[Gmecm] [Diy_efi] direct injection

Tue Oct 4 20:09:25 UTC 2011

Phil....

You assumption that the reason for DI EFI is to run lean is not quite correct.  Most DI systems in stringent emissions countries are meant to run in "homogenious Charge" mode, which means stoichiometery for the USA.
There are some in Europe that are experimenting with true lean combustion.  These are stratified charge engines and the combustion chambers, compression ratios and location and sray type of the DI injectors are different than the homogenious charge engines.

Your "leaness" of operations comes from the start and warmup phase, as there is no port wall-wetting with DI.  It all goes into the chamber.  That and the positioning and timing of the injectors gives very good combustion characteristics, so minimal warmup hydrocarbons are generated.
This and the consistent and better combustion lends itself to higher compression ratios for a given octane fuel without excessive spark knok and low misfire incidence.

Also, most DI systems the injectors are high voltage (boosted voltage to 80 to 100V) peak and hold systems, and this fast response allows some system to do multiple injections per cylinder event (sort of like latest diesel technology),  This further increases the output of the engine while maintaining low exhaust emissions and either avoiding or toleraring EGR better.  When mixed with variable full-range valve timing for both intake and exhaust, it allows small engines to generate bigger toque and HP outputs while still achieving the extremely low emissions required today.

So if you want to create or adapt a DI system for your engine, you need to look at all the parts involved, including the induction system, the valve timing system, and the shape and layout of the combustion chamber and intake/exhaust ports and the location of the injector and how they cool it!

A final note.  The high pressure pumps generally have a PWM to modulate the output pressure, and the timing of the pump lobes for pressure generation is also timed to reduce stress on the camshaft(s).

Hope this helps.

Hank

-----Original Message-----
From: Phillip Kuhn <pmkls1 at yahoo.com>
To: David Allen <davida1 at hiwaay.net>
Cc: Milano Guy <milanoguy at gmail.com>; Do It Yourself fuel injection <diy_efi at diy-efi.org>; gmecm at diy-efi.org <gmecm at diy-efi.org>
Sent: Tue, Sep 27, 2011 11:32 pm
Subject: Re: [Gmecm] [Diy_efi] direct injection

     Just an fyi before I type my reply, I sent that last e-mail out last thursday and it just went through. For some reason my emails sent to the list are taking forever to get through.

     David, I believe that you are correct in your assumption because I am sure that the engineers are trying to get the engines to run on as lean a mixture as possible. To jump slightly off topic for a second . The very unrealistic demands regarding auto manufacturers increasing their CAFE ratings is placing a large strain on the auto makers, more specifically the engineers. They are resorting to unconventional means and pushing limits to try to squeeze every last mpg out of every car that they can.  Right now a decent portion of driveability complaints that we see are actually considered "normal operation". The root cause of those complaints is how the ecm and tcm are programmed to operate, which of course are aimed at improving fuel economy. Having said that, I am sure that the DI engines are running pretty lean. 

     Milano Guy, I'm not sure if there is any available information that I haven't disclosed. If you have any specific questions I can do my best to answer. Honestly, we have had very little DI specific issues to date so I don't know much past the layout and operation of the systems. I forgot that the only 2 injectors that we have replaced at my shop were not the problem with the respective vehicles either and were just replaced as a shot in the dark. The only actual fuel system issues we have seen from the DI systems was the high pressure pump leaking fuel into the crankcase and causing the vehicle to set rich codes. This was on the 2.4 Ecotec engines only. The only other issues would be due to the DI, but not actually the DI itself. There have been issues with the 3.6 v6 engines developing excessive carbon deposits on the intake valves and actually causing the valves to stick at low RPM's resulting in intermittent misfires.  That issue is common enough nationwide to result in a bulletin being released, but overall is still rare and can be at least partially attributed to the maintenance and driving habits of the owners. Other than that and some updated calibrations the systems in the GM vehicles seem to be operating well. One complaint that we see a lot would be the ticking noise you can hear from the pump and injectors which is absolutely normal. There is a significant amount of noise dampening material used around the pump and injectors and you can still easily hear the ticking, especially when the engine is cold. This is of course due to the pressures the system operates at along with the way that the pump operates in respect to the commanded fuel pressure. Which, again, this is completely normal, but for those unfamiliar with DI vehicles it "seems" like an engine problem that causes them to believe that the engine has an issue. Glad that I can contribute to the list for a change. And did everyone get the e-mail that I sent out with links to the photos I took of some DI components ?

Phil

From: David Allen <davida1 at hiwaay.net>
To: Phillip Kuhn <pmkls1 at yahoo.com>
Sent: Tuesday, September 27, 2011 5:23 AM
Subject: Re: [Diy_efi] direct injection

Interesting information, and it does make sense.

