DIY exhaust AVCS controller?

[NSFW]

It's time for me to do a motor build. My car currently has a single-AVCS motor, (like every 05 Legacy GT) and I'm thinking about getting a dual-AVCS motor put in.

Of course that would require something to control the exhaust cam timing, so I'm wondering if it would be practical to build an exhaust-AVCS controller around an Arduino or Atmel AVR or similar small microcontroller.

I write software for a living and I have some embedded exerience, so I'm not worried about that part. If there's a roadblock here it would be in the hardware to interface the AVCS sensors and actuators with the microcontroller. For that, I could use some help.

Does anyone know where I can find some info about the signals from the exhaust cam angle sensors, and the signals to the oil control valves?

Thanks!

Edited to add: If I end up trying this, whatever I come up with will be totally open-source, GPL v2.

* Registered users of the site do not see these ads.

[NITROS]

I'm willing to help you out with this. i was thinking about doing this with a netduino instead of arduino but ultimately it doesn't matter.

[NSFW]

I replied to your PM before I even saw this thread... I'd rather do netduino too actually. I spend all day in Visual Studio already.

[NITROS]

I looked into netduino, but it is too expensive and the amount of community support is much smaller than arduino. The netduino is significantly more powerful but for this project the arduino gives us enough power.

I just ordered my ardunio leonardo and a starter kit.

[86Dreams]

REALLY interested in this project please post updates. Interested in helping but not sure what I could actually contribute. ~3 years ago when I graduated college I used 4 arduinos to create a 3-phase Inverter, havent touched them since tho.

[sabreofsd]

You're going to need A/D and D/A converters... which probably means voltage dividers on the input side at least. The two big hurtles you're going to encounter:
To mimic the stock ECU's controlling, you'll need to create a table that is looked up for a value based on intake pressure (boost) and RPM. Getting the RPM is easy, the boost might be a little more tricky.
Now for the hard part: Making sure that the Arduino has enough power to maintain nyquist theorem's input/output PWM signals.

So....
A->D conversion
Table look up
D->A conversion of PWM signal to AVCS actuator
Making sure that you do this fast enough to keep up with the engine... otherwise you have the possibility (unlikely) of valve contact.

Check out this link (just googled) for some high speed ADC stuff: http://www.microsmart.co.za/technica...d-arduino-adc/

[NSFW]

It's my understanding (possibly wrong) that the cam position sensors are send pulses, not analog signals. And the AVCS control valves just need 12v PWM at relatively low power, so I don't think D/A or A/D will be needed.

Manifold pressure can be had from the existing manifold pressure sensor, or a separate sensor T'd off of an existing line. They're cheap and well understood, I've got an AEM sensor wired in place of the factory sensor on my car.

One of the biggest unknowns IMO is exactly how the AVCS signal pulses work. I'd kinda like to get a source for the sensor plugs so I could make something that would easily wire into a dual-AVCS motor with no cutting or splicing, just to analyze those signals. I don't have a dual-AVCS motor so I'd have to use someone else's for most of the testing.

[86Dreams]

Does the cam position sensor work the same way that a crank position sensor does? Crank position sensors work with a toothed wheel and magnet sensor (Hall-effect?) I believe. When the tooth passes by the sensor it produces an analog pulse, you know the number of teeth on the wheel so you know how fast the wheel (and therefore cam) is moving by counting the teeth. The 'beginning' of a new revolution is marked with either an extra tooth or a missing tooth.

Do the AVCs have drivers that accept a low power signal?

can you get an oscilloscope on an existing AVCs?

[sabreofsd]

Some threads of note for you:
http://forums.nasioc.com/forums/showthread.php?t=871645
http://forums.nasioc.com/forums/showthread.php?t=437418
http://forums.nasioc.com/forums/show...1078290&page=8
http://forums.nasioc.com/forums/show....php?t=2084497

From Phil's post, it's definitely a PWM signal!

[NITROS]

Originally Posted by 86Dreams

REALLY interested in this project please post updates. Interested in helping but not sure what I could actually contribute. ~3 years ago when I graduated college I used 4 arduinos to create a 3-phase Inverter, havent touched them since tho.

I would love any help you can offer. I'm really just learning circuit design for this project, so it will be a while before I have anything tangible.



Answering this as I read the posts, expect edits. I also have a thread going on chippernut http://www.chippernut.com/forums.htm...-im-l-4864009/

1) Does the arduino have enough power to run avcs. I would say so? The is a guy that created a piggyback ecu for a boosted IS300. The concept is very similar. Also there is another arduino project that is being developed as a standalone ecu for mazda engines. It's based off of the Megasquirt v1. Chippernut created a shiftlight using a arduino, with RPM readout, this is relevant as I plan to use the same mechanism to pickup RPM for a ref. He also utilizes the 2JZduino for his own IS300.

IS300: https://thedeltaecho.wordpress.com/2jzduino/
Speeduino: https://www.youtube.com/watch?v=Q475JN-t63E
Chippernut: http://www.chippernut.com/

[86Dreams]

Originally Posted by NITROS

I would love any help you can offer. I'm really just learning circuit design for this project, so it will be a while before I have anything tangible.



