By on May 3, 2012

“Ask an Engineer” is hosted by Andrew Bell, a mechanical engineer and car enthusiast. Andrew has his MASc in Mechanical Engineering from the University of Toronto, and has worked on Formula SAE teams, as well as alternative fuel technologies in Denmark and Canada. Andrew’s column will explore engineering topics in the most accessible manner possible.

Even though every other car nowadays seems to offer gasoline direct injection (GDI), Mercedes-Benz was the first to exploit this technology in the 1955 300SL. But it wasn’t until the mid-1990’s that other automakers started to use GDI in mass produced vehicles. GDI promises marginal increases in fuel economy (3% reduction in BSFC ) but its real benefits include reduced cold start/low load emissions and higher power outputs. While the technology offers engineers incredible flexibility from an engine design perspective, it is not without faults. As with any new technology it is important to understand both the positives and negatives before you choose, say a compact car with GDI or one regular fuel injection. If you want to keep your car for a long period of time, the long-term reliability of a GDI engine is an important factor.

The effect of increased percentages of ethanol on injector longevity.

The percentage of ethanol in gasoline at the pumps is steadily increasing. Ethanol has a tendency to increase the corrosion rate of the various metals used in an engine. Add this to the elevated fuel pressure and the fact the injector is directly exposed to in-cylinder combustion events, and you have a recipe for a recall. Furthermore, these injectors are very sensitive to fuel quality due to outrageously tight tolerances. It is very important to use high quality fuels and keep the filters clean.

Higher pressures in general.

GDI requires significantly higher fuel inlet pressures than port injection. This puts a great deal of strain on every piece of the fuel delivery chain. This is not a problem on a new engine. 50,000 miles down the road, and it may be. Manufacturers have been relatively proactive in this department by specifying robust, stainless steel fuel lines and connections. That hasn’t stopped fuel pump recalls from already occurring

Carbon buildup on intake valves.

This is the big problem with most current GDI engines. Due to modern unburned hydrocarbon (UHC) regulations, vapors from the crankcase are usually vented into the intake stream in order to prevent oil droplets from escaping through the exhaust. In a port injection engine, these droplets are ‘washed off’ the neck of the intake valve by a relatively constant stream of gasoline droplets. In a GDI engine, the gasoline doesn’t touch intake side of the valve. As a result, the droplets have a tendency to bake onto the valve and significantly reduce performance. To add to this effect, many advanced GDI engines also include exhaust gas recirculation in order to lean out the combustion mixture and reduce in-cylinder temperatures for certain combustion modes (reducing NOx emissions). Since GDI combustion has the ability to produce far more soot than premixed combustion (port injection), the problem is magnified.

Even more alarming is that these deposits can dislodge and damage other downstream components (turbochargers, catalytic converters, etc.). Manufacturers have added systems to capture these oil droplets and particulates, but no system is 100% effective. As a result, there are many disappointed early adopters with large repair bills. Even diesel engines haven’t been immune to these issues.

The reason these issues have slipped through to production is that they won’t show up in a 500,000 mile torture test. These types of issues will appear after years of short trips (preventing the engine from reaching operating temperature), bad batches of fuel, etc. As we approach the efficiency limits of the internal combustion engine, the engines themselves (and associated support systems) have become more complex. As with the transition from carburetors to electronic fuel injection, there will be some overlap between relatively bombproof port injected engines and the unproven, first-generation GDI engines.

 

Get the latest TTAC e-Newsletter!

130 Comments on “Ask An Engineer: GDI Problems In A Nutshell...”


  • avatar
    cacon

    Good overview of the issues inherent with these new GDI engines. Thanks for the reading.

    Can you recommend the most effective method to clean up the carbon build ups in the intake valves?

    I’m interested since I own a SEAT (VW) with a TSI 1.8 liter engine (the BZB one). And most of its use is in stop and go traffic :(

    • 0 avatar
      ciddyguy

      For most port injection motors,just get it out on the highway on occasion and occasionally floor it, but drive it at a fast, steady speed for at least an hour to blow out any buildup. The idea is to get the motor good and hot and keep it moving for a while before you shut it down.

      I believe it’s called an Italian tuneup.

      While not perfect, it has been reported to help keep carbon buildup from building up too much.

      The trick is to do it like at least once a month or so if you don’t do much if any highway driving.

      • 0 avatar
        turbobrick

        Or you can just drive at a reasonable speed and drop it down a gear so the revs go up. I usually do this on my way home from work, let everything come up to operating temp and then just switch the OD off.

        • 0 avatar
          mephy

          Hello,

          for fix the carbon oxide buildup I find a solution.

          I use AMSOIL INTERCEPTOR in the gas at ratio of 1:500.

          The concept is very simple, adding this 2 cycle oil in the gas the gas becomes better especially the one with ethanol.

          Better gas better combustion.

          It gives even a lubrication to all fuel system from the gas pump till the injectors
          and has anti rust property.

          It is engine safe and catalyst safe, now is 4 years that I use it.