Our natural gas engines have a system where an enriched fuel/air mix is held near the spark plug and less rich in the remainder of the chamber. It allows for cleanly and completely burning mixtures that would otherwise be too lean to ignite of the mixture was completely blended.

I bet GM is doing the same thing by the placement of the fuel spray and the shape of the chamber.

David

----- Original Message ----- 
From: Phillip Kuhn 
To: ScottyGrover at aol.com 
Cc: Do It Yourself fuel injection 
Sent: Thursday, September 22, 2011 9:39 PM
Subject: Re: [Diy_efi] direct injection

I know that this topic has gone kind of cold, but I spoke with a field engineer that came to the shop today and he had some information I thought I would share with you. When I told him about your inquiry and the information that I had been unable to obtain he also had some additional information that I was unaware of as well. He told me that the injector placement was critical and that was one of the major obstacles that took GM engineers some time to overcome. Evidently, the location of the injector in the combustion chamber and more importantly where the spray is focused is crucial to the engine performing properly. Since they were modifying existing engine designs with the intention of making as few major modifications to the cylinder heads as possible it was quite difficult to get the injectors spraying fuel into the right part of the cylinder for the engines to run right. As a matter of fact I have replaced a cylinder head on a DI 3.6 v6 recently. I noticed the injector actually came in through the side of the combustion chamber ( this was probably due to the only available location to mount the injectors ) and the top of the piston and combustion chambers were very specifically shaped. They were profiled so that the fuel spray focused into a recess in the middle of the combustion chamber directly under the spark plug. I am guessing that this was to get the flame to start in the middle of the chamber and spread outward in a controlled manner. Anyhow, I thought this information was important enough to pass on. Don't get me wrong, I would love for you to successfully utilize a DI system on your project and anyone else for that matter. I just want to provide enough info so as to aid in this being completed successfully without running into so many obstacles that anyone who attempts this will not give up. I also don't want to see someone ruin a possibly rare and/or expensive cylinder head just to mount the injector and find out that the rest of the project is too difficult to complete. Good luck and feel free to ask anymore questions that come to mind.

Phil

From: "ScottyGrover at aol.com" <ScottyGrover at aol.com>
To: diy_efi at diy-efi.org
Sent: Friday, September 9, 2011 12:15 AM
Subject: Re: [Diy_efi] direct injection

Thank you very much for the information.  I am designing a system for an air-cooled, 2.7L flat six and haven't been able to get any co-operation from Delphi or from my local dealer (he says I don't need to know.)
I have studied the patents (Delphi and others) and have a good idea of the layout but I need the physical size (particularly the diameter) and the application (HP rating, engine size, and of course the part number.)

Scotty

In a message dated 9/8/2011 7:46:41 P.M. Pacific Daylight Time, pmkls1 at yahoo.com writes:
I have been a tech at GM dealerships since 2001. I don't have the slightest clue about the other manufacturers using DI, but know a decent amount about the GM systems. So far I haven't seen any common issues at all with the GM stuff. I have seen a few injectors go bad and some issues with the high pressure pump leaking fuel into the crankcase ( just like the old mechanical pumps did), but nothing major or widespread overall. As far as the systems themselves go I will give a detailed description of how they work etc. Currently, only the Ecotec engines have DI versions, and all late model high feature v6 engines (2.8, 3.0, 3.6) are DI. They all use an in-tank electric pump running at ~60 psi to send fuel up to the high pressure pump. The high pressure pump is mounted on the engine, is driven off of a camshaft, and operates exactly like the regular old mechanical fuel pumps do ( save for the pressure output ). The entire system is a returnless system and
like most newer vehicles has no serviceable fuel filter. The supply pressure is regulated by the ecm and a fuel pump flow control module via a pwm signal to the electric in-tank pump. The fuel pressure on the high pressure side of the system is also controlled by the ecm too via an actuator mounted inside the high pressure pump. The system operating pressures on the high pressure side are 500-800 psi at idle and around 2500 psi at max output iirc. From the high pressure pump forward they use heavy stainless lines with flare and ball connections and are supposed to be a one-time use deal. Anytime we take a line loose it is supposed to be replaced, although I have reused several without any leaks. There is also a fuel rail pressure sensor located on the fuel injector rail to send pressure readings back to the ecm and there is no other means of testing pressure on the high pressure side. The injectors look like normal injectors on the fuel rail side and
use o-rings that appear to be the same size or similar to that of regular fuel injectors( I still can't figure out how the seals don't blow out at those pressures). The other end of the fuel injector is very long and narrow and uses special teflon seals to seal it into the cylinder head as they go directly into the combustion chamber through the intake side of the cylinder head. Special tools are required to install and size the teflon seals similar to the tools used for non-cut teflon seals used in transmissions. There are also special tools required to remove the injectors from the cylinder head as carbon will quickly cause them to get stuck in the close-tolerance bores. Every time the injectors are removed all of the seals have to be replaced. The injectors do require a special high voltage driver, but I do not know the specs right now. Overall, the systems operate like a regular sequential port system does and are designed similar to a diesel
system. Both fuel economy and power are improved on the engines GM is using DI on currently. The latest 3.6 in the base camaro is up to something like 317hp now. As far as being able to fit DI to any other engine, I suppose it may be possible on some engines. Even then it would require significant cylinder head modification and some fabrication. You would also have to use an ECM designed for DI and be able to modify the calibrations and flash the ecm which I am unsure if that is possible currently. Hope this answers a few questions.

