waterinjection.info  

Go Back   waterinjection.info > Aquamist specific system Q and A and new products > Aquamist HFS-4 forum

Reply
 
Thread Tools Display Modes
  #1  
Old 26-06-2017, 02:49 AM
makro86 makro86 is offline
Junior Member
 
Join Date: Jan 2012
Location: Melbourne
Posts: 7
Default Port injection - frequency of FAV

Hi guys,

Long time HFS-4 v1 owner here - the system is great and it hasn't missed a beat. I've been using it with one pre-TB jet up to this point and am now looking to go to port injection for a new engine I am completing at the moment.

I have been discussing the port injection concept with my mechanic friend who is also looking at purchasing a HFS-4. We've searched the forums and web in general to try find some information on this, however have not had much luck in solving the following.

It seems the recommended frequency for the FAV is approximately 30Hz. My assumption here is that the HFS-4 will also be controlling the FAV at this frequency. It is a fixed frequency and the HFS-4 uses PWM duty cycle to regulate the amount of WM injected at any point.

So when switching to a port injection configuration, this leads us to the following scenario...
RPM: 9000
FAV outlet connected to port injection manifold with 4 outlets, 1 to each intake runner.
9000 / 60 = 150 revs per second.
150 / 2 = 75 intake strokes every second, 1 every 2 engine revolutions.

Now if the FAV is injecting at 30Hz, wouldn't that result in the injections of WM being out of phase with the engine?
I can see from the great results that people have with port injection that it definitely does work, so I'm not doubting that, however we are really interested in how it actually ends up working.

My only guess here is that the pulse from the FAV is somehow spread between the intake events.

Also, what is the highest frequency the FAV can be run at? My friend is looking to use his Motec to directly control the FAV as a supplementary injector. To state the obvious, frequency and duty cycle is going to change dynamically as the RPM increases, so we're looking at something like a 65 - 75 Hz / pulses per second rate for the FAV.

What would be the best practice for this type of setup to get the best distribution of WM across all cylinders?

Any guidance or correction to my logic above greatly appreciated.

Regards,
Matt
Reply With Quote
  #2  
Old 26-06-2017, 10:19 AM
theboostshack theboostshack is offline
Junior Member
 
Join Date: Mar 2017
Location: Sydney
Posts: 14
Default Re: Port injection - frequency of FAV

G'day Makro86,
We are also interested in this topic. Currently in the process off upgrading to larger, 260 cc, direct port jets. As such, the plan is to set up a custom table in the ecu to with a pwm output to control injection. We recently had a conversation with a fellow forum member how has implemented this strategy. Below is the information he provided about how he has it set to. Hope you find it helpful and please keep us informed as you progress.
Cheers.

"My AquaMist control with the Link has evolved over time and it’s pretty complicated but it seems to work fairly well despite the complexity.

First off, I connected the Aqm’s input wires up to an Aux output, rather than tap into an injector. Then, to control the rate of the Aqm’s injection, I set the Aux output to GP PWM and created a 3D table with Axes on RPM and MGP to control it. I then proceeded to do some testing, with the nozzles hanging off to the side of the car, to see at what PWM value the FAV started pulsing and at what PWM value the FAV is wide open with no pulsing. What I learned is that with my setup, with a PWM output value of 18, the pump activates, at PWM=20, the FAV just starts to pulse, and at 82, the FAV is wide open. So I set up my PWM table with 18’s a few psi below where I want it to start injecting, that way the line is pressurized as soon as the FAV starts pulsing. Those values will probably be different for you because they are determined by those pots on the Aqm gauge.

Here’s where it starts getting interesting/complicated

Rather than just making a second fuel table for when the methanol is spraying (what most people do I think), I connected the Aqm’s flow sensor output (0-5v) up to an analog input on the Link and I use the flow reading as an axis on a 4D fuel table to drop fuel out and also a 4D Ignition table, to advance the timing based on the flow sensor’s reading. This way if a nozzle gets clogged and the flow is reduced, the system will respond dynamically. It also allows me to change or add/remove nozzles with minimal retuning. The one condition it doesn’t protect from would be if a methanol line came off or cracked after the flow sensor. But I have other ways of protecting against that.

If you are not injecting a substantial amount and/or have your methanol nozzle(s) close to the throttle body or have port injection, that may be all you need. But I have 3 nozzles right at the intercooler exit for maximum cooling and mixing. What I experienced was that during a prolonged pull, the mix would get gradually richer as the 2.5 feet of intake piping and intake manifold wetted with methanol. Some people may just learn to live with that effect and tune it slightly lean so as it richens up to doesn’t get too rich. But stuff like that bothers me, so I implemented a flow timer. Basically once the flow meter registers 0.9 volt, I start a timer. Then I use that timer as an axis on my 4D fuel table, the other axis being the flow meter. So as the timer counts up, I increase the % of gas pulled.

That definitely helped keep my AFRs more stable, but it wasn’t enough. I also set up a second fuel table as an Overlay table with axes on MGP and AFR. So if AFR still goes lean or rich, this table provides additional correction. I also have Closed Loop Lambda turned on at all times with a high gain so it responds quickly. With all these compensations in place, the AFR tracks the target pretty well though it’s still not perfect 100% of the time and probably never will be.

For protections, I have the Aqm failsafe output signal wired into the Link and if that’s triggered, all the methanol-related tables and adjustments get switched off. In addition, I have closed-loop boost setup with AFR as an axis so if it drifts lean, the boost is automatically dialed back. I use AFR as an axis on a 3D MAP limit as well but under normal conditions the MAP limit should never get tripped since the boost target is set lower than the MAP limit.

I currently have three nozzles in place: the 0.8, 0.9, and 1.0 and at full spray the output from the flow meter is just under 4 volts and that is about the limit for my setup. I tried installing 1.2mm nozzles but they resulted in very poor atomization. So bad in fact, that even many seconds after a pull, the engine could suddenly go extremely rich as methanol laying in the piping suddenly found its way into the engine. I did crank up the pump pressure by adjusting the regulator screw so I figure I’m probably flowing around 1200 ~ 1300cc/min."
Reply With Quote
  #3  
Old 26-06-2017, 12:52 PM
makro86 makro86 is offline
Junior Member
 
Join Date: Jan 2012
Location: Melbourne
Posts: 7
Default Re: Port injection - frequency of FAV

Hi boostshack,

Thanks for your reply. That's quite an interesting and complex setup there.
My preference would be to eliminate a lot of that extra tuning by just going to port injection and a smaller jet up stream, maybe pre-turbo.

Looking forwards to some other replies regarding the frequency and spread of WM across cylinders which may help us both out

Regards,
Matt


Quote:
Originally Posted by theboostshack View Post
G'day Makro86,
We are also interested in this topic. Currently in the process off upgrading to larger, 260 cc, direct port jets. As such, the plan is to set up a custom table in the ecu to with a pwm output to control injection. We recently had a conversation with a fellow forum member how has implemented this strategy. Below is the information he provided about how he has it set to. Hope you find it helpful and please keep us informed as you progress.
Cheers.

"My AquaMist control with the Link has evolved over time and it’s pretty complicated but it seems to work fairly well despite the complexity.

First off, I connected the Aqm’s input wires up to an Aux output, rather than tap into an injector. Then, to control the rate of the Aqm’s injection, I set the Aux output to GP PWM and created a 3D table with Axes on RPM and MGP to control it. I then proceeded to do some testing, with the nozzles hanging off to the side of the car, to see at what PWM value the FAV started pulsing and at what PWM value the FAV is wide open with no pulsing. What I learned is that with my setup, with a PWM output value of 18, the pump activates, at PWM=20, the FAV just starts to pulse, and at 82, the FAV is wide open. So I set up my PWM table with 18’s a few psi below where I want it to start injecting, that way the line is pressurized as soon as the FAV starts pulsing. Those values will probably be different for you because they are determined by those pots on the Aqm gauge.

Here’s where it starts getting interesting/complicated

Rather than just making a second fuel table for when the methanol is spraying (what most people do I think), I connected the Aqm’s flow sensor output (0-5v) up to an analog input on the Link and I use the flow reading as an axis on a 4D fuel table to drop fuel out and also a 4D Ignition table, to advance the timing based on the flow sensor’s reading. This way if a nozzle gets clogged and the flow is reduced, the system will respond dynamically. It also allows me to change or add/remove nozzles with minimal retuning. The one condition it doesn’t protect from would be if a methanol line came off or cracked after the flow sensor. But I have other ways of protecting against that.

If you are not injecting a substantial amount and/or have your methanol nozzle(s) close to the throttle body or have port injection, that may be all you need. But I have 3 nozzles right at the intercooler exit for maximum cooling and mixing. What I experienced was that during a prolonged pull, the mix would get gradually richer as the 2.5 feet of intake piping and intake manifold wetted with methanol. Some people may just learn to live with that effect and tune it slightly lean so as it richens up to doesn’t get too rich. But stuff like that bothers me, so I implemented a flow timer. Basically once the flow meter registers 0.9 volt, I start a timer. Then I use that timer as an axis on my 4D fuel table, the other axis being the flow meter. So as the timer counts up, I increase the % of gas pulled.

That definitely helped keep my AFRs more stable, but it wasn’t enough. I also set up a second fuel table as an Overlay table with axes on MGP and AFR. So if AFR still goes lean or rich, this table provides additional correction. I also have Closed Loop Lambda turned on at all times with a high gain so it responds quickly. With all these compensations in place, the AFR tracks the target pretty well though it’s still not perfect 100% of the time and probably never will be.

For protections, I have the Aqm failsafe output signal wired into the Link and if that’s triggered, all the methanol-related tables and adjustments get switched off. In addition, I have closed-loop boost setup with AFR as an axis so if it drifts lean, the boost is automatically dialed back. I use AFR as an axis on a 3D MAP limit as well but under normal conditions the MAP limit should never get tripped since the boost target is set lower than the MAP limit.

I currently have three nozzles in place: the 0.8, 0.9, and 1.0 and at full spray the output from the flow meter is just under 4 volts and that is about the limit for my setup. I tried installing 1.2mm nozzles but they resulted in very poor atomization. So bad in fact, that even many seconds after a pull, the engine could suddenly go extremely rich as methanol laying in the piping suddenly found its way into the engine. I did crank up the pump pressure by adjusting the regulator screw so I figure I’m probably flowing around 1200 ~ 1300cc/min."
Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +1. The time now is 08:58 PM.


Powered by vBulletin® Version 3.8.4
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.