          I tried different kind of 2 cycle oil and different brand at different ratio, I end up with this one at this ratio, it works great for me.

  • avatar
    aristurtle

    Motorcycle engines (or, at least, sportbike engines) tend to focus on peak power per displacement over other considerations like engine longevity and such, yet GDI has not, to my knowledge, been used in a modern bike engine. Why do you think that is, and when do you think it will start to be introduced?

    • 0 avatar
      protomech

      Motus planned to use a GDI setup for their $30k touring bike, but ended up dropping it claiming lack of industry support.

      The motorcycle business is pretty conservative.

      http://www.asphaltandrubber.com/news/motus-mst-drops-gdi-favor-port-fuel-injection/

      • 0 avatar
        aristurtle

        Well, I can understand that; Motus is a very small shop and can’t really afford the R&D risks involved with a new engine design, but I’m surprised that, for instance, BMW hasn’t tried it yet.

      • 0 avatar
        ellomdian

        MOTUS is trying to start a company selling a $30k sport-touring bike based on a (admittedly) new-design motor in a market segment that has only really existed for the last 3-5 years (and is currently a fierce competition among longstanding manufacturers.)

        I would not be surprised to hear them claim that they had considered anti-gravity suspension but that the establishment was too conservative to consider it.

  • avatar
    TonyJZX

    the toyota subaru FT86 has a 2nd set of injectors in the old port position to clean off deposits

    its also been my impression that the VW VAG cars are about the worst for direct injection deposits

    there can be a lot almost like stalagmites developing behind the valve

    • 0 avatar

      Toyota’s DI engines are all this way so far. But it’s an inherently compromised design: twice as many injectors, and the compression ratio probably can’t be as high.

      • 0 avatar
        CJinSD

        The Lexus IS350 has an 11.8:1 compression ratio and the FT86 has a 12.5:1 compression ratio. While lower than Mazda’s Skyactiv, it would take a true newbie to be surprised when the Mazdas have engine issues early and often. It is pretty obvious to me that all current choices are compromises, but Toyota’s compromise does the best job of delivering the benefits of DI without the built in crippling durability flaws. Are Toyota injectors even half as failure prone as those used by other GDI pushers? That goes a long way towards making up for having twice as many even before your get to leave the head on the engine past 100,000 miles.

        • 0 avatar
          darrelld

          My Lexus IS350 had carbon build-up to the point of needing service at 38,000 miles. Symptoms were rough idle and engine stumble from off idle acceleration. The dealer charged $150.00 for something the service ticket referred to as induction cleaning.

      • 0 avatar
        Athos Nobile

        Toyota’s solution will be shortly implemented by other OEM, we were shown in class some schematics of an Audi engine with both systems..The second set of injectors help solve some emissions or drivability (I can’t remember ATM) problems at low RPMs.

        If I understood correctly, the PFI would work until certain engine speeds and from there the DI would pick up.

        Everything you touch in this world is a compromised design Mr. Karesh. I wouldn’t expect reduced compression ratios because of this.

      • 0 avatar
        racingmaniac

        I wonder though if can’t alter their timing as such(a la Skyactive) that they basically sacrifice C/R at the partial load Port Injection mode.

        I am also not sure VAG cars are necessarily the worst. Maybe by volume as they are probably the most prolific.

      • 0 avatar
        ajla

        Does the 4GR-FSE use both? I thought that one was DI only.

      • 0 avatar
        sportyaccordy

        The doubling of injectors/fuel systems is legit, but it also allows for the best of both worlds- including the benefits of DI that enable super high compression.

        Honestly though I have seen a lot of 12+ to 1 pump gas built motors in little wind up Hondas. I think the real issue with conventional injection and high compression is emissions.

      • 0 avatar

        Sorry, I was thinking only of Toyota’s US-market DI engines. I should have qualified the statement.

        I imagine that there might be no need to reduce the compression ratio if the port injectors are used under low load conditions when there’s little risk of premature detonation. Many variables in play, I’d love to read a thorough account of the different trade-offs made for various engines.

        Update: I hadn’t realized that the 2.5-liter V6 in the IS 250 lacked port injectors. So Toyota does have a purely direct-injected engine in the U.S.

      • 0 avatar
        wumpus

        The compression ratio can be just as high as long as they can turn off the secondary injectors at WOT (or close enough to otherwise cause detonation). If the issues are seen in grocery getting but not in torture tests, the engine can handle pure DI under such conditions, but not always under lower loads.

        It is more an issue about getting the company to pay for the parts needed for reliability. Expect to see it on Toyotas and Hyundais (since its their dime).

  • avatar
    supersleuth

    This is why, if I were in the market now, GDI would be a dealbreaker. Of course, by the time I actually am in the market, if I want to buy new I may not have much choice…

    • 0 avatar
      TonyJZX

      i think you will be safe for a long while

      ford and GM do not have many (or any?) passenger cars with direct injection

      Ford have a DI Coyote in the wings but that isn’t an economy car so…

      The Cruze delivers almost 40mpg with plain port fuel

      the new Dodge Dart won’t be DI (as far as I’m aware)

      i would say PFI will live on in economy cars and domestics for quite a while

      • 0 avatar
        Ubermensch

        Ummm, what are you talking about? Ford and GM both have tons of models with GDI engines right now including volume sellers like the Focus and new Escape.

