G35 carputer climate controls

**2k1Toaster** · 10-23-2010, 08:10 PM

I was assuming he was going to desolder the buttons if doing the relay way, but you are correct soundman. If you want to leave the buttons in place, it will only work if they are NO.

**doctr_nick** · 10-23-2010, 08:15 PM

Originally Posted by soundman98

well, you do need to determine if the switches are NO, or NC. if they are NC, connecting a relay along-side the switch will not work--because you would need to break the connection for the system to recognize the button press..

But i would be hooking the relay up such that it bridges the contacts on the PCB just like the original button. Normally, I have to push the button down for anything to happen, meaning that the button has to complete some sort of cicuit. If I mount the relay as described above it would act as a surrogate button and be able to do the same job when given input from the fusion brain. Thus I don't need to worry about the fusion brain sending a signal directly to the MCU, but rather sending a signal to the relay to close the circuit. Am I wrong....?

**2k1Toaster** · 10-23-2010, 08:28 PM

Originally Posted by doctr_nick

But i would be hooking the relay up such that it bridges the contacts on the PCB just like the original button.

That would be correct assuming it is a NO button. So when resting, no connection. You press the button, then there is a connection. You close the relay, there is a connection. You press both at the same time, a circuit is made.

However, if it is a NC button then when it is resting is it is connected. So you press the button, and the connection breaks. You flip a relay, the connection stays connected and nothing happens. If you press the button and have the relay on, it stays connected and nothing happens.

Originally Posted by doctr_nick

Normally, I have to push the button down for anything to happen, meaning that the button has to complete some sort of cicuit.

I think this is where you are misunderstanding. Again, a NO button would behave that way. The logic would be "If the pin goes from low (inactive) to high (active) then someone pressed the button, and I should do _______". However the same circuit with a NC would be "If the pin goes from high (inactive) to low (active) then someone pressed the button, and I should do _______".

Originally Posted by doctr_nick

If I mount the relay as described above it would act as a surrogate button and be able to do the same job when given input from the fusion brain. Thus I don't need to worry about the fusion brain sending a signal directly to the MCU, but rather sending a signal to the relay to close the circuit. Am I wrong....?

For a NO button, all good. Not so for a NC button.

So with the NC logic from above, when nothing is pressed it sees logic high. When you close the relay, you are just providing an alternate path for the logic high. So it never goes logic low, and never triggers the MCU to do anything.

**gaoes** · 10-24-2010, 03:36 PM

so determining if its a NO or NC is really the biggest obstacle. if it IS a NC button, then just use a SPDT relay (single pole, dual throw) and hook your contacts from the button on the 30 and 87a pins, so when you energize, your breaking contact, just like normal (85 and 86 are power and ground).

**soundman98** · 10-24-2010, 05:30 PM

Originally Posted by gaoes

so determining if its a NO or NC is really the biggest obstacle. if it IS a NC button, then just use a SPDT relay (single pole, dual throw) and hook your contacts from the button on the 30 and 87a pins, so when you energize, your breaking contact, just like normal (85 and 86 are power and ground).

mostly correct, just make sure that the relay your using has a NC 87a terminal. some relays use 87a and 87 for NO..

**gaoes** · 10-24-2010, 10:59 PM

Originally Posted by soundman98

mostly correct, just make sure that the relay your using has a NC 87a terminal. some relays use 87a and 87 for NO..

sorry, used to working with aircraft relays, easier to understand

**gaoes** · 10-27-2010, 12:03 PM

so any update on this? i might be interested in doing something similar in my MS3 if at all possible, would make room for a vertical 10" screen lol

**doctr_nick** · 10-27-2010, 08:20 PM

Originally Posted by gaoes

so any update on this? i might be interested in doing something similar in my MS3 if at all possible, would make room for a vertical 10" screen lol

Sorry guys, no progress yet. I've been out of town for one of my clerkship rotations and haven't had a chance to start testing things with the multimeter. I'll try to get around to it this weekend, and post back.

In the meantime perhaps I could get an answer to another one of my questions. I've reviewed the FAQ's regarding powering a carputer, ect. and all seem to indicate that it is ideal to run a new wire direct from the battery. If I were to use a nettop that only consumes ~30W max, would this still be necessary? Is it still too risky to tie into the factory radio harness? I mean, right now I have a Pioneer AVIC f700bt hooked up, which is essentially a low power computer, so it would be interesting to find out what the power draw of one of these units would be in comparison.

**gaoes** · 10-28-2010, 03:47 PM

you can try using one of the empty fuse slots. 12v x 5 amps = 60 watts. TFT uses about 12 watts iirc, not sure on the fusion brain.

**soundman98** · 10-28-2010, 05:04 PM

in your case, it should not be a issue to use the radio wire.

one of my reasons for supporting running a new wire is because you can control all aspects of it, and know everything about it-- where it is run, what size fuse and wire, and what is connected to it--in my eclipse, the radio fuse also powers the dome lights, and clock, and those devices would skew the total allowable power capacity of any other add-on device..