Learn Me: Direct Injection maintenance

z31maniac said:

bobzilla said:

In reply to alfadriver :

I'm surprised this hasn't happened sooner to be honest. There are a lot of really good things about DI. Better fuel control, more power from smaller packags due to better fuel distribution and higher compression etc. But like everything there are downsides. More fuel dilution requiring better oils, pre-det issues, caking the valves etc. I imagine the complexity of having two fuel systems working in tandem has what has kept it an either/or as long as it has. 

Toyota has had the system for nearly a decade.

Well over.  They were doing it back in 2007 when we were developing our 2010 car.  But the reason they did it was because they could not meet emissions with DI only.  Back then, it was a very expensive solution, and we found that DI let you do some really cool tricks and actually make much better emissions.  That continues today.

But it was the cost and complexity that forced companies to find better solutions to that problem.

Now, there are other reasons to re-add PFI.  One of them is power- instead of adding another high pressure pump, it's cheaper to just add PFI injectors (since the drivers never left any of the ECU's).  

Once added, it lets you do some more tricks for even better emissions.

 There are a lot of really good things about DI. Better fuel control, more power from smaller packags due to better fuel distribution and higher compression etc.

Well, all the 10+ years of DI up to this point has just been the run-up to what Mazda is doing, where you run a gas engine like a diesel. That is 'the' main advantage of a DI system in my opinion, but it takes a lot of fancy controls to get there and this is how long it took. 

In reply to Vigo :

Can't really say that I agree.  HCCI isn't the revolution everyone thinks it is, and I really don't see many going down that path anymore.  Everyone had research on it, and like the rotary, Mazda is the only sticking with it.  Which, to me, doesn't mean it's a robust solution.  

And if you read how Mazda uses it, it's not really doing much.  Especially for the cost and effort.

DI still allows for significant increase in compression.  And lowers the exhaust temp a significant amount (which is really a big deal, in reality).  And lets one do some really trick things to keep emissions really low.  That's more than worth it.

(And DI is almost 20 years old in it's recent form- Mitsubishi had it in the late 90's)

Once you have DI and the level of precision required to have HCCI, you have everything in place to boost a gas engine to the moon without detonation. Stock gas cars have already been running nearly 20psi of boost for over a decade. How much further is useful when you have no real chance of detonation? 

Vigo said:

Once you have DI and the level of precision required to have HCCI, you have everything in place to boost a gas engine to the moon without detonation. Stock gas cars have already been running nearly 20psi of boost for over a decade. How much further is useful when you have no real chance of detonation? 

That sounds good on paper.  And while it might kind of work in the real world, the dynamics means that it's not nearly as helpful as it sounds.  And I mean that in terms of knock and in terms of driveability.  HCCI won't prevent detonation, btw, and neither will the most precise fuel injection.  

And I bristle as the comment that DI is so much more precise in fuel delivery than PFI.  It's not.  The hype has made people believe that it is, but the reality is that there has been some good steps to improve accuracy for small injection times, but the times are SO short that the open and close delay times are still a huge problem (let alone the dripping and leakage issues).

In reply to alfadriver :

TECHNICALLY DI is over 90 years old. Diesels have doing it all along. ;)

bobzilla said:

In reply to alfadriver :

TECHNICALLY DI is over 90 years old. Diesels have doing it all along. ;)

 

And gas DI is nearly as old, since the germans were playing with it in the 30's and it did make it into the iconic gullwing.

Actually, IMHO, the SPICA fuel injection that Alfa used for it's cars is capable of DI.  I wonder if anyone tried...

edit- and I think it's much longer than 90 years old....  Diesel first came up with the idea and prototyped it almost 130 years ago.

So this is unscientific to say the least, but here's my personal experience.  

