Planet 9

Just have a couple quick thoughts about DFI.

I've wondered about the effects on intake valves such as carbon build up due to the lack of fuel to constantly keep them clean. Also the effects on the engine oil due to any present blow by that would force even more fuel into the oil, creating fuel contaminated oil. Think of this in terms of a diesel engine that can almost never have uncontaminated oil even after you change it.
Wouldn't DFI increase the compression ratio slightly as well?

Or would it be the exact opposite of the above by providing a cleaner complete burn.

Just some random thoughts....care to share?

It is the fuel sprayed behind the valves in a port-injected engine that causes the build-up on intake valves, which is not carbon. Carbon builds up inside the combustion chamber. DFI doesn't increase fuel content of the oil either; all DFI does is change the point at which fuel is injected.

DFI by itself does not increase the compression ratio, but most engines employing DFI get a compression boost during the redesign, because the fuel injected directly into the combustion chamber is cooler (because the hot intake valves haven't heated it), and this allows a higher compression ratio to be tolerated with a given level of anti-knock in the gasoline, increasing power.

DFI has two potential drawbacks. First, the super high pressure fuel pumps sometimes make a lot of noise. Second, the super high pressure fuel lines can be a maintenance complication, and a safety hazard if handled improperly.

I've heard that DFI engines often don't sound as good, but don't quote me on that.

__________________
'08 E90 M3 (replaced '07 Guards Red Cayman S, which will be fondly remembered)
'08 Miata GT
'08 Honda Ridgeline RTL
'07 KTM Super Duke 990
'07 KTM 300 XC-W

Very interesting question. I hadn't really thought about it. I think blueone covered your questions very well.

I wonder if maybe we'll see some effect of heat on the intake valves, valve guides and intake ports, but I sort of doubt it.

As with any new technology or product evolution, there are bound to be unanticipated drawbacks. Like blueone said, the high pressure fuel rails may prove to be maintenance items over time and I wonder how much fuel may be blasted out of them in the event of a serious crash.

__________________
Gator Bite
PCA Cayman Register Advocate

A complaint commonly levied against the 997.2, yes, but it has been pointed out in every review I have read that the 987.2 does not suffer the same fate. Is that down to engine placement and less use of sound deadening material? Who knows, but we'll find out in a few months when they start to arrive.

The biggest problem is probably what blueone pointed out above. I follow the BMW boards as well and they have had some significant issues with the 335i's high-pressure fuel pump failing and needing replacement. Seems like it's the sort of product that will fail early or not at all, but I guess that, too, we will see with time.

All else being equal (and it probably isn't), the answer would be that the higher fuel pressures would be a higher risk. This would be primarily due to the increased misting effect that is made possible by the higher pressures - bigger vapour cloud, higher peak ignition pressures due to the better atomization.

There would be ways of mitigating that increased risk but it would take a much closer look at the system design to understand what, if anything, has been done.

Just lending a safeguarding approach to this question that is utilized in chemical/refining process design. ie. minimize volumes and minimize pressures where reasonable.

__________________
GT Silver CS
1980 Honda CBX
2004 Suzuki SV1000
2006 Subaru Legacy
2002 Honda Insight

Actually it is gasoline's heat of vaporization (the heat energy absorbed converting the atomized fuel into a vapor) that creates the cooling effect allowing for a slightly higher compression ratio. Regards, Joe

Because it occurs in the combustion chamber, and not behind the intake valves. The effect you're describing takes place wherever you perform vaporization, but direct injection makes it occur in exactly the right place. And then the hot intake valves heat up the mixture. Actually, both of our original descriptions could use improvement.

Plus the higher pressures enable better vapourization which increases the effect mentioned. Also, there is more consistent vapourization, which tends to mitigate 'hot spots' which can cause uneven burn and therefore possible knock - therefore compression ratio can be raised even further without fear of knock. One of the reasons for more consistent vapourization is that liquid 'pooling' on the back side of intake valves is eliminated - there are high speed video's of this effect on non-DFI motors.

Note that better gasolines are also part of the equation - as time marches forward, any deposits can cause a further increase in CR or cause hot spots, so better fuels help mitigate that issue.

__________________
GT Silver CS
1980 Honda CBX
2004 Suzuki SV1000
2006 Subaru Legacy
2002 Honda Insight

Well said.

Absolutely correct, as is gtscayman (didn't want to quote such big a post).

Particularly on our cars, most of the sound is generated from the exhaust rather than the intake, as such, it's not really the advent of DFI that changes it's noise. This is extra true of our premium segment where "how it sounds" is of huge importance to Marketing. Magazine reviews up to date have stated the 997.2 is quieter with DFI, while the CS with DFI is louder/better. All this tells me is the marketeers decided "a base 911 should be more refined, if they're into their noise they can get a PSE" and "a Cayman customer will always be more performance-oriented, thus make them all sound huskier".

Now bear with me on this one. I'm aware almost all of you think of a 911 as a race-car, but remember. The Cabrios, the non-Ses, there's an awful lot of 991s that go to customers who really aren't racers at all, and having a refined base product makes more market sense than in the case of a Cayman, as the same customer ("hairdresser" type as it were, no insult intended) would simply buy a Boxster.

Regarding reliability issues on DFI, I'd be massively impressed if DFI engines were less reliable. Blow-by is something that can be very accurately measured, compensated for and corrected. It's also something which can (conceivably) impact engine-life before it's warranty time is due. And for a brand whose reputation is built around 30-year old cars still capable to be raced, that's unlikely.

The hard science of a DFI's combustion has been very nicely explained above, but also take into consideration that a DFI has a lot fewer moving parts and a smaller total parts count. This is undoubtably an indicator that at least mechanical reliability is improved tangibly.

