Need help with simple (I hope) auto circuit

[KeepItSimpleStupid]

alec:

The comparitor interfaces to the swich input to the PCM. When it goes below 400 mV (the contact closure), it pulls down the input of the ELM411?

Note that te comparior is availabe with ref connected to - and ref connected to +.

Fast collapse is just as important as power up. The power supply cant have a long time constant.

The comparitor has to be chosen, so its open collector output is ON when the switch is pressed.

A DPST swich COULD be used to eliminate the comparitor AND isolate the indicator circuit.

I built an automotive circuit that happened to use the regulated ant pre amp supply as a trigger for a timer. Issues could occur in my case with the on/off while starting, ON behavior of the radio supply. He may not have that issue. During starts some accessories go off like radio and heater.

 

[alec_t]

The comparitor interfaces to the swich input to the PCM. When it goes below 400 mV (the contact closure), it pulls down the input of the ELM411?

Couldn't the switch connect directly (or possibly via a protective resistor) to pin 4 (In3) of the ELM411? Closing the switch would provide a negative edge to clock the flip-flop.

 

[KeepItSimpleStupid]

Yea, probably, BUT the ELM411 has an internal pull-up already. I think it's safer to use the comparator. Now, it might be possible to diode isolate the two circuits too.
diode drop + switch activates the tow/haul mode and a diode drop + same switch activates the ELM411. It just depends on logic levels. The ELM is nothing but a programmed PIC. That would have to be tested.

The use of a DPST switch is safer yet.

 

[KeepItSimpleStupid]

jeepin jeff:

You need to do a few things:
1) Measure the open circuit voltage at the pin you ground
(guessing it will be 5V)

2) Can you momentarily ground the pin via one diode, and again with two diodes in series and see if the system toggles. If you can monitor the voltage at the pin you would normally ground that would be great. I think you will see 5 to 0.6 or so and 5 to 1.2V or so.

The band(s) of the diode(s) should point to ground. In one case it should be around 0.7V and the other around 1.2V. Around, not exactly.

All that means is grab two 1n400x diodes (RadioShack) connect it to the pin that needs to be grounded with the band pointing to ground.
Do the same with two diodes in series. Check if in both cases the tow package toggles.

 

[jeepin jeff]

Sorry not to get back until now, I can do all the above and may try to get out to do it today.
Thank you

 

[jeepin jeff]

Just ran into this, https://www.hptuners.com/forum/showthread.php?8887-Performance-Mode-Switch-Option/page5
There are 5 pages (that started in2006) devoted to the solution I'm looking for and a frenchman seems to have come up with something but it gets pretty quiet after that. KISS I think you and Alec are going down a similar path, what do you think? Just more food for thought, if you're still hungry!

 

[alec_t]

what do you think?

I think the French circuit (a) risks damaging the ECU because SW1 is switching a relay coil without a back-emf suppression diode, (b) the relay is totally unnecessary anyway and (c) pins 8-11 have been left floating, so the IC may behave erratically.

 

[KeepItSimpleStupid]

What Alec said. I think they are trying to use the relay to debounce, but the relay contacts itself will bounce. It's 12 V totally, and that's what we're waiting on confirmation for. I'm asking for the one or two diode drops just to see if the logic levels are CMOS which means that the switch ground has to really really be close to zero volts.

In the circuit above, the pushbutton turns on the relay as long as you hold it down. What's the point of that?

 

[jeepin jeff]

In the circuit above, the pushbutton turns on the relay as long as you hold it down. What's the point of that?

If you watch the video in the thread its doing exactly what we are after. I am going after parts soon and will get the info you guys are after. I just think its interesting that this same common silly little problem has been an issue since at least 2006!

 

[alec_t]

Sorry, I'm not convinced by a video of a LED going on/off. It tells us nothing about the possible problems with that circuit.

 

[jeepin jeff]

Ok guys, I went out and did a little checking, it is a 12 volt source. These were taken with the engine running and in drive.
Haven't ran in to radio shack yet but I got to thinking about this in regards to the diode test: Check if in both cases the tow package toggles
I have an axle out right now and can't actually drive to feel the difference in shifting mode, do you still want me to go through this or does it need to wait?

 

[alec_t]

Just to confirm, are those the wire 71 voltages with and without the button pressed, respectively?

 

[jeepin jeff]

Yes sir, the blue wire in the photo is connected directly to pin 71 of the power control module.

 

[KeepItSimpleStupid]

The diode tests are less important than knowing the open circuit voltage is 12 V.

There are a number of approaches that can be used (I never like like suggesting the best, nor do I list them in order - beware)

1) Sort of an isolated approach. The indicator circuit is totally separate. This would need a double pole switch.
2) A semi-isolated approach where diodes attempt to isolate the circuits somewhat. The "diode test" would confirm if this is possible.
3) A monitor of the 12 V to ground signal without a direct regard as what is the logic low. Just as long as the switch takes it below logic low.
i.e. if the logic low were 10V, and we detected 400 mV and the switch took it below 400 mV were OK.
4) (Not considered). Measure short circuit current and getting a better idea of the logic thresholds. Opens up using an opto-isolator. Current measuring is too difficult.

As I said earlier, some circuits don't like their inputs to go above the power supply voltage. There is usually an inherent diode in these devices that can draw too much current if that happens. "protecting" the inputs is one way to solve it. Not powered and having a 12 V signal applied means the power supply of the circuit is 0V, not its intended operating voltage.

The automobile environment is "hostile". Take a look at figure 3a here: https://www.littelfuse.com/~/media/...utomotive_tvs_diodes_application_note.pdf.pdf

That note puts the environment it in very simple terms.

 

[KeepItSimpleStupid]

Aside:

I tried to do some simple research, these are indicators that you really don't want to turn off completely. I have indicators like my traction indicator that won't dim and that one is annoying. VARIABLE DIM would turn the indicator off completely if the dash lights were turned down.

With the addition of a ULN2004 (3 out of 7 sections) and some glue logic, You can make the circuit accommodate LOGIC DIM or VARIABLE DIM. I don't like the way the industry defines them. ILLUMINATION & DIMMING

LOGIC DIM::= 12 V lowers the intensity (typically provided by the tail or headlight circuit)
VARIABLE DIM::= Older cars had a source of 0-12 V controlled by the dash rheostat. I think this signal has disappeared.

These signals are typically used for after market stereo systems.

 

[shokjok]

My 1990 Dodge truck has a built-in LED inside a dash-mount O/D switch. Is there some way to gain access to the PCM to get a voltage from the pin 71 feed?
If the vehicle defaults to regular mode upon startup, that signal could be used as a power-on-reset for a flip-flop driving a transistor. When you depress the TOW MODE button, the flip-flop can feed another transistor driving the LED and the PCM pin 71. I suggest obtaining a TOW/ HAUL switch from an auto yard to attach to the dash.

 

[KeepItSimpleStupid]

I have no idea what happened to the OP. It might matter as to what dash the vehicle has. It was specified as GM not Dodge. Therefore a "Dash message" seems very probable.

 

[shortbus=]

Most times the pcm is sinking(grounding) a voltage, not supplying voltage. The exception is a 5V out to sensors.