTON” on the outside – this actually goes through the cup and allows the magnetic field to “leak” out around the button – and this sweeping past the pickup coils generates the trigger pulses.

 As this is a simple magnet and coil assembly – the amplitude of the generated pulse is proportional to the RPM on the engine – small at cranking speed up to quite large – 30+ V at high RPM.

 Honda in the other bikes uses a “biased” coil – this is a coil wrapped around its own magnet that reacts to the close proximity of ferrous metals. The amplitude of this output pulse is much more constant and is mostly a function of the proximity of the metal. There is some speed dependence and low RPM during cranking but the output amplitude is much more stable. Another benefit of this approach is a much lower incidence of noise pickup from the spark firing and general electrical noise.

 In Practice

The CX coils tend to pick up much more environmental noise and in particular the firing of the spark introduces a pulse into the pickup.

 The original CX CDI circuit features fairly heavy filtering on the input from the pickups to combat this but is also benefitted by the fact that once the CDI is triggered – it cant be triggered again until it is recharged – and this is a function of the location of the charging coils on the stator.

 I suspect that this ability of the pickups on the CX to pickup the firing noise is a primary reason for the move to the biased coil arrangement utilized in later bikes – and still used in many today.

 So what does this mean with my experiments in the shed.

Filtering out this noise is not a trivial problem as its very similar to the noise (trigger pulses) that we do want to see and as we are using them for timing determination – we don’t want to delay them too much.

 We can say – the CX pickups are acceptable for the use of standard coils – 3-4ohm as these generate low enough noise to be filtered out.

 When using Coil On Plug (COPS, stick coils) the current is roughly double per coil and the dwell is of the order of 2mS. This means the induced noise spike is higher and right in the region of interest where we are expecting the real trigger pulse to occur at some of the RPM range –

 Through extensive experimentation and optimization I’ve arrived at a set of values that work well for the standard coil option and also work well for the COPS.

 The Rae-San CX Ignition Module COP version will undergo some further testing and possibly some time on the dyno to tweak the very best out of this arrangement, but the standard version is GOOD TO GO.

Eventually the very best solution will be the use of the Rae-San HALL EFFECT sensor for using COPSs on a CX – instead of the pickups, as this avoids noise pickup altogether.