skiwi

stage 2 of my on-demand tip conversion is finished, in that i have designed a circuit which i believe will work.

part 1 is posted at https://forums.audiworld.com/a8/msgs/120204.phtml

to summarise, i want to use my steering wheel tip buttons in an on-demand manner, like the later tip cars.

the wiring diagram is below:

<img src="http://pictureposter.audiworld.com/53706/tip_switch_circuit.jpg">

the circuit has the following features:

1) when either the "tip up" or "tip down" is pressed, the relay will switch the "tip recognition" circuit on. this has the effect of allowing the tcm to recognise the tip commands

2) there is a delay circuit on the trigger side of the relay designed to hold the "tip recognition" circuit closed for a number of seconds. in other words, once a tip button is hit, the "tip recognition" circuit will stay closed until the relay is released by the circuit

3) changing the value of the capacitor in the delay circuit will alter the circuit "hold" time

4) my calculations are for approximately 5 seconds, but i will need to test to be sure.

as yet unknown is whether i will need to assert the "tip recognition" circuit momentarily *before* the tip "up" or "down" signal. i will need to test this.

any suggestions or help is welcome.

stage 3 is to fabricate the circuit and test before "productionising" (stage 4).

banal

how does it know to go back to auto?

PaulW

Goes back to auto by the RC circuit timing out

ttuling

The relay will not stay energized once you let go of the + or - circuit. Also, you are depending on the TCM drawing enough current through the + and - switches to energize the relay. Does it? (For stock operation, It doesn't need to in theory, but maybe you've measured.) The capacitor just serves to turn off the relay once current from the TCM through the switch and relay has charged it enough.

You might consider an approach where each tip switch connects to the cathode (bar ends) of two diodes. The anode of one diode would go to your circuit and would provide a solid ground when the switch is pressed (albeit at 0.7 V and limited to the switch's current carrying capacity). The anode of each switch's second diode would go to the TCM input for that Tip switch. It protects the TCM from any positive voltage from your circuit, which for all I know will be higher than what the TCM ever wants to see internally on that wire (5 V?). This scheme lets the Tip switch work so long as the TCM recognizes 0.7 V as switch activation.

Tie together the anodes of each switch's diode that feeds your circuit, and you have a single wire that drops to near ground when either Tip switch is pressed. It's probably enough to run a relay.

Charging a timing capacitor through this wire might be problematic because of high charging current exceeding what the Tip switches can handle. For example, if you want to run a 5 mA relay coil (and that's not much as relays go) for 2 seconds after a 0.05 s click, that's 200 mA average during the click -- probably much more at the beginning when the empty capacitor starts to charge. Are the switches up to it? You could solve that with a bit of electronics, but I know you want to keep it simpler than that.

Tom

skiwi

your are correct about the relay. i got the diagram wrong, the rc circuit should sit across the relay.

anyway, i've been thinking about abandoning the rc circuit idea in favour of a ne555 timer circuit with a pot to vary the timeout and some diaodes to protect things. a bit more complex so i'm still trying to figure it all out. it has been a long time since i've done any of this sort of thing.....

thanks for the helpful comments, i need all the help i can get :-)

tozoM8

Clever thinking. If you have enough voltage from the TCU it will work. Hopefully it will not send any feedback to the TCU.

ttuling

I was wondering if they still make those. They do. http://focus.ti.com/lit/ds/symlink/ne555.pdf is the data sheet from September 1973, revised March 2004. Marvellous.

It looks like the circuit on page 9 will do the job, except that the relay coil will be between OUT and GND, not OUT and VCC. OUT can supply 200 mA. Make sure to put a hefty diode across the relay coil, reverse biased when the relay is on, to absorb the voltage from the magnetic field collapsing when the relay turns off. (Not good to send that back to transistors.)

A 0.7 V trigger is okay -- pull it up to VCC with a resistor.
The capacitor between ground and CONT is not the timing capacitor -- note A on page 10 says it can improve operation. No idea how big it needs to be, but probably not electrolytic, just something small to filter noise. In fact, the timing capacitor probably shouldn't be electrolytic either -- tantalum would do better for stability with time and temperature.

From the data sheet, it looks like the SA555P or SE555P would be the choices for a DIP package with adequate temperature range.

Regarding how to power the thing, I'm not sure it would be long lasting if VCC is simply connected to car +14V. There can be multi hundred volt spikes on the line when the starter operates, for instance. Some combination of reverse polarity blocking diode, zener diode to ground and/or inductor in line, and capacitor may be needed. But I haven't ever known best practices in that area.

Tom