GDI - High Pressure Fuel Pump Control

It's all about the code!
Post Reply
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

GDI - High Pressure Fuel Pump Control

Post by AndreyB »

Direct Injection is giving us a completely new domain - now we need to control fuel pressure. There is the strategy of deciding on target fuel pressure based on load, RPM etc and there is the tactics of actually achieving that desired fuel pressure.

A relevant picture from Slack, this fuel entry solenoid is also peak and hold.
Attachments
high_pressure_fuel_pump.jpg
high_pressure_fuel_pump.jpg (1.17 MiB) Viewed 15465 times
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
puff
contributor
contributor
Posts: 2961
Joined: Mon Nov 11, 2013 11:28 am
Location: Moskau

Re: GDI - High Pressure Fuel Pump Control

Post by puff »

where is the fuel pressure sensor?
i'd love to see the circuitry for that one.
mck1117
running engine in first post
running engine in first post
Posts: 1493
Joined: Mon Jan 30, 2017 2:05 am
Location: Seattle-ish

Re: GDI - High Pressure Fuel Pump Control

Post by mck1117 »

puff wrote:
Tue Feb 04, 2020 10:12 pm
where is the fuel pressure sensor?
i'd love to see the circuitry for that one.
The fuel pressure sensor is on the fuel rail. They're not particularly exciting, actually: just normal linear analog pressure sensors with two outputs.

Here's a Ford Fiesta ST as an example:

Image

1. fuel pump
2. fuel injector connectors (injectors aimed straight down in to head)
3. fuel pressure sensor

edit: here's the video that's snipped from, showing disassembly of the DI system on that car:
puff
contributor
contributor
Posts: 2961
Joined: Mon Nov 11, 2013 11:28 am
Location: Moskau

Re: GDI - High Pressure Fuel Pump Control

Post by puff »

and what's the nominal pressure? by any chance, do you have part numbers/pictures of those pressure sensor and solenoid valve?

I still have a similar challenge - build a cheap diy version of FrankenCIS, which has essentially the same concept: control fuel pressure (withstanding up to 150psi) based on coolant temperature, load and RPM using a certain type of fuel injector instead of a solenoid valve.
In FrankenCIS they suggest using megasquirt to control it - still not sure if it was P&H or the normal one.
Last edited by puff on Wed Feb 05, 2020 5:57 am, edited 1 time in total.
mck1117
running engine in first post
running engine in first post
Posts: 1493
Joined: Mon Jan 30, 2017 2:05 am
Location: Seattle-ish

Re: GDI - High Pressure Fuel Pump Control

Post by mck1117 »

Here are the tables that define the fuel flow/pressure controls on my 2013 Ford Focus ST.

Image

Observations:

- Injector flow is proportional to the square root of the pressure (within about 2%, this is true of all injectors). Roughly g/s = 0.142 * sqrt(kpa) for these injectors.
- Injection pressure is higher under higher loads and speeds (duh): around 250psi at idle, and 2000 PSI under load at speed

These two effects combine to give a pulse width that's relatively constant: it's always in the 1-5ms range, because for direct injection you have to. You only have limited time to inject fuel, so you have to get it done in a hurry, but waste a minimum of extra power driving the pump harder (higher pressure).

Another observation is that it always runs at high pressure (~2000psi) for about 30 seconds after engine start, at which point it switches to the table. You can audibly hear this happen outside the car, because the pump gets silent for around 1 second while the fuel pressure drops (being emptied out the injectors - the pump can only increase pressure in the rail).
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: GDI - High Pressure Fuel Pump Control

Post by AndreyB »

Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: GDI - High Pressure Fuel Pump Control

Post by AndreyB »

@abricos made a step forward - OEM box from BMW 760 is in charge of high voltage injectors for now. This gives us a baseline and a way to test the fuel rig before we migrate to rusEfi own high voltage injection control.

Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
Abricos
contributor
contributor
Posts: 849
Joined: Mon Aug 18, 2014 12:32 am
Location: Carteret, NJ 07008

Re: GDI - High Pressure Fuel Pump Control

Post by Abricos »

Flow test ... 20bar pressure ...
* delay 100, cylinder #3, 5ms ON, 1000ms OFF, repeat 2 times
* fuelbench2 100 3 5 1000 2

dammogreen
Posts: 2
Joined: Sun Jun 07, 2020 9:37 pm

Re: GDI - High Pressure Fuel Pump Control

Post by dammogreen »

Hello , I've been lurking and just registered.

Maybe I missed it but I am not sure if anyone has mentioned that the solenoid that controls fuel pressure on GDI pumps is *timed* to the driving lobes, By ecu control. It's not just a simple untimed pwm pid drive loop.

It was not something I had envisioned when looking into GDI injector and pump control, but makes sense.

Just trying to save people some dev time. read below: https://www.pi-innovo.com/gdi-fuel-pump-control/
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: GDI - High Pressure Fuel Pump Control

Post by kb1gtt »

Yeah the pumps can be a pain. I understand that you can typically over drive the pump which simplifies it but at the cost of shorter pump life and lost hp. For now there is a flow bench that uses a brake pump to generate pressure. The proto PCB had been getting broken on the east coast then repaired on the west coast. The pcbs have been getting shipped back and forth. We are making progress. Little by little we are chipping away at it. Thanks for the discussion.

Do you have a test bench or interest in gdi?
Welcome to the friendlier side of internet crazy :)
dammogreen
Posts: 2
Joined: Sun Jun 07, 2020 9:37 pm

Re: GDI - High Pressure Fuel Pump Control

Post by dammogreen »

Sorry no test bench yet, eventually I may build a GDI driver box to be driven by a stand alone ecu. Whenever I get to that I will share any information.
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: GDI - High Pressure Fuel Pump Control

Post by AndreyB »

dammogreen wrote:
Mon Jun 08, 2020 12:54 am
eventually I may build a GDI driver box to be driven by a stand alone ecu.
Have you seen our progress on https://rusefi.com/forum/viewtopic.php?f=4&t=1337? We have some relevant hardware and even a bit of software. Where are you located? Maybe a "hello world" thread? :)
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
mck1117
running engine in first post
running engine in first post
Posts: 1493
Joined: Mon Jan 30, 2017 2:05 am
Location: Seattle-ish

Re: GDI - High Pressure Fuel Pump Control

Post by mck1117 »

dammogreen wrote:
Sun Jun 07, 2020 9:43 pm
Maybe I missed it but I am not sure if anyone has mentioned that the solenoid that controls fuel pressure on GDI pumps is *timed* to the driving lobes, By ecu control. It's not just a simple untimed pwm pid drive loop.
There are actually both types. Most GDI are what you describe, where you control the closing of the inlet valve to the pump to decide how much volume gets forced in per cycle. Some diesel pumps are actually continuous (asynchronous) PWM controlled.
Post Reply