My last two daily drivers have been GDI.  First was a 2012 Kia Soul, the base model with the 1.6 GDI engine (only way I could get a manual transmission).  About a year after I bought it, I began seeing articles and posts about problems with carbon buildup on the backs of the intake valves being a potential problem.  Once the car got to about 60K miles, I bought myself a cheap borescope and inspected the intake valves by snaking the borescope down the intake manifold runners.  All of them were pretty darn clean.  No real buildup at all.  I put a total of about 180K miles on that car, and inspected the intake valves about every 30K or so.  Never saw any appreciable buildup. 

Next up, and my current DD is a 2015 Honda Fit.  1.5 GDI (and another manual gearbox of course....) It has just over 60K on it now, and I recently inspected it's intake valves.  They are in great shape.  

So, the observation that some OEMs have problems with this and some do not seems plausible, even likely to me.

Another possibility is that there is something to be said for the popular theory that GDI cars that are driven in a "spirited" manner, rather than just puttering around, will have less problems with this.  I pretty much drive these cars like I stole them, so I suppose that could be part of the reason too.

Like I said, not scientific, but take it for what it's worth.  

HCCI won't prevent detonation, btw, and neither will the most precise fuel injection.  

The only thing you have to do to prevent detonation is never have more fuel in the chamber than you actually want to be burning. PFI fuel is 'all in' before combustion starts. DI fuel doesn't have to be (isnt in diesels), although emissions constraints might make it so in OEM applications.   That's what i was referring to. 

Vigo said:

HCCI won't prevent detonation, btw, and neither will the most precise fuel injection.  

The only thing you have to do to prevent detonation is never have more fuel in the chamber than actually you want to be burning. PFI fuel is 'all in' before combustip starts. DI fuel doesn't have to be (isnt in diesels), although emissions constraints might make it so in OEM applications.   That's what i was referring to. 

Except DI has to run stoich just like PFI does.  HCCI may be "like" diesel, but the system can't run that way 100% of the time- both for emissions reasons and it's virtually impossible to actually pull that off.

The only time DI isn't 100% injected on the intake stroke is the cold start, where tricks are used to get the emissions system lit off.

HCCI had a lot of "promise" but it never really panned out in a robust enough way to head down that path just like diesel.  And that's ignoring the emissions issues.  BTW, HCCI is gas auto ignition.  Which isn't knock, but incredibly close to it.

jimbbski said:

I have a DI engine in my F150 (3.5L Turbo) with 150K on it. No problems.

What I have read is you need to change the oil regularly and to use "good" oil when you do.

Doing this reduces the build up "gunk" on the backs of the intake valves because it's the oil that has broken down that causes the most buildup.

+1 oil must say “resource conserving” or something else that I can’t recall.

 

I'm on the F150 Forum and it seems to be common to change the oil at no more then 5K intervals.

Those that push it longer seem to have issues with their engines while those that change more often have less.

This is just my opinion based on reading various posts there but what I said in the second line is something

I've read in a auto trade publication.

Round two: the opposite bank of my V8
Yup, got stranded again because of a stuck injector.  Should have replaced them all the first time.  The silver lining is that I did it myself this time and saved $1,000.

My car is 10 years old but has only 65K miles.  The injectors and injector ports were filthy.

In reply to nderwater :

The tips are not that bad- I've seen a lot worse.  Odd that the injectors are sticking, though. 

bobzilla said:

I plan to do a half a can of seafoam in the Koup every 30k (getting close to number 2). Then again, with no EGR and a relatively decent pcv system, the Koreans arent AS bad as some of the other brands (*cough GM *cough).

Bobzilla,

Is that just a 1/2 can in the gas tank every 30K? 

In reply to Wxdude10 - Mike :

no through the PCV hose. Speaking of... its about time to do it again. 

alfadriver said:

In reply to nderwater :

The tips are not that bad- I've seen a lot worse.  Odd that the injectors are sticking, though. 

This was my first time pulling this style of injector so I had no idea what to expect.  The two with silver tips in the photo had been wiped a bit with a cloth.