Pages: [1]
Author Topic: MED 9.1 Passat low fuel pressure control - open or closed loop?  (Read 1028 times)
russian
Newbie
*

Karma: +0/-0
Offline Offline

Posts: 7


« on: October 20, 2020, 10:01:48 PM »

I am trying to build my own ECU and that makes me interested in how does Bosch ECU control it's low pressure fuel pump so that I can repeat that.

I know there is a 20Hz signal from ECU to fuel pump control module (not actual fuel pump) and my question is if it's open loop or closed loop? a curve or a map? what are relevant table names?

https://www.youtube.com/watch?v=swU3I_cTnNY
Logged
gman86
Hero Member
*****

Karma: +42/-127
Offline Offline

Posts: 699


« Reply #1 on: October 21, 2020, 02:23:12 PM »

I am trying to build my own ECU and that makes me interested in how does Bosch ECU control it's low pressure fuel pump so that I can repeat that.

I know there is a 20Hz signal from ECU to fuel pump control module (not actual fuel pump) and my question is if it's open loop or closed loop? a curve or a map? what are relevant table names?

https://www.youtube.com/watch?v=swU3I_cTnNY

Open loop. KFNTBKS and KFFLAF control the pressure request > duty cycle conversion.
Logged
russian
Newbie
*

Karma: +0/-0
Offline Offline

Posts: 7


« Reply #2 on: October 21, 2020, 10:41:19 PM »

Thank you, that helps!

Logged
BlackT
Hero Member
*****

Karma: +45/-26
Offline Offline

Posts: 914



« Reply #3 on: October 22, 2020, 01:53:09 AM »

Looks like great work, keep going. I am also building my own ECU.
What MCU are you using?
Logged
curamrdan
Full Member
***

Karma: +1/-0
Offline Offline

Posts: 93


« Reply #4 on: October 22, 2020, 02:40:02 AM »

But ecu measuring fuel pump pressure, comparing to an target and then makes changes on PWM signal, so that means its closed loop, isnt?
Logged
Blazius
Hero Member
*****

Karma: +50/-29
Offline Offline

Posts: 890



« Reply #5 on: October 22, 2020, 03:32:00 AM »

Thats a rusefi is it not? Then stm32 with tle8888 ?
« Last Edit: October 22, 2020, 03:41:07 AM by Blazius » Logged
russian
Newbie
*

Karma: +0/-0
Offline Offline

Posts: 7


« Reply #6 on: October 22, 2020, 05:18:32 PM »

But ecu measuring fuel pump pressure, comparing to an target and then makes changes on PWM signal, so that means its closed loop, isnt?
yes, comparing actual pressure to target pressure in order to adjust PWM is closed loop.

this car has very different setups for low-pressure and high-pressure. right now I am trying to start it for the first time without high pressure pump control.
Logged
russian
Newbie
*

Karma: +0/-0
Offline Offline

Posts: 7


« Reply #7 on: October 22, 2020, 05:20:11 PM »

Thats a rusefi is it not? Then stm32 with tle8888 ?
Wow, thank you for recognizing it Smiley yes, stm32 MCU and tle8888 which does not really help here and PT2001 for GDI control.
Logged
russian
Newbie
*

Karma: +0/-0
Offline Offline

Posts: 7


« Reply #8 on: October 22, 2020, 10:04:13 PM »

What about fuel rail pressure X injector flow curve or map? I can use map names of maybe even data screen shots pretty please.

I was able to run the engine for a second with pressure build by OEM ECU but at the moment I am flying totally blind in terms of injection duration.
Logged
curamrdan
Full Member
***

Karma: +1/-0
Offline Offline

Posts: 93


« Reply #9 on: October 23, 2020, 04:00:07 AM »

If you are trying to make ecu for this engine, my experience with LPFP is the least important to solve for you. I just did some testing in real, and on one car, i modified PWM signal to controller, to be as a switch and instead pump it just switching relay, which sending full voltage before module to a stock pump. So pump runs 100% all time (hitting 5,8bar regulator value in fuel filter)   and then i modified many pressure request, trigger values to ecu target pressure just to follow pressure drop in WOT by rpm and not throw DTC in occasional few bar pressure difference in possitive way if engine is not in full load. That way i can run almost 400ps on stock fuel pump and faulty old module Smiley
As rail pressure is seperate system, i did not found any influence on injected fuel, running 1bar or 5bar before rail.
So in you situation, just switch pump by relay and not bother with it until solve engine management itself.
Logged
russian
Newbie
*

Karma: +0/-0
Offline Offline

Posts: 7


« Reply #10 on: October 23, 2020, 08:57:18 AM »

For low pressure I just do 50% duty cycle, for high pressure I am doing nothing so far.

Next step is to see if I can idle it on just low pressure fuel pump, that's why I am looking for at least some injector flow data.
Logged
curamrdan
Full Member
***

Karma: +1/-0
Offline Offline

Posts: 93


« Reply #11 on: October 23, 2020, 12:39:32 PM »

OEM when there is some problem with Highpressure regulator, its just run 7bar on rail with limited throttle opening.

KLTIKRPR  is correction accroding rail pressure. Its same for 200hp or 265hp engine versions.
KLTVTSV is deadtime correction accroding rail pressure.
Logged
gman86
Hero Member
*****

Karma: +42/-127
Offline Offline

Posts: 699


« Reply #12 on: October 23, 2020, 10:53:23 PM »

But ecu measuring fuel pump pressure, comparing to an target and then makes changes on PWM signal, so that means its closed loop, isnt?

It measures low pressure to adjust the N276 on the high pressure pump. The ECU doesn't change duty cycle based on low pressure. It has a fixed lookup for duty against requested pressure vs fuel mass flow. If it over or under shoots, it won't correct. It'll just flag a fault if it trips a threshold.
Logged
curamrdan
Full Member
***

Karma: +1/-0
Offline Offline

Posts: 93


« Reply #13 on: October 24, 2020, 02:22:56 AM »

It measures low pressure to adjust the N276 on the high pressure pump. The ECU doesn't change duty cycle based on low pressure. It has a fixed lookup for duty against requested pressure vs fuel mass flow. If it over or under shoots, it won't correct. It'll just flag a fault if it trips a threshold.

My experience, when logging LFPF pressure and module DC,  i clearly see, when pressure undershooting, ecu rise the DC even up to alloved maximum limit(for example 92%). And also, when i rise KFFLAF values too much, i can see overshoot of pressure in situation wher fuel mass flow quickly rising(and fuel pump is strong enought) and then ecu adjust DC again, itself.
Maybe on other engines or software versions it behave differently, if so, its even degrading importance of managing LPFP system Smiley
Logged
Pages: [1]
  Print  
 
Jump to:  

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Page created in 0.021 seconds with 16 queries. (Pretty URLs adds 0s, 0q)