Hi,
been doing the schematic for my final year project, its for automotive, but in a standalone module that doesnt communicate through can or obd or anything, nor does it require any modification of the car itself, but if the mods have a problem, Apologies!
is it possible to get some feedback on my schematic? its approx 8 A3 pages and is by far the most complicated piece iv done before!
Thanks in advance

 

It would help if you did a brief description of your project.

It seems to me that you are building a track circuit lap tracking device. Am I right? There seems to be quite some wires feeding info from the depths of your car, but since I 'm no expert I 'll let others say what they think too.

 

good point! ok..
its effectively a data logger for cars.. incorporates a USB interface to transfer the file saved onto the onboard microSD to memory stick.. has accelerometer, GPS, accelerator and brake pedal position sensors. (not using CAN/OBD/OBD2) implementing EEPROM for saving of settings, 4 button control interface, OLED screen, onboard temperature sensors, with LDR to control brightness of screen.
in brief.. a datalogger.. with information display and plugins for future development..

 

Sheet 2 is missing.
Sheet 3, A1, Q1 - in order to turn Q1 ON, you will need to get Vgs (voltage on the gate relative to the source) to 4.5v or more. So, in order to get signal "Ign" to 3.3v, you will need to raise the gate to 3.3v + 4.5v = 7.8v.

If you are driving it from a uC that has a 3.3v Vdd/Vcc, you won't get there. Instead, consider using a very low-threshold P-ch enhancement mode MOSFET.
Also, you don't have a signal name on the gate, so it will be difficult to determine where it's coming from.

If the connector is disconnected, the gate will float. This is not good. You need a ~10k resistor on the gate to the source to keep it turned off.

Sheet 3, D6, "IGN COIL NEG" signal - you may not realize that the negative side of the coil can hit several hundred volts when the transistor turns off. R10 will probably dissappear in a puff of smoke.

To the right, you have a signal called "CAR_RPM". If the car is revolving, you have problems. You might be tempted to call it "ENGINE_RPM" instead, but it won't be engine RPM either. You could call it "TACHOMETER" or "TACH" for short, because that would be the more correct nomenclature for the signal.

With a 4-cycle Otto engine (which most modern vehicles except for rotary engine and hybrids are) you get one spark per cylinder every other rotation of the crankshaft. Starting sometime in the 1980's, they started using one coil per cylinder instead of one coil per engine and a distributor. Nowadays, the ECU puts out a tach signal, so you don't have to do anything with the primary side of the ignition.

Sheet 4, on the left - you have several tank circuits showing. They will have one frequency where they reject current flow in either direction. Above that, the capacitor will couple the signal across; below that, the inductor will couple the signal.

If you want to block RF, consider using toroidal baluns and/or ferrite beads.

I see you're using more than one switching regulator. But for the 12v output, you're using the input from the 3.3v supply. This could result in quite a drain from the 3.3v supply, and you're taking a double hit on efficiency by running the current through two regulators. Why not use an isolated buck-boost configuration for the 12v supply instead?

 

What information does the datalogger actually read directly from the car? How is that done? Where do you tap to the signal sources?

 

Some of the sheets havent been labelled (1 or 2 etc) properly, in my rush to get this up i just exported what i had finished. shall all be labelled and filled in soon

Sheet 3, A1, Q1 - in order to turn Q1 ON, you will need to get Vgs (voltage on the gate relative to the source) to 4.5v or more. So, in order to get signal "Ign" to 3.3v, you will need to raise the gate to 3.3v + 4.5v = 7.8v

being driven from 12V Input from the ignition. (turn the key on, the uP sees the pulse as an external interrupt waking up the unit =]) and in terms of it pulling it down.. would a 100k be better just in terms of power consumption? and it doesnt need to be a strong pull down anyway..

Sheet 3, D6, "IGN COIL NEG" signal - you may not realize that the negative side of the coil can hit several hundred volts when the transistor turns off. R10 will probably dissappear in a puff of smoke.

I have been informed that the circuit will work, though the values may need adjusting.
id prefer to stay away from depending on an ECU for the tach signal, i know someone whos done some work with rev limiters and launch control so i should ask him what the best way of doing that is maybe?

on sheet 4, i put the inductors and capacitors because i wasnt sure which would be better for decoupling? i know that a capacitor is better in theory, however logically, a small inductor would allow dc balancing of the ground plane to the case/chassis.

the reason i attached a boost switcher to the 3V3 for the 12V was i thought it would be a better more stable supply compared to the input voltage which could be anywhere from 11-15V. wasnt sure how else to do it.

What information does the datalogger actually read directly from the car? How is that done? Where do you tap to the signal sources?

The only signals from inside the car (at the moment) are RPM, Road Speed (though this may not be implemented in software until later - implementing a dead reckoning function), and Brake and Accelerator Pedal position, receiving signals from my linear position sensors.

hope this answers all queries, thanks for the feedback! spend so long doing the schematic you go googlie eyed!