      • 0 avatar
        ciddyguy

        I know Fiat has talked about eventually doing GDI in the 1.4L MultiAir, but it’s in the wings for now as far as I know.

        I can’t remember if the 900cc TwinAir has GDI or not and don’t have time to look it up right now.

      • 0 avatar
        DannyZRC

        the ecoboost and all of the turbo ecotecs are DI, are they not?

      • 0 avatar
        supersleuth

        No, they are not. For example, the 1.4 in the Cruze and Sonic is not.

      • 0 avatar
        oldworntruck

        actually ALL ecoboost engines are GDI and gm has been employing gdi since 2003 in certain models including the solstice etc..

      • 0 avatar
        supersleuth

        But as I said, GM still has plenty of turbo Ecotecs that are not GDI. I’m pretty sure they have GDI versions ready to go should they so choose, but they’re getting excellent fuel economy from the 1.4 without it so I hope that stays port injected. (The Cruze Eco with stick would be on my short list if I needed to replace my car right now.)

      • 0 avatar
        Mud

        @Tony
        Not correct. Our 2011 Equinox has the DI 3.0 V6.

      • 0 avatar
        TireIrony

        GM has been putting direct-injected V6s in their cars and SUVs since 2008 — first the Cadillac CTS, expanding to basically every car and SUV with a V6 at this point.

      • 0 avatar
        Aqua225

        Supersleuth,

        the GM 1.4L Turbo in the cruze is GDI, though GM does a very good job of obscuring it on their website.

        http://www.motivemag.com/pub/news/GM_Announces_1_4-Liter_Turbo_for_Cruze_1_4.shtml

        And Andrew Bell here is clueless.

        There has been much discussion on technical and engineering sites about the GDI carbon problem.

        Currently, the GM V6 doesn’t appear to have it, the jury is out on the ecoboost motors, and Audi has a serious problem in their GDI V8s.

        Some engineers with some of these car companies say it boils down to cam phasing based on airflow, temperature, and throttle demand. Once you have the magic, you will stop having buildup.

        EGR contents, PCV burps, etc. will not phase the typical GDI engine. Think about it this way: for ages, the same burps and gaseous emissions have gone through the *really cold in comparison* intake manifolds, and generally speaking, you don’t have buildups there.

        Andrew is just participating is some good ol’fashioned FUD.

      • 0 avatar
        supersleuth

        No, the Cruze / Sonic 1.4 isn’t GDI. Your source is wrong.

      • 0 avatar
        Educator(of teachers)Dan

        3.6VVT V6 DI engine in the last year W-body Impala? Anyone? Anyone? Bueller? Bueller?

      • 0 avatar
        Andrew Bell

        @Aqua225
        The purpose of this forum is to share issues that do not get discussed when a potential buyer walks into a showroom. Of course these issues have been discussed at length on engineering forums. As I pointed out, the technology isn’t exactly new. You are on the right track with your comments regarding cam phasing, etc.

        What you are missing is that the back of the intake valve on a GDI engine can get much hotter than an equivalent port injected engine mostly because there is not a steady stream of fuel washing it off every cycle. Whatever gets deposited on the back of the valve gets baked on as a result.

        Thank you for your comment.

  • avatar
    dave504

    Great article. The last two paragraphs really underscore the fact that most of the major automakers are still having issues with implementation of DI technology. VW/Audi FSI engines have major carbon/sludging issues, BMW has gone through six revisions of the HPFP for their N54 engines, (and owners are now reporting fuel pump issues with the new N55 engines as well), and Hyundai recommends a special additive for their GDI engines. Toyota seems to have the best reliability using a combination of both port and direct injection (interesting technical info: http://www.ft86club.com/forums/showthread.php?t=3172), though I have not heard of many issues with Ford’s Ecoboost either. I will soon be in the market for a new vehicle, but finding one without DI and a computer controlled dual clutch transmission (another technology that’s not quite there yet) is getting difficult.

    • 0 avatar
      TonyJZX

      i am not one of of those people who gets rid of a car after the 3 or 5yrs warranty is up so maintainability after warranty is a concern

      i will not even look at stuff like the VW Golf with DI turbo and DSG… or the Focus with the powershift gizmo

      where I am they are soon to have a 1,6 ecoboost that is supposed to be good but I think I would err on the safe side and get a Cruze with plain PFI, turbo and 6 spd manual… or the 6 spd auto… but GM autos are generally quite terrible

      • 0 avatar
        Luke42

        @TonyJZX: In your case, there are a lot more reasons to avoid Volkswagen than just GDI.

        I’m a burned and bitter former Volkswagen owner. I owned an out-of-warranty Volkswagen Jetta for 20k miles once, and the old beater Ranger with 160k miles on the clock (at the time) that the Jetta was supposed to be replacing turned out to be far more reliable and cost-effective in the long run.

