Feature Request: DeadTime vs. fuel pressure differential

Testing and development of Megasquirt 3

Moderators: jsmcortina, muythaibxr

Post Reply
Dennis_Zx7r
Experienced MS/Extra'er
Posts: 374
Joined: Thu May 26, 2016 1:25 pm
Location: Germany

Feature Request: DeadTime vs. fuel pressure differential

Post by Dennis_Zx7r »

With a pump with fixed fuel pressure, there's a non-constant fuel pressure differential due to the the varying MAP while the engine is operating. In Fuel Pump and Pressure Control, there's already a correction built in for this (Fixed Pressure setting), which compensates by proportionally increasing/decreasing the calculated PW(-DT?) if I understand it correctly.

However, a different differential pressure also means the Deadtime of the injectors changes. For example, with the injectors I have here, DT is about .9ms on 380kpa fuel pressure differential (FPD), compared to .62ms with 300kpa (same voltage).
I think a FPD-dependent factor may be sufficient, similiar to how the voltage-correction was done in MS2. A curve like DT% vs Voltage in MS3 would be even better.
I think this would be nice for better low load control when using unreferenced pumps, especially with big injectors where DT is a significant part of the fueling equation. And of course for PWM controlled pumps, if different FPDs are targeted on purpose.
My project: Link
AbatelliCristian
Super MS/Extra'er
Posts: 855
Joined: Sun Oct 10, 2010 6:33 am

Re: Feature Request: DeadTime vs. fuel pressure differential

Post by AbatelliCristian »

I confirm. The fuel pressur is very influential for deadtime.... Some injector with high pressur not open and make missfire...
CRSTune
Experienced MS/Extra'er
Posts: 338
Joined: Thu Oct 20, 2016 9:12 pm
Location: Manassas, Virginia

Re: Feature Request: DeadTime vs. fuel pressure differential

Post by CRSTune »

I bet this could easily be accomplished by changing the current deadtime curve into a 2D table (Voltage vs Pressure).
I'm in the VA/DC area! Let me know if I can help you locally! I offer tuning and troubleshooting services.
http://www.msextra.com/forums/viewtopic ... 37&t=64269
Email me directly at CRSTune@gmail.com

Personal Vehicle:
'92 Nissan 240sx, KA24DET, GT2860RS, MS3X, Coil-on-Plug
Dennis_Zx7r
Experienced MS/Extra'er
Posts: 374
Joined: Thu May 26, 2016 1:25 pm
Location: Germany

Re: Feature Request: DeadTime vs. fuel pressure differential

Post by Dennis_Zx7r »

CRSTune wrote:I bet this could easily be accomplished by changing the current deadtime curve into a 2D table (Voltage vs Pressure).
I have no idea what's better for the implementation. I guess a table would also work.
A curve might be more user friendly though, especially as it could just be ignored when using a compensated pump or if one doesn't care about it.
My project: Link
David_G
Helpful MS/Extra'er
Posts: 98
Joined: Wed Apr 02, 2014 1:14 am

Re: Feature Request: DeadTime vs. fuel pressure differential

Post by David_G »

I would love to see that feature too!

I have a MAP-referenced fuel pressure regulator but even with that, fuel pressure is not entierly constant.
With 1000cc injectors, my actual PW are very low and therefore DeadTime has a very high influence. Its so high, that when upping the fuel pressure from 43.5psi to 75psi for testing purposes (and PW being constant), the AFR actually gets LEANER by 10%.

Of course as PW get higher, that effect gets smaller, so there is no way I can compensate for that in the current firmware.
whittlebeast
Super MS/Extra'er
Posts: 2221
Joined: Tue May 04, 2004 8:20 pm
Location: St Louis
Contact:

Re: Feature Request: DeadTime vs. fuel pressure differential

Post by whittlebeast »

I could see going to a table with voltage up the side, fuel pressure differential along the bottom and dead time in the field but how would you ever figure out how to populate the field? Does anybody else do anything similar?