Answering this as I read the posts, expect edits. I also have a thread going on chippernut http://www.chippernut.com/forums.htm...-im-l-4864009/

1) Does the arduino have enough power to run avcs. I would say so? The is a guy that created a piggyback ecu for a boosted IS300. The concept is very similar. Also there is another arduino project that is being developed as a standalone ecu for mazda engines. It's based off of the Megasquirt v1. Chippernut created a shiftlight using a arduino, with RPM readout, this is relevant as I plan to use the same mechanism to pickup RPM for a ref. He also utilizes the 2JZduino for his own IS300.

IS300: https://thedeltaecho.wordpress.com/2jzduino/
Speeduino: https://www.youtube.com/watch?v=Q475JN-t63E
Chippernut: http://www.chippernut.com/

Im subscribed, good luck.

[NITROS]

My arduino finally came in. I'm going to try to build some simple circuits to basically just read the RPM and cam position.

Does anyone know how to verify whether the cam position sensor is a hall effect or VR? I know I can use a scope but I don't have one. (First project build a arduino scope?)

[sabreofsd]

https://www.lab-nation.com/

[86Dreams]

Originally Posted by sabreofsd

https://www.lab-nation.com/

honestly for that price

http://www.amazon.com/Siglent-SDS1052DL-Digital-Oscilloscope-Frequency/dp/B00GQNN70A/ref=sr_1_1?ie=UTF8&qid=1437145821&sr=8-1&keywords=oscilloscope
might as well just get a regular o-scope...assuming youre not working in a parking lot that is.

[sabreofsd]

That's a very decent scope for the price! Better resolution, but no outputs. Different tools for different applications

[NITROS]

Looks like I found a solution for the vr signal. www.jbperf.com/dual_VR/v2_1.html

[NSFW]

Which sensor uses VR? Is that for the intake AVCS angles?

[NITROS]

Originally Posted by NSFW

Which sensor uses VR? Is that for the intake AVCS angles?

Its my assumption at the moment that the Exhaust AVCS cam position sensor is VR. The intake cam position is hall effect. This is based on my research. I found that VR sensors are 2 pins and hall effect snesors are 3 pin. On the ej20x the exhaust cam position sensors are 2 pins. the intake cam position sensors are 3 pin.

I still need to verify by hooking it up to an actual scope.

[fragment]

Subscribed.

My 04 FXti is equipped with DAVCS heads, but no exhaust control. I'm making about 40 hp less than I should given my mods. My tuner suspects the lack of advance is the major issue.

[NITROS]

You can mechanically advance the cams a bit by adjusting the cams by two teeth ccw, I think. There is a thread here on nasioc. This would put the exhaust cam in roughly the position of a single access head. Do a search.

[fragment]

Originally Posted by NITROS

You can mechanically advance the cams a bit by adjusting the cams by two teeth ccw, I think. There is a thread here on nasioc. This would put the exhaust cam in roughly the position of a single access head. Do a search.

This is my current thinking. Pity I didn't find that thread before I drove 6 hours one way for a tune.

My searching failed me.

[NITROS]

This is the thread I referred to.

Http://forums.nasioc.com/forums/showthread.php?t=2582784&page=3

Originally Posted by fragment

This is my current thinking. Pity I didn't find that thread before I drove 6 hours one way for a tune.

My searching failed me.

[NITROS]

I received my VR conditioner circuit. I will start experimenting building the circuit this weekend.

[NSFW]

Originally Posted by fragment

Subscribed.

My 04 FXti is equipped with DAVCS heads, but no exhaust control. I'm making about 40 hp less than I should given my mods. My tuner suspects the lack of advance is the major issue.

Were there any other drawbacks to running no exhaust cam advance? How's startup & general driveability? Did the break-in go OK?

I was thinking about getting a dual-AVCS motor put in now and figuring out the cam control later, but I wasn't sure the motor would even start... Sounds like that fear was a little overblown though.

[NITROS]

Originally Posted by NSFW

I wasn't sure the motor would even start... Sounds like that fear was a little overblown though.

I am running a JDM ej20x davcs engine on a 05 sti ecu and wiring harness. The car drives fine but it does definitely seem low on power.


Here are some items i found interesting in the 05 and 13 sti manuals. I compared those results with the legacy ej20x/y manuals. I found some differences and also some very helpful diagrams.

One difference I noticed between a EJ20x and DAVCS sti heads is that subaru has changed the cam position sensor from VR on the EJ20x to HallEffect on the DAVCS STI. This is good news for those with STI DAVCS heads as the purchase of a VR conditioner circuit is not required.


Looking at the image below both the 05 sti and 13 cam position in relation to crank position are identical.

2005 STI


2013 STI


If the above is true then I can use this data + some maths, I can calculate crank angle and the positions of the cams. After some more maths I should be able to figure out the needed PWM signal to the AVCS solenoids. I wonder if oil pressure will result in a necessary modification in the PWM signal sent to the AVCS solenoid.




I wonder of the LH intake position signal is a misprint as it shows rise in voltage instead of a drop?