        I spent something like $7k maintaining that Jetta over those 20k miles, and I spend about the same amount to maintain my out-of-warranty Ranger over 100k miles and 8 years. If you care about out-of-warranty maintainability, Volkswagen is not for you!

    • 0 avatar
      bd2

      German and Japanese GDI engines (not counting the newer Toyota GDI setups) have had a history of carbon buildup; it’s actually GM’s GDI powerplants that have had the best success with limiting carbon buildup.

    • 0 avatar
      ppxhbqt

      Hyundai does NOT recommend a special additive for their GDI engines. Hyundai recommends “Top Tier” gasolines, like many companies do. This applies even for cars like the Elantra, Elantra Touring, and Santa Fe, none of which offer any GDI engine. The manual says that if Top Tier gas isn’t used and drivability problems arise that an additive is recommended with every oil change. The wording is the same for cars with and without GDI. Plus I’ve yet to find anything on the webs about problems with Hyundai’s, GM’s, or Ford’s systems. And all have been on the road long enough to have both high mileage and granny’s grocery getter short trip examples.

  • avatar

    The Nissan set of engines haven’t had many issues I know of. I have a 2005 Pathfinder (first year of the VQ 4.0) and despite my best efforts, my wife continues to put cheap gas in it. Despite that we are at 130k on the clock without any issues outside a fan clutch and thermostat.

    An uncle just unloaded a Jetta TDI after spending $2200 at the VW dealer (Had 36,300 miles on it with a 36,000 mile warranty). They replaced two injectors and still had to replace the EGR system to make it run right. I would be weary of Diesels with direct injection until they can iron out these issues. Especially German ones.

    • 0 avatar
      Fred55

      You will be weary of probably every diesel ever made then. Diesel engines are inherently direct injected. http://en.wikipedia.org/wiki/Diesel_engine

    • 0 avatar
      racingmaniac

      I don’t believe the VQs are GDI. Nissan have build GDI variant of VQ but never for US AFAIK. There were VQ25DD and VQ30DD for Japanese domestic(maybe Asian) market. The ones we’ve been getting were still conventional port injected variant…

    • 0 avatar
      brettc

      VW has been selling TDIs since 1989, so most of the problems are known with the older generations. That being said, no one knows how the current common rail TDIs will be for long-term costs with the addition of DPFs and the other associated emission control equipment. I’m sorry your uncle got bent over at the dealer. I’d love to have details of what that dealer actually did for $2200 because it sounds like he really got hosed. Overall, I’d trust a TDI over a GDI engine since the TDI is a little more proven.

  • avatar
    Vipul Singh

    Good initiative.

    Out of interest, are there any non-Fuel Stratification (FSI) implementations of GDI, or do all GDIs have FSI/TFSI?

    Can there be any advantages of going for a GDI approach without FSI?

  • avatar
    NN

    Are there any reliability issues showing up with the new generation Sonata’s since they are now a couple years out? There should be plenty on the road with 50000+ miles.

    • 0 avatar
      supersleuth

      The manufacturer’s recommendation for regular use of a special fuel additive is already a question mark on these. And it’s the 200K mile mark I’m really interested in.

      • 0 avatar
        srogers

        Do any manufacturers design for durability beyond 200k miles? I’d be satisfied if any vehicle of mine made it to that point without anything beyond routine maintenance.

      • 0 avatar
        supersleuth

        Lots of cars- nearly all Toyondas and plenty of domestics- will do that quite routinely. I even had a Saturn- talk about a POS- that was almost there and would have made it easily if my daughter hadn’t wrecked it. Failure of the new crop of GDI engines to do this with ease would be a considerable step backward. So some of us will wait and see.

      • 0 avatar
        MadHungarian

        200K? No sweat in a Town Car. Low acquisition cost, really low maintenance cost, decent mileage for the size. If you want reliability, always pick the drivetrain that’s been around long enough to work all the bugs out, and avoid the Mr. Wizard gee-whiz specials.

      • 0 avatar
        Aqua225

        I’d love to take a fox body car and cram a 2.3L Turbo with aftermarket fixin’ in it, and see what the mpg & performance would be like, 0-60mph in 5, 100 in 12sec with a 13s quarter mile would be my target, with 35mpg with low or no boost settings.

        Maybe with the right gear optimizations? All I know is that it is sad to see the fox go, since it was the last of the machines, other than trucks, where you could do some really weird powertrain transplants with.

        Most front drive power trains are over customized for their chassis (rear, too, nowadays), for it to be plausible to do weird engine grafts on.

  • avatar
    ComfortablyNumb

    This is fear-mongering. You make it sound like these failure are common to all DI engines. In fact, the majority of DI engines are very reliable. We’ve been doing ethanol and DI diesels for years, and these effects are well-known. Any new engine, DI or not, will have growing pains. Being an engineer with a little more knowledge than the average observer does not mean you can make unsubstantiated generalizations about this technology.