I haven't looked into the construction of the type of pump used, but could this possibly be an item to watch for concerning reliability? Or if not the reliability, I wonder how much the pump costs? Diesel pumps aren't cheap for instance and they have the 'luxury' of pumping something with some lubricity. Any thoughts or info on that aspect?

__________________
GT Silver CS
1980 Honda CBX
2004 Suzuki SV1000
2006 Subaru Legacy
2002 Honda Insight

Excellent point. DFI is not too unlike what modern diesels do already. I'd like to remind people of the history of diesels in cars, in particular American cars. Remember those good old days with GM and the V8's that were actually converted gas engines? Remember how many fuel pumps failed? Remember the cost?

While I'm absolutely positive modern engineering and materials will improve this area, the extremely high fuel pressures, combined with the significantly lower flash points of gasoline and the overall difference in the heavy duty nature of a real diesel engine vs a gasoline engine for weight savings would make me take pause and a wait and see attitude. While I'm less afraid of crashes, time and miles would be a concern for safety and reliability.

--CC

I was thinking about this assertion... why is it that DFI systems have "a lot fewer moving parts and a smaller total parts count"? There's no mechanical throttle (I forgot about that in the efficiency area), but there's the same number of injectors. I suppose on some BMW engines with a butterfly per cylinder there's a big mechanical savings, but that wouldn't be true for Porsche. True? The engine management firmware looks about an order of magnitude more complicated, the high pressure fuel system and injectors look like mechanical risks, and the fuel rail plumbing is somewhat more difficult. Am I missing something?

PS - don't confuse my discussion with negativity. I'd rather have a DFI engine.

re: reliability of the new DFI hardware, I would guess that Porsche doesn't make any of this. It is likely a Bosch product and Bosch has been making DFI units by the millions for years.

High pressure DFI or just DFI? HDFI would use smaller injection orifices, run higher combustion temperature, require more computing power and control parts, etc from what I can see.

I was surprised to see DFI has been around for years in gas engines (aviation first, then Mercedes in the 50's) but these appear to be entirely different. Then HDFI appears to have shown up seriously about 10 or so years ago, but OMC basically went bankrupt with their design (2 stroke though), and even the success of the Audi R8 in LeMans can't vouch for its reliability in consumer cars (unless you carefully prepare the fuel for your car and rebuild the engine every 24 hours!).

That being said, a Google search didn't come up with any recent HDFI issues and there have been more and more of them on the road in the last several years. I would hate to be a young guy though, buying one of these used in 10 years - trouble shooting has to be a pita!

__________________
GT Silver CS
1980 Honda CBX
2004 Suzuki SV1000
2006 Subaru Legacy
2002 Honda Insight

You know, I had the exact same question. I read the fewer moving parts malarkey in one of Porsche's press-releases. I personally suspect that most of it has something to do with the cooling system rather than the engine aspiration, but they did state they had fewer moving parts.

Now here's the part I'm not entirely clear on. Direct Fuel Injection only means the fuel is sprayed into the cylinder rather than into the air at the (plenum?) pre-head stage. It does not necessarily mean there's no throttle/buttefly.

Initially I also assumed it'd be like a Diesel Cycle, always ingesting a standard quantity of air and injecting (varying amounts of) fuel into it, resulting in nonsensical (and wildly fluctuating) values of Lambda. Having discussed this a little further with colleagues (diesel researchers) we decided that it's rather unlikely.

There are absolutely definite gains to be had from Direct Injection on throttled engines, mostly related to improved combustion efficiency (notice increase Compression Ratio), but while I've been led to believe (by the hype, the implications of press-releases, and the forums) that DFI engines really run like Diesels fuelled by petrol, I find it hard to believe. Recently I was in a Shell Engine Research presentation where a fuel scientist explained why Petrol was the Diesel of the future (practically explaining why a Diesel-cycle Diesel engine running Gasoline in Compression-Ignition was possible). It was extremely clear that this kind of technology was 15-20 years away, and it looked suspiciously like what we're led to believe DFI production engines are.

I'm not inclined to believe that Shell engine research is finding what Audi/Jag/BMW/others already sell exciting pioneering cutting-edge stuff that's 20 years away.

There really is only one way to definitively prove that DFI is not in fact petrol-fueled Diesel-cycle, (rather, it's a Direct Injection Otto-cycle throttled petrol engine). And that's to check if there's a Catalyst on your DFI cars. If there is, it'll absolutely only work at the correct stoichiometric ratio of Lambda=1 (or small variation) and hence prove the engine is in fact running Otto.

And hence, long story short: DFIs do have a throttle. I think.

An insightful post, thanks.

Let's ask Sharky to check his 997.2 and see if it has a throttle! Sending PM now...

Some do, some don't. BMW's "Valvetronic" engines don't have butterflies, for example, and vary airflow by adjusting intake valve lift. They're not running like diesels. I, perhaps wrongly, assumed that since Porsches had variable valve lift they were doing the same thing. The Cadillac DFI V6 still has a throttle, for example, and I think the Cayanne V8 does too, but for some reason I thought the flat-sixes didn't. Old age plays tricks with one's memory, sometimes.

Here's an excellent tutorial on DFI I found a while back:

1115 Ppt Presentation

In the presentation you'll notice the author calls DFI more complex than port fuel injection.

Just to add a little more confusion to this thread, I believe the statement that there are fewer parts in the new DFI engine referrs to the design of the bottom end (ie, no separate cradle holding the crank) rather than to theisystem. The elimination of the crank cradel should forever eliminate the RMS leaks.