An exercise in overthinking - Junits15's street build

[Duece McCracken]

A focus ST is a great car, it’s a golf without all the euroboy showmanship.

No plaid=no win

Sponsored

 

[junits15]

No plaid=no win

When I got the ST the combination of plaid seats, good stereo, and manual was not an option on a golf.

A total let down

 

[Duece McCracken]

When I got the ST the combination of plaid seats, good stereo, and manual was not an option on a golf.

A total let down

Total BS, damn krauters

 

[junits15]

Also initial messing around with the WMI controller is showing that all systems are working! No leaks either!

All sensors, the solenoid and the pump work. I took the hoses off the injection plate set the pump duty to 100% and verified that we had WM flow.

This is after I tested the pump, you can see it sprayed fluid all over the motor.

The flow sensor and pressure sensor need to be calibrated, and I also need to figure out the torque to WMI mapping in this controller. It seems like I can choose to PWM the pump, PWM the solenoid or both. Not sure how that works.

I’m very happy though, no leaks, and all systems working is an incredible thing to see after working on this for months.

The syvecs WMI module is very complicated for sure.

 

[junits15]

It’s pressure driven, which makes sense. So first you map pump duty and rpm to output pressure so the controller knows what duty to go to for what pressure you command.


Then you map pressure and rpm to driver demand torque. So that way when you request a torque with the pedal that requests the WMI controller to set the pump to that pressure.


So
Pedal requests torque > torque requests pressure > pressure requests pump duty.


Neato


It has closed loop mode too where it can use the pressure sensor as part of a feedback loop to hit the requested pressure accurately. However that doesn’t really seem important right now, I’m leaving open loop.

 

[junits15]

More messing around and found a few more things.

1. The controller is very flexible without being overly complicated, the more time I spend with it the more it makes sense. This is likely the case with any aftermarket ECU I would assume, since its designed to be individually tuned, the interface/and methodology is simple to grasp. Wish aftermarket ECU's were more commonplace and could pass emissions.
2. AEM and Snow use PWM to control the pump directly while holding the solenoid wide open, this Syvecs controller allows more:
   1. Open loop PWM pressure driven pump control. You ask the pump for a pressure at a given torque and RPM, which translates to a duty cycle which you map.
   2. Closed loop PWM pressure driven pump control. You calibrate the pressure sensor, then the system uses it to automatically adjust the pump duty until it hits your requested pressure.
   3. On/Off pump control. Very simple thresholding, pump turns on full blast when the desired RPM, Torque, and Throttle Position are met.
   4. Open loop PWM solenoid control, the controller adjusts the signal going to the solenoid based on RPM and Throttle position.

Number 2 is the biggest and most interesting thing about this unit. I'm using a combination of 3 and 4, so I command full system pressure after 240 ft-lbs of torque is requested, the throttle position is greater than 95%, and the RPM is greater than 1500. Then I open the solenoid based on RPM so long as the throttle remains above 95%.

This is nice because it means if you free rev the car, it wont spray water, as the torque request will be low. Basically with it set up like this it will only spray water when the car is in the OP VCT schedule. That means all part-throttle driving will be 100% identical to stock with no issues.

My pressure sensor saturates at 100PSI, its physically capable of handling 350PSI, but after 100PSI it stops reporting. I'm going to upgrade it and see if the closed loop pressure control works, that would be really nice as it will reduce the amount of time the pump spends dead-headed and help it run cooler. It will also allow me to run the solenoid wide open, as anything after 60% duty on the solenoid is pretty much wide open anyway.

 

[junits15]

We want to ignore anything below 5k RPM, I didn't start the pulls at the same speed, but they're both 3rd gear.

So this is the very first comparison, I picked up an extra 2 degrees from 5500 to redline, and I think with some adjustments I can pick up even more across the board. The solenoid only goes wide open after 5500 RPM, I'm going to have it open earlier.


I believe i need more that the 10 Gal/Hr of nozzle that I've got as crazy as that seems.

Also this is a 100% stock timing tune, I'm just spraying WMI into the stock tune and seeing what it does. The additional timing advance is evidence of its efficacy, but what's throwing me off is that I am not seeing any effect on fuel trims, I would expect to see massively negative fuel trims but I'm not seeing any difference at all. I'm running 50% meth, there should be a difference.

Its making me wonder if its really spraying at 100% duty cycle or if the solenoid is being held partially closed by something in the WMI tune. WMI should impact STFTs as its un-accounted for fuel, the fact that it isn't is weird.

 

[junits15]

There's plenty for me to play with here

• Larger nozzles
• mixtures changes
  • more water
  • more meth
  • pure water
• Pump pressure driven control

At this stage I can say that it works, its just a question of what it needs in order to work great.

 

[engineermike]

Subd.

It looks like something else is limiting your midrange timing in the log originally posted.

 

[junits15]

Subd.

It looks like something else is limiting your midrange timing in the log originally posted.

Appreciate the sub!
Here's a better picture, what's limiting me there is actually knock lol

^ WMI ON

^ WMI OFF

I'm going to open the valve fully at a lower RPM, if the system is actually injecting I'd expect to see something change, namely more timing advance lol.

 

[Chris Barnes]

I'm not surprised you're not seeing huge swings in fuel trims. I was using an Alky system with a dual nozzle setup with a 10 gal/hr and 5 gal/hr. I don't recall seeing the trims go negative more than about 7%. I was using 60/40 meth to water by weight. This system looks really intriguing. Glad you are figuring it all out so we can learn.

 

[junits15]

I'm not surprised you're not seeing huge swings in fuel trims. I was using an Alky system with a dual nozzle setup with a 10 gal/hr and 5 gal/hr. I don't recall seeing the trims go negative more than about 7%. I was using 60/40 meth to water by weight. This system looks really intriguing. Glad you are figuring it all out so we can learn.

Thank you for the reply this is really helpful!

Selfishly I want ethanol-like performance without ethanol, but for the community if I can get it to work and prove it, this setup can easily be downgraded to be around $600 in parts, which actually would make it a viable option for people who like me, have no ethanol.

But its too early to count those chickens, its still entirely possible that this is all its got. We will see!

 

[engineermike]

https://www.mustang6g.com/forums/attachments/1684346359230-png.758048/

This is what I was looking at but I misunderstood the line colors.

 

[junits15]

https://www.mustang6g.com/forums/attachments/1684346359230-png.758048/

This is what I was looking at but I misunderstood the line colors.

Ahh yes that was some octane booster stuff I played with a while ago. I should have left the labels on that to make it easier to read.

 

[junits15]

I'm not surprised you're not seeing huge swings in fuel trims. I was using an Alky system with a dual nozzle setup with a 10 gal/hr and 5 gal/hr. I don't recall seeing the trims go negative more than about 7%. I was using 60/40 meth to water by weight. This system looks really intriguing. Glad you are figuring it all out so we can learn.

Do you have any other info on your setup? It's helpful to see how other people have done this and what they used it for. I assume you used it boosted?

Sponsored