    • 0 avatar
      DC Bruce

      I don’t know who “we” is who have been “doing ethanol and DI diesels for years,” but the OP seems to make a fair point. Volkswagen’s reputation for reliability is bad on so many fronts (see Consumer Reports owner surveys and Trued Delta), it’s hard to isolate whether their introduction of turbocharged GDI and CRD engines has added to the woes of their owners or not. Ford’s 2 liter DI engine in the Focus is new this year . . . too soon to assess longevity. Toyota has had a quasi DI system in the 3.5 liter V-6 that first appeared in the Lexus, but as the article points out, that system uses port injection as well.

      Diesels have always been “direct injection,” but prior versions of automotive diesels used a pre-chamber and relatively low pressure and simpler fuel injection systems. The higher pressure, common-rail injection systems on modern automotive diesels (which eliminate the pre-chamber and associated heat loss, thereby boosting efficiency) are also relatively new and untested. The legendary Mercedes automotive diesels of the 1980s and earlier, which would run for many hundreds of thousands of miles with proper oil and fuel filter changes are quite different from the CRD systems Mercedes and other manufacturers use today.

      So, it is fair to say that the GDI and CRD engines that have been introduced in the past few years are “unproven” over long use.

      Interesting that Mazda’s “skyactiv” 2.0 liter port-fuel injected engine makes all of 2 h.p. less than Ford’s GDI engine of the same displacement. Neither uses forced induction.

    • 0 avatar
      jz78817

      it’s not fear-mongering. it’s an issue with engines from at least a few automakers, and it’s something to be concerned about across the board until we determine how the industry as a whole is doing it. I’ve heard much more about deposits in VW/Audi engines, but that might be because people are looking at them more. The only things I’ve found about e.g. Ecoboost deposits are firms selling valve cleaning services.

  • avatar
    Sundowner

    I have or have had a total of five direct injecton motors in my driveway, all are or were made by VW. The 2.0T in my wife’s 2006 A3 needed the valves cleaned at 80k miles after a fair but constant 5 years of service. The valve cleaning cost $600, and the car has been reliable since. the other three cars are driven by what I would call more ‘spirited’ drivers and have about 100k miles between them with no injector issues. The last one, the 2.0 TDI diesel motor had multiple problems after only 8k miles. I’d say that my experiences support the article, it’s not fear mongering, but it’s not a deal breaker, either. Just another hassle on a newer car, which is inevitable. I’d buy another VW 2.0T car, but I would never buy a TDI again.

    • 0 avatar
      ismail

      It looks like the new generations of the vw/audi engines will use the same trick employed by toyota:

      http://forums.vwvortex.com/showthread.php?5468544-NEW-EA888-1.8-TSI-2.0-TSI-Engine-Details

    • 0 avatar
      chuckrs

      On a Porsche Cayman forum, this has been discussed quite a bit. The consensus is to drive it like you stole it. This is pretty antithetical to our driving habits and needs which seem more attuned to lazy V8s with a lot of low end torque.

  • avatar
    NormSV650

    Lots of DI Cadillac CTS’s and now minivan mamas get the 3.6 DI in the Chevy Traverse.

    Ttac is a couple of years behind on this one. They should stick to non-techical articles.

  • avatar
    racingmaniac

    Nice to see a fellow Skule Alumni writing on TTAC. What year were you on the FSAE team? I’ve been around from 03-07.

    With regards to the GDI, the foreword mentioned GDI promises reduction in low-load emission, is this specifically CO2 measure? Toyota’s D2S(or something like that) and the next gen VAG TSI motor both incorporates a 2nd set of injectors for port injection at low load application. I think VW’s reasoning was that in low load/sub-optimum temp operation the particulate emission is worse and thus using port injection at those time is actually better for emission purpose. Has carbon build up issue been as much a problem in diesel vehicle? Or perhaps their higher operating temperature curb this somewhat?

    • 0 avatar
      Andrew Bell

      @racingmaniac
      I wasn’t much of a contributor to the Toronto team but I did a lot of work out West when I was at UVIC. It was a great opportunity to get some hands-on engineering experience!

      The key point with respect to my comment on emissions is that direct injection allows for incredible flexibility. You have the ability to precisely adjust the timing and duration of an injector event and can even produce multiple injections in a single cycle. This is a huge advantage for the engineers responsible for meeting increasingly challenging emissions standards.

      As with gasoline DI engines, some diesel engines are worse than others. To avoid starting WW3, I will leave it at that. Thank you for your interest.

  • avatar
    DannyZRC

    Mr Bell, is the 3% BSFC gain in N/A applications specifically?

    I ask because I know that DI has a charge cooling effect, and DI turbo engines tend to be higher compression than their conventionally injected counterparts, with compression ratio being one of the larger factors impacting fuel efficiency (specific consumption/output).

  • avatar
    nickoo

    So I take it the valves build up carbon on the backside from the combustion chamber? It sounds like the lowest cost and easiest way to handle this, (although not an elegant solution) is to just have the valves periodically cleaned. Perhaps the motors can be engineered to facilitate an easy valve cleaning method or there could be a combustible cleaning solution/gasoline released from an atomizing device that is sucked through the air intake and over the valves?