Andy
CRSTune
Experienced MS/Extra'er
Posts: 338
Joined: Thu Oct 20, 2016 9:12 pm
Location: Manassas, Virginia

Re: Feature Request: DeadTime vs. fuel pressure differential

Post by CRSTune »

Both the AEM Infinity and the Haltech Elite implement fuel flow and deadtime tables in their software. They're completely optional of course.

The absolute best way to populate the tables would be to bench test the fuel injector flow. Next best way would be using manufacturer provided data. Both Injector Dynamics and Fuel Injector Clinic provide deadtime based on pressure and voltage.
I'm in the VA/DC area! Let me know if I can help you locally! I offer tuning and troubleshooting services.
http://www.msextra.com/forums/viewtopic ... 37&t=64269
Email me directly at CRSTune@gmail.com

Personal Vehicle:
'92 Nissan 240sx, KA24DET, GT2860RS, MS3X, Coil-on-Plug
Dennis_Zx7r
Experienced MS/Extra'er
Posts: 374
Joined: Thu May 26, 2016 1:25 pm
Location: Germany

Re: Feature Request: DeadTime vs. fuel pressure differential

Post by Dennis_Zx7r »

David_G wrote:[...]Its so high, that when upping the fuel pressure from 43.5psi to 75psi for testing purposes (and PW being constant), the AFR actually gets LEANER by 10%.
Of course as PW get higher, that effect gets smaller, so there is no way I can compensate for that in the current firmware.
Something like this is exactly why I think this makes sense, especially for PWM-controlled pumps and varying fuel pressures. The MS already (at least partially) supports PWM for fuel pumps. It also has a correction for different fuel pressures, but the change in DT is not respected. So if one uses this correction, it effectively adjusts the fueling differently for high PWs than for low PWs.
Personally, I'm nearly done building an external PWM control-loop to lower fuel pressure in Idle and have significantly higher fuel pressure under full load. This is already standard practice for at least 8 years on some mass-produced bikes like the BMWs (no idea about car engines). DT is a significant factor in these applications as dutycycle is kept very low on purpose.
My project: Link
Matt Cramer
Super MS/Extra'er
Posts: 17499
Joined: Thu Apr 16, 2009 8:08 pm

Re: Feature Request: DeadTime vs. fuel pressure differential

Post by Matt Cramer »

whittlebeast wrote:I could see going to a table with voltage up the side, fuel pressure differential along the bottom and dead time in the field but how would you ever figure out how to populate the field? Does anybody else do anything similar?

Andy
It would require a lot of flow bench testing with an adjustable voltage power supply. But some injector shops have this data available.
Matt Cramer -1966 Dodge Dart slant six running on MS3X
David_G
Helpful MS/Extra'er
Posts: 98
Joined: Wed Apr 02, 2014 1:14 am

Re: Feature Request: DeadTime vs. fuel pressure differential

Post by David_G »

One could also calculate what the theoretical AFR change with changing Fuel pressure/injector flow should be and adjust the pressure-based Deadtimes until the actual AFR matches the calculated ones.
Basically the same way how I found out that my above problem is Deadtime related.

Or as you said, take the data from the injector shops (Deatschwerks also has this info on their website, so at least the biggest shops have this data published).
Dennis_Zx7r
Experienced MS/Extra'er
Posts: 374
Joined: Thu May 26, 2016 1:25 pm
Location: Germany

Re: Feature Request: DeadTime vs. fuel pressure differential

Post by Dennis_Zx7r »

I did some quick graphs with data obtained from Injector Dynamics for pressure differentials between 40 and 75psi with 45psi as 100% pressure differential. Used data from their ID725, ID1000 and ID2000. The number stands for the flow.
Seems to be fairly linear at least with these injectors, so I think interpolating/extrapolating would work out ok and not too many measurements would have to be taken for DIY-calibration.
Looking at this data, I think a table would be best as the slope varies quite a bit with voltage.

Image
My project: Link
Post Reply