For anything you'd like to see added to Speeduino
User avatar
By kimdrummel
#6824
Hi everyone!

Some may know i have previously requested the Toyota 24 + 2 (should we call it that?) trigger to be implemented into the code.

To further aid this, i have done some research with oscilloscope and dived into how the trigger is wired to the OEM ECU.

I have used the distributor from a european Toyota Celica with an NA 3S-GE engine. Some say the 4A-GE and many other toyota engines have the same type of trigger but i can not confirm that though as i only have the 3S-GE stuff readily available.

This distributor contains one 24 teeth trigger down at the bottom of the dizzy. No missing tooth. And on top, there's 2 additional trigger sensors that are located on each side of the dizzy. One of these extra trigger sensors is placed at the 0 degrees TDC point and the other one at the 360 degrees TDC point (crank degrees).

All of these triggers use the same ground. I don't know if this is a problem though. At least it worked fine picking up the signals on my OWON oscilloscope using a power drill at what i believe is far higher speeds that what a camshaft operates at. The images provided of the signals though are taken at sensible speeds.

The cam sync lobe seem to have two trigger points. One that i would call the primary trigger, a very sharp pointy tooth. And then it has what i would call a "dip" at the other half of the lobe. Maybe this is nothing to be concerned about though. I don't really know if a MAX9926 would pick up that dip on the lobe as a triggering signal. See images of it below.


Heres an image of the 24 teeth trigger wheel inside the dizzy:

Image


Heres a detailed image of the cam sync lobe. Both the pointy sharp primary trigger and the "dip" on the other side:

Image

This image describes the actual signals. In this particular one, Channel 1 (RED Curve) is the 24 teeth signal, and Channel 2 (BLUE Curve) is the signal coming from the cam sync trigger sensor located at the 0 degrees TDC point. It also shows a dip in the sync signal when the dip on the lobe passes the trigger sensor:

Image

In this image i have provided a signal i discovered practically by accident when by mistake connecting both cam sync trigger sensors to the same probe. This signal seems pretty similar to the Audi 135 trigger with one seperate signal every crank turn on top of the constant analog signal from the 24 teeth trigger. The difference being that instead of 135 teeth per revolution, there's only twelve. Although i would believe the previous signal using only one sensor for syncing would be more reliable and maybe even easier to write a decoder for since that signal can tell you exactly when the engine has made a complete cycle, instead of a half one, having the ECU not really know what cycle the engine is in:

Image


And here's a zoomed in image of the same signal, using both sync trigger sensors on the same probe. Although i believe the speed is slower here as well seeing as the signals produce far less voltage than before:

Image



So, think this trigger would be possible to implement into the code?

Personally i believe the method of using only one of the sync trigger sensors would be the best way to go because, as i said, it would be able to tell the ECU when the engine has done a complete cycle. Maybe making it possible to eventually run fully sequential.
Last edited by kimdrummel on Sun Mar 27, 2016 9:39 am, edited 1 time in total.
By noisymime
#6825
If I'm reading and understanding this correctly then with just the crank + one sync trigger sensor connected you get 12 evenly spaced pulses per crank revolution and 1 pulse on the secondary per 720 crank degrees. If this is correct then it should work just fine with the existing dual wheel decoder (though obviously with a VR conditioner by the looks).

Set the primary teeth to 12, secondary teeth to 1. The trigger angle represents the degrees ATDC of the first crank tooth after the cam tooth, which appears to be 0 based on your excellent pics.

If you have a VR conditioner board handy already, feed it the crank signal and the 0 degree TDC signals then scope the two outputs. That should confirm exactly what will be fed to the arduino, but based on what you've described I'd expect it would work just fine with the above config.
User avatar
By kimdrummel
#6826
noisymime wrote:If I'm reading and understanding this correctly then with just the crank + one sync trigger sensor connected you get 12 evenly spaced pulses per crank revolution and 1 pulse on the secondary per 720 crank degrees. If this is correct then it should work just fine with the existing dual wheel decoder (though obviously with a VR conditioner by the looks).

Set the primary teeth to 12, secondary teeth to 1. The trigger angle represents the degrees ATDC of the first crank tooth after the cam tooth, which appears to be 0 based on your excellent pics.

If you have a VR conditioner board handy already, feed it the crank signal and the 0 degree TDC signals then scope the two outputs. That should confirm exactly what will be fed to the arduino, but based on what you've described I'd expect it would work just fine with the above config.

Yeah i was thinking the same thing actually. The only thing i'm concerned about though is that "dip" in the signal. Unfortunately i don't have a VR conditioner board ready. But i am starting to realize i should probably get one of those MAX9926 boards, maybe from JBPerf. Or make my own based on the same circuit i plan on using.

And if anyone could provide some info on other popular Toyota engine triggers, for example the 4A-GE, i think many would appreciate if we could determine if the patterns are compatible with speeduino. :)
Last edited by kimdrummel on Sun Mar 27, 2016 9:59 am, edited 1 time in total.
By noisymime
#6827
kimdrummel wrote:Yeah i was thinking the same thing actually. The only thing i'm concerned about though is that "dip" in the signal. Unfortunately i don't have a VR conditioner board ready. But i am starting to realize i should probably get one of those MAX9926 boards, maybe from JBPerf.
Yeah that's a good question. It can only really be answered by scoping what the output from the VR conditioner looks like I'm afraid. My guess is that it will be ignored by an adaptive conditioner like the 9926, but its just a guess.

If you joined the 2 signals together it would actually still work just fine with the same settings, but you won't be able to go sequential when the support for it arrives in firmware.
User avatar
By kimdrummel
#6828
noisymime wrote:
kimdrummel wrote:Yeah i was thinking the same thing actually. The only thing i'm concerned about though is that "dip" in the signal. Unfortunately i don't have a VR conditioner board ready. But i am starting to realize i should probably get one of those MAX9926 boards, maybe from JBPerf.
Yeah that's a good question. It can only really be answered by scoping what the output from the VR conditioner looks like I'm afraid. My guess is that it will be ignored by an adaptive conditioner like the 9926, but its just a guess.

If you joined the 2 signals together it would actually still work just fine with the same settings, but you won't be able to go sequential when the support for it arrives in firmware.
Yeah, i think the best thing would be to only use one of the signals. It's easy enough to just put a jumper between the two on the board if needs be. But sequential is good stuff so that is definitely something i will prepare for. Ignition is still single coil control with dizzy cap though so there's only need for sequential fuel. At least on the 3S-GE. If the 4A-GE has the same pattern, some dudes in that field would probably like the possibility to run sequential COP's as well. :)

Gonna make me a MAX9926 board and see how it fares.
User avatar
By PSIG
#6832
Don't forget there is a "JBPerf replacement" MAX9926/Opto board already here that you can order from OSHPARK.com now. The circuit Josh designed is in your Speeduino download, under speeduino\reference\hardware\VR Conditioner. I have a handful of these PCBs and can throw one in an envelope to get you going if you're in or not too far from the US.

David
User avatar
By kimdrummel
#6857
PSIG wrote:Don't forget there is a "JBPerf replacement" MAX9926/Opto board already here that you can order from OSHPARK.com now. The circuit Josh designed is in your Speeduino download, under speeduino\reference\hardware\VR Conditioner. I have a handful of these PCBs and can throw one in an envelope to get you going if you're in or not too far from the US.

David
Unfortunately i live in Sweden. :(

Posting overseas from the US is usually expensive, but i am ready to take you up on your offer if you are willing. :)
User avatar
By PSIG
#6876
kimdrummel wrote:
PSIG wrote:I have a handful of these PCBs and can throw one in an envelope to get you going if you're in or not too far from the US.

David
Unfortunately i live in Sweden. :(

Posting overseas from the US is usually expensive, but i am ready to take you up on your offer if you are willing. :)
Hopefully, the tiny bare PCB will go as a standard letter. PM me your address and let's see what it will take to do.
By Hasse.69
#6883
I don't want to interrupt , but i do it anyway. :D
I also live in Sweden and got those boards......
Just sayin..

//Hasse.
  • 1
  • 2
  • 3
  • 4
  • 5
  • 15

you could set 'skip revolutions' to 3, then it wou[…]

Hi, I am trying to assign Signed values to the x-a[…]

blitzbox

I've finally figured out why MAP and Lambda weren'[…]

Hello, I bring news!! Let me tell you that after […]

Still can't find what you're looking for?