    • 0 avatar
      Truckducken

      I’d like to know if filling the intake with a can of Seafoam and letting it steep for an hour will do the trick, like it seems to for non-DI engines? If so, my normal maintenance routine isn’t going to change much.

      • 0 avatar
        brokeguy

        That’s exactly what nickoo described, although instead of filling the intake you just let the engine vacuum pull a small amount of Seafoam into the intake through a vacuum line that feeds into the intake manifold. Some of the old-timers swear by hot water, they say it’s an even better solvent than Seafoam. Of course, if you do this incorrectly there is a possibility you could damage the engine.

    • 0 avatar
      TR4

      “a combustible cleaning solution/gasoline released from an atomizing device that is sucked through the air intake and over the valves?”

      My car has two of those. They are called carburetors.

  • avatar
    missmySE-R

    Would love to have TrueDelta’s Michael Karesh do a bit of data analysis on this topic. Within the same model, how does engine reliability compare in DI vs. standard FI varients? Cobalt SS (’08+) vs. std Cobalt. Solstice Redline vs. std Solstice. Taurus SHO vs. Taurus AWD. MazdaSpeed3 vs. Mazda3. Jetta GLI vs. std Jetta. Cooper S vs. Cooper.

    • 0 avatar

      I’d need far larger sample sizes. Or for these cars to get older.

      For cars new enough to have DI we tend to have very few reports of engine problems. Plenty for fuel pumps in BMWs a year or two ago, but not for gunked-up valves. Maybe there will be more reports in a year or two as the miles accumulate.

    • 0 avatar
      Andrew Bell

      I am also looking forward to more hard reliability data as these cars age. TrueDelta is a fantastic resource.

  • avatar
    foojoo

    My father recently purchased a 2012 GDI car, and the manual states that one should only use Top Tier fuel. If you Google it you’ll see a list of companies that are supposed to be using a sufficient amount of additives to minimize problems with carbon buildup.

    Anyhow, is Top Tier fuel a marketing gimmick or necessary to minimize carbon buildup?

    Either way, he has been using fuel by Top Tier companies.

    • 0 avatar
      Ubermensch

      I don’t think the quality of fuel is going to help with the carbon buildup issue from what I understand about what causes it. Most of the carbon is coming from oil vapors from the crank case and from unburnt fuel routed back to the intake by the EGR so much of the carbon is coming from the engine oil and not the fuel. Without the fuel, top tier or otherwise, to “wash” the valves, the carbon builds up anyway.

      Personally, I think this is being blown out of proportion as it seems that some relatively inexpensive preventative maintenance can eliminate the issue. It is looking like intake cleanings every 50k miles, or maybe even less often, can remove a lot of these deposits. I think this is a small price to pay for the added benefits of DI. It is the fuel pump issues that companies like BMW are having that is more worrisome and inexcusable at this point.

      • 0 avatar
        supersleuth

        I beg to differ. For me this is clearly more of a pain than the marginal benefits of DI are worth- and they clearly are marginal, less for example than the power / economy benefits of a small turbo as with the 1.4 Ecotec.

      • 0 avatar
        redmondjp

        Same here. I see a negligible increase in performance/economy along with increased complexity, higher maintenance costs and lower long-term reliability.

        Let’s say that the seafoam-into-a-vacuum-line method doesn’t work for the car you buy and they actually have to remove the intake manifold to clean the valves every 50K miles at a cost of $500-800. Will you ever gain this added cost back by the small increase in economy over a port-injected engine? I doubt it.

        My 1988 Buick 3800 was at 221K miles when I sold it, still running fine with the original fuel pump and injectors, and the injectors were never removed and cleaned. Am I going to get that kind of low-cost, worry-free reliability out of a DI motor? I did exclusively use Shell gas in the Buick over its lifetime, so I’m a believer in the fuel additives/detergents.

      • 0 avatar
        Aqua225

        I agree with supersleuth and redmonjp, a down-to-the-valves cleanup every 30 to 50K is unacceptable. If GDI requires that, as the only possible way to remedy the problem, it is as good as dead as a technology.

        But luckily, that does not appear to be the case. There are several GDI motors not having the issue the Germans appear to be currently having en masse.

  • avatar
    bryanska

    Will laser ignition help this problem?

  • avatar
    ComfortablyNumb

    They absolutely are fair points, but there is a difference between pointing out issues, and saying DI engines “have a tendency” to do something. It ignores the fact that there are millions of miles of trouble-free operation from DI engines running diesel, ethanol and gasoline. This article makes it seem like you’re walking on eggshells if you buy a DI engine, and that’s just not true.

    My point is that while DI may exhibit the symptoms described above, the symptoms are not a guarantee for every DI engine. Discussing them is fine, but suggesting that it’s a gamble to buy a DI engine is disingenuous.

    • 0 avatar
      Andrew Bell

      @ComfortablyNumb
      It was not my intention to attempt to destroy the credibility of GDI technology. The purpose of writing this article was to bring to light some of the common issues encountered with GDI engines. Some engines have proven quite bulletproof (the GM V6 comes to mind). A fair number however have proven quite problematic for various reasons (no names will be mentioned). I am merely suggesting due diligence prior to making what is a substantial purchase for most people.

