Flasher/Warning CPU module

[Casey]

I've posted this rx7fb.com, but thought I might as well post it here too, in case it might be of help to others, now or in the future.

Here is the unit in question, which seems to appear in versions A thro F (white lettering on the module). My investigation is on a version "E" module. Circuit board has been removed in the photo below.

The unit is located on the panel near the driver's right knee in a RHD model, near the fusebox. It's a bit fiddly to remove as the three nuts holding it in place are a little hard to access. There is one large connector block and one smaller, single (or double on some versions I believe) connector. Once the three nuts are the two connections disconnected, it is easily removed.

[Casey]

3 rubber grommets hold a metal shield on the back of the unit in place. Slide these grommets off to remove the metal plate, then two self tapping screws hold the circuit board in place, Remove those screws and unclip the board from the edges of the casing to reveal:

[Casey]

The next stage was to work out the components and circuit layout, so I could do a bit of reverse engineering. There is very limited (and even confusing!) information about what functions are controlled by this board, based on the Mazda wiring diagrams. It doesn't help that some functions are not present on the UK version - hence a lot of missing components on the board.

After a lot of puzzling out, it turns out there are 4 functions on the board, as delineated by the dashed white lines on the circuit board top. I've marked those functions on the photo below. It took quite a while to figure out the last ("?") circuit function, but it turned out this controls the single wipe and intermittent windscreen wipe function.

So, the four functions are:

1. Single wipe and intermittent windscreen wipe function.
2. Ignition key barrel lamp timer (on for ~30 secs after door opened)
3. Door open, ignition off, lights left on, warning buzzer.
4. Hazard and indicator flash timer.

[Casey]

On the top side:

"?" = wiper circuitry

[Casey]

These Mazda wiring diagrams helped to identify some of the external connections to the module. It's made harder as the diagrams cover functions not available on UK models, such as seatbelt warning buzzers, etc.

[Casey]

However, the letters M, C, K, J, G and E on the above diagram enabled me to sketch out this circuit and determine it provided an intermittent signal to the dashboard mounted buzzer to sound when a door was open and the car lights were left on (left hand part of the circuit below).

Additionally, the right hand part of the circuit provided a roughly 20-30 second "ground" to the ignition barrel light to illuminate the key area, triggered when a door was opened. I have managed to confirm these circuits and their operation by hooking up suitable inputs and outputs.

I'll tidy up this diagram later and post up a better version!

[Casey]

This Mazda wiring diagram showed letters R, Q J and identified the componentry on the board relating to the hazard and flasher function:

[Casey]

I traced the circuit connected to pins R, Q and J (J is a common "ground" for all the circuits).

I was able to hook up a 12 volt supply and a pair of 21w lamps to simulate the real load and confirmed the flasher functioned as expected.

[Casey]

That just leaves the last circuit connecting to pins A, B, F, H and L on the module. I finally found reference to these, hidden on the windscreen wiper part of the Mazda diagrams.

[Casey]

I traced the circuit and confirmed "B" and "H" are inputs from the windscreen wiper stalk, controlling intermittent wipe and single wipe functions. "L" connects to the wiper motor, which has a switched +ve battery on it's other terminal. This circuit provides a ground level output to kick the wiper into action.

[Casey]

I have now completed and tested a prototype circuit which is capable of plugging into the hazard unit to operate all the functions - thanks to DKWW2000 (Pete) for the donated plug  .

BTW, kiwidave kindly lent me his CPU module to reverse engineer and design the tester. Everything except the 30 sec "ignition barrel illumination function" seemed to work. If you look back a page to the rough cct diags I discovered two transistors (Q5 and Q6) had blown. I replaced those with some similar spec NPN transistors in my collection and it's all working again - very satisfying :thumbup:

Next task is to construct a little test box that can be plugged into the module to run a full function test, in situ, or on the bench.

Parts all assembled, but will take a few weeks to complete. Wiring harness to connect module to test box already done.

I'll update later......

[Eccentric shaft]

Mindblowing stuff!