frankenferrari V12

[mk e]

Somehow 14 O2 sensors seemed like a better plan in my head than it seems looking at 14 O2 sensors, bungs and connector in a pile on the floor!

Same for the box of 12 LS-2 coils....actually a box of 16 coils because they were sets of 8 so I guess that like project looks worse than it actually is . They are Chinese coils so I figured having 4 spares was probably a good thing.

[stevieturbo]

I was sold chinese LS3 coils by an unscrupulous seller on egay.

Out of 8 one didnt work from the word go, and another 3 failed within 20 minutes.

Needless to say I removed them after that. There are sellers selling genuine units for very little money, buying chinese crap just makes no sense.

[mk e]

I was sold chinese LS3 coils by an unscrupulous seller on egay.

Out of 8 one didnt work from the word go, and another 3 failed within 20 minutes.

Needless to say I removed them after that. There are sellers selling genuine units for very little money, buying Chinese crap just makes no sense.

I've heard counterfeit coils are prevalent and crap.

Hopefully the ones I have that are Chinese and branded as such are actually better. I spoke to the place I got them and they said they've sold 30-40 sets with no complaints so hopefully the coils are good.

I'm short a could analog inputs (because I'm WAY too cheap to buy an LTCD) so I spent some time looking at the board under a scope and I'm pretty sure the circled resistors are the pull-ups for the NBO2 1&2 and AT5&6 and removing them will make those plain only AV inputs.. I'm planning to use 12 NBO2 NB 1&2 can stay but I've only got 3 temp inputs so I think those 2 resistors should come off........

[SprinterTRD]

I wouldn't remove anything from the ECU as this will affect the calibration of the inputs.
Needless to say there will be no warranty on the ECU.

Why don't you see if the sensors you want to connect can drive the input with the pullups intact.

[mk e]

I wouldn't remove anything from the ECU as this will affect the calibration of the inputs.
Needless to say there will be no warranty on the ECU.

Why don't you see if the sensors you want to connect can drive the input with the pullups intact.

Oh....that ship has sailed

The manual says the AT circuit has a 1k pullup and I only found 6 1k resistors on the whole board which is how many 6 AT circuits need and these 2 were off by themselves which makes sense as the M150 has 6 AV but the M130 has just 4 and they're attached to pins A1&2.....so the person who layed out the board almost certainly took the M130 design and expanded it to the M150. I'm pretty sure removing them is all there is to it and removing the 2 resistors next to them would convert the 2 lambda channels to plain AV pins as well.

Those parts are TINY! I used 2 soldering irons as tweezers to heat and lift so as not to damage anything and it seems to work nicely....so AT5&6 should now be effectively AV18&19 and I'm good to go.

[greenamex2]

I am going to be really interested if this works....I have a surplus of temp inputs and a deficit of voltage inputs.....

[Scott@FP]

A standard potentiometer such as a TPS or linear or rotary position sensor, or a multiplexed position switch works just fine on a 5v biased AT circuit. Instead of using 3 wires you use two.

[mk e]

A standard potentiometer such as a TPS or linear or rotary position sensor, or a multiplexed position switch works just fine on a 5v biased AT circuit. Instead of using 3 wires you use two.

That was my 1st thought (TPS to the Temp input) and I weren't planning DBW (or using TPS for load sensing) I'd probably do exactly as you describe. The problem there though is you're flowing current through the sensing contact, so the voltage reading becomes a function of the contact resistance so output will vary over the life of the sensor and vary sensor to sensor vs the 3-wire setup where contact resistance is nearly irrelevant....which is why TPS is always 3 wire from the factory. 2 wire is functional but significantly less than ideal.

On a lambda or knock input with 1M and 160k pullups respectively I don't see why you couldn't ignore the pullup and wire as normal as the current flow would be so low I seriously doubt you'd ever know the difference. On my setup I'm already planning to use those inputs though

[mk e]

I am going to be really interested if this works....I have a surplus of temp inputs and a deficit of voltage inputs.....

It will be about a month before I start wiring it in I think but I'll post the results as soon as I have them. I'd bench test it now but haven't settled on a FW package so there's not much I can actually test.

[Scott@FP]

A standard potentiometer such as a TPS or linear or rotary position sensor, or a multiplexed position switch works just fine on a 5v biased AT circuit. Instead of using 3 wires you use two.

The problem there though is you're flowing current through the sensing contact, so the voltage reading becomes a function of the contact resistance so output will vary over the life of the sensor and vary sensor to sensor vs the 3-wire setup where contact resistance is nearly irrelevant....which is why TPS is always 3 wire from the factory. 2 wire is functional but significantly less than ideal.

Microamps ain't going to make a difference. 1K pullup and 100K pulldown. AT and AV pin is loaded with the same 100K pulldown. Most DBW throttle bodys and pedals are actually hall effect position sensors, and TP is treated a little bit differently as far as ECU logic is concerned, there are min and max limits for voltage at the input that must never be exceeded, such as 0.25v and 4.75v, so using traditional methods is best.


BTW awesome project.

Highly configured applications that use almost all inputs/outputs aren't easy to get sorted out, you're going to have to prioritize and make some compromises, regardless of the ECU used.