      GDI will be everywhere soon enough as it provides many benefits. Similar to the transition from carbs to EFI, there have been, and will continue to be growing pains here as well.

      Thank you for your comments.

  • avatar
    stryker1

    Very interesting.

  • avatar
    Mervich

    This is NOT a new problem only associated with DFI. Carbon build-up on the valves became a problem 40 years ago, mostly associated with port fuel injection. This is simply the latest chapter. The number one cause is crappy fuel. Those who insist on using unbranded, convenience store gas because it is two cents cheaper will certainly experience carbon build-up on the valves and therefore reduced performance and diminished fuel economy from ANY fuel injected engine. Chevron gasoline has long been the fuel of choice with the best additive package (Techron) to aid in prevention, if used exclusively and in the grade (octane) recommended by the manufacturer. There may now be a better way to remove existing carbon build-up, but the old tried and true method of elimination was blasting crushed walnut shells through the engine. BMW used the walnut shell method for years…and still may.

    • 0 avatar
      George B

      I’m a cheapskate who buys the least expensive gasoline available. Usually QuikTrip and RaceTrac. (Similar to 7-11 and Speedway) QuikTrip is on the Top Tier list while RaceTrac is not. No fuel injector problems in 20+ years.

      Ask a Camry, Silverado, or Grand Caravan owner what maintenance they do to prevent deposits on intake valves and you’ll get a confused look. What maintenance? Somebody should make German engineers commute to work by car, work an extra hour or two of overtime, and cut their vacation time back to two weeks a year until they fully understand how expensive time in the shop is to American consumers.

      • 0 avatar
        Mervich

        Aside from the fact that a German company, Robert Bosch Corporation, is credited as being the inventor of electronic fuel injection, carbon build-up on valves is not, nor has it ever been an automotive brand or country of origin specific problem. Once again, those who insist on using unbranded, convenience store gas because it is two cents cheaper will certainly experience carbon build-up on the valves and therefore reduced performance and diminished fuel economy from ANY fuel injected engine. You may not believe your cheap fuel causes such a problem, but it is there…robbing your engine of performance and fuel economy. The build-up is usually gradual, so you may not have noticed the decline. Many dismiss lost performance and economy as the engine is just getting older and worn. Go around to any shop when heads are pulled and you’ll see first hand. If the valves are clean, they’ve been using top tier fuel…if not, you can bet on cheap gas.

    • 0 avatar
      Aqua225

      Mervich, nice sentiment from a German-car-o-phile.

      http://www.autoobserver.com/2011/06/direct-injection-fouls-some-early-adopters.html

      It’s a serious flaw, and GM motors aren’t exhibiting it like the German marques.

      Also, a history lesson: Robert Bosche was NOT the inventor of port fuel injection. Port fuel injection using electronic PWM modulation of injectors was originally a design implemented for aircraft engines, along with analog computer controls (no digital back then). What country you ask? The good ol’US of A and our mighty Military Industrial Complex. The first production car with mechanical injection was the ‘Vette. The first electronically fuel injected car was the ’58 Chrysler De Soto. The systems provider was Bendix.

      http://www.allpar.com/cars/desoto/electrojector.html

      • 0 avatar
        JREwing

        That autoobserver.com article was rather illuminating. It puts my mind at ease that the carbon buildup issue is preventable, whether or not the manufacturer chooses a hybrid port/GDI strategy or not.

      • 0 avatar
        Mervich

        @Aqua225
        I don’t own a German car, plus, I wouldn’t describe myself as being a “German-car-o-phile”. You do have me on a technicality, though. Allow me to rephrase. One of the first commercial gasoline injection systems was a mechanical system developed by Bosch and introduced in 1952 on the Goliath GP700 and Gutbrod Superior 600 (http://en.wikipedia.org/wiki/Electronic_fuel_injection). In 1967, Bosch was the first to develop an automotive electronic fuel injection system (D-Jetronic) that actually worked well enough to be mass produced. The Bendix developed Electrojector system was never made available to the public..the early electronic components were not equal to the rigors of underhood service and were too slow to keep up with the demands of “on the fly” engine control.

        You have entirely missed my point, though. Cheap gasoline is a major offender in carbon build-up.

      • 0 avatar
        Aqua225

        Mervich,

        There were units shipped with the EFI system on them.

        I never made any claims on reliability though. I think had it been reliable, it would have ushered in EFI on everything about 20 years before it actually started showing up again.

        They may have had better luck if they had used milspec components, in the pictures, it appears to use quite a few electrolytic capacitors as well. All a recipe for rapid tolerance drift.

        Actually, I was just reviewing that article and it appears his fixes were along the lines of my thoughts to get that system back up and running.