Phil

----- Original Message -----
From: Avery Nisbet <anisbet at gmail.com>
To: Fred Cooke <fred.cooke at gmail.com>
Cc: ScottyGrover at aol.com; diy_efi at diy-efi.org
Sent: Thursday, September 8, 2011 4:07 PM
Subject: Re: [Diy_efi] direct injection

I have heard from VAG mechanics that they still have trouble with
this. This is with the current TSFI engines found in the audi's and VW
in the US.

Though TDI engines have been around for years I only hear of them
clogging up the EGR system not the intake ports.

I think most of this is due to EGR issues and dirty air.  Where else
would any substance that would/could coke on the intake port/valve
come from if you have DI.  The washing action of Port injection
probably helped with badly designed EGR systems in the past.

For DIY at home "testing" of DI, EGR systems could probably be avoided.

-Avery

On Thu, Sep 8, 2011 at 1:00 PM, Fred Cooke <fred.cooke at gmail.com> wrote:
> Early examples, from at least 1998, perhaps earlier, are to be avoided as
> they are usually prone to coking of the intake manifolds. Very bad coking -
> to the point of blockage! The earlier cars had a lot of trouble in many
> areas and were not reliable at all. If we're talking about DI. if we're
> talking about BMW Piezo DI, I have no idea.
>
> On Fri, Sep 9, 2011 at 7:18 AM, Avery Nisbet <anisbet at gmail.com> wrote:
>>
>> A lot the current car have these.
>>
>> GM has been using them in the eco-tec(FWD 4 cylinder) engine for a
>> while.  Im sure there a few different injector styles.
>> The prices should come down in a few years.  You may be able to find
>> specs by looking up after market replacements.
>>
>> The cruze should have these depending on market.   The later model
>> year US market cobalts had them too.
>>
>> -Avery
>>
>> 2011/9/8 Mike <niche at iinet.net.au>:
>> > All I know its pretty high pressure, the latest bmw twin turbos use this
>> > with piezo driven
>> > injectors that cost $2200+ (AUD) each !  But they do get pretty
>> > reasonable
>> > fuel consumption of
>> > around 6L/100Kms with sizable power on demand - straight six 2 to 3L,
>> > seems like the next tech step but geesh does it cost !
>> > regards
>> > Mike
>> >
>> > At 02:48 AM 9/9/2011, ScottyGrover at aol.com wrote:
>> >
>> > Has anyone done any work on this type of fuel injection? I can't get any
>> > data from Delphi, even as to the physical size (length, stem diameter.)
>> >
>> > Scotty from Hollyweird
>> > _______________________________________________
>> > Diy_efi mailing list
>> > Diy_efi at diy-efi.org
>> > http://lists.diy-efi.org/mailman/listinfo/diy_efi
>> >
>> > _______________________________________________
>> > Diy_efi mailing list
>> > Diy_efi at diy-efi.org
>> > http://lists.diy-efi.org/mailman/listinfo/diy_efi
>> >
>> >
>> _______________________________________________
>> Diy_efi mailing list
>> Diy_efi at diy-efi.org
>> http://lists.diy-efi.org/mailman/listinfo/diy_efi
>
>
_______________________________________________
Diy_efi mailing list
Diy_efi at diy-efi.org
http://lists.diy-efi.org/mailman/listinfo/diy_efi

_______________________________________________
Diy_efi mailing list
Diy_efi at diy-efi.org
http://lists.diy-efi.org/mailman/listinfo/diy_efi

_______________________________________________
Diy_efi mailing list
Diy_efi at diy-efi.org
http://lists.diy-efi.org/mailman/listinfo/diy_efi

_______________________________________________
Diy_efi mailing list
Diy_efi at diy-efi.org
http://lists.diy-efi.org/mailman/listinfo/diy_efi

_______________________________________________
mecm mailing list
mecm at diy-efi.org
ttp://lists.diy-efi.org/mailman/listinfo/gmecm

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.diy-efi.org/pipermail/gmecm/attachments/20111004/018b8bfa/attachment-0001.html 