  • avatar
    Advance_92

    I’ve got questions about the first two issues:
    Ethanol’s been in gas for a number of year now, and levels as high as E85 for the last five. Why aren’t all fueling components in a new car designed to handle it yet? It’s one thing for plastic carburetor floats in a car from the late 70s, but a new car shouldn’t be having any issue with ethanol.

    Second, if fuel delivery is so bad with direct injection how do diesels manage it with even higher pressures?

    Are issues with direct injection due to the technology itself or a case of not using the right components?

    • 0 avatar
      jz78817

      there’s more to E85 compatibility than the fuel system materials. Ethanol has different chemical properties which affects things like burn rate and combustion chamber optimization. Being “flex-fuel” is a lot more than just making sure your fuel lines don’t dissolve.

      • 0 avatar
        Aqua225

        Not really. Unless you are running turbos, there is not much ethanol optimization occurring, and even then, most manufacturers do not allow their engine controllers to aggressively use ethanol by increasing allowable boost pressure .

        To take advantage of ethanol, you need higher compression or higher boost.

        Since flex fuel vehicles need to be normal on gasoline, they have zero optimization for ethanol, other than some software tricks related to ignition timing. These tricks are also activated passively, it takes run time & careful O2 + knock sensor monitoring to slowly allow more aggressive ignition timing.

        According to the information on the net, fuel type detection in the fuel system was long ago abandoned.

        Starting fuel, etc. Is optimized to start the engine in most temps on either. Once the engine is on closed loop control, the software just attempts to optimize combustion. There is no magic “is burning ethanol” button.

  • avatar
    Philosophil

    This is a great idea for a series. Keep it up!

    It is my rather vague, amorphous fear of this kind of thing (which Andrew has articulated very well) that made me wary of anything GDI. It’s also one of the reasons I went with a Honda (the Element) in my most recent vehicle. Rightly or wrongly, I see the complexity and fine tuning of today’s engines (and systems in general) as implying lower tolerances overall and a greater risk of critical failure in the long run. I might be overreacting here, but I usually prefer systems and methods that are simpler, and ‘tried and true’ as it were, especially when it comes to things that you want to last a really long time…

    • 0 avatar
      dejal1

      I believe it was a couple of weeks ago that Honda was ripped a new one on TTAC for being slow to jump on the GDI bandwagon and they were having their lunch money stolen because they were leaving a couple of MPGs on the table.

      • 0 avatar
        supersleuth

        Because, you know, it doesn’t take any engineering prowess to make most of your vehicles lighter than your competitors’ so you don’t need GDI to get competitive fuel economy.

        Anyway, with the new Accord we’ll get a chance to see if Honda can get GDI right. I expect it’ll still be slow to trickle down to the Civic and Fit- even Hyundai used to say that GDI made little sense on smaller engines, until, one guesses, the marketers wanted it.

  • avatar
    Mad_Hun

    I have been following this issue for years now, and have seen evidence of severe carbon / oil sludge buildup on the intake valves of VAG 1.8/2.0Ts and the V-8s, Porsche V-8s, Land Rover / Jaguar V-8s, BMW N54s and Lexus IS250s (which use DI only, as opposed to the hybrid system in the 350). I haven’t yet heard / seen the problem arise in the newer GM and Ford applications, or the Audi 3.0T, but I suspect it’s only a matter of time. With respect to the 3.0T, it may be the case that the gunk is being trapped farther upstream in the supercharger inlet vanes.
    Someone above stated that there are millions of miles of trouble-free service from GDI engines. That may be true, but only because all of those miles have been logged early in the engines’ lives. I’m convinced that without the valve washing effect of port injection, every GDI engine will suffer from buildup over the course of time. Italian tune-ups do nothing to help the problem, as gas never touches the valve.
    As stated above, I believe the long term solution will be an industry-wide adoption of Toyota’s hybrid system.

  • avatar
    WheelMcCoy

    Nice addition to TTAC’s roster. I look forward to more articles like this.

  • avatar
    stuntmonkey

    I wonder if anybody is going to try GDI + Atkinson-cycle? At low RPM, you still squirt into the chamber on the upstroke, but part of the air/fuel mixture gets squeezed out into the intake port, thereby washing the deposits. At WOT, it matters less because the EGR isn’t in play, and you could return it to conventional timing.

    At the other end of the spectrum, GDI is not absolutely necessary for maximal horsepower either. It’s not been adopted in F-1 (to my knowledge); my understanding is that those engines wouldn’t be able to utilize the cooling effect effectively.

    • 0 avatar
      racingmaniac

      Thats actually an interesting thought, considering part of the strategy for Skyactiv engine on Mazda is to mimic Atkinson cycle through timing manipulation and increase overlap at part load(not sure if the injector fires when the overlap happens), it might be an interesting way to tackle this without the use of duplicate injector…

    • 0 avatar
      Aqua225

      There are a lot of claims on being able to control predetonation better in high load situations (think F150 ecoboost towing a trailer up a hill). This would be a partial throttle, yet high load situation. Due to extra heat from operating partial throttle in a port injected engine, combined with higher cylinder temperatures from both waste heat from the pumping losses of part throttle, plus operating under heavy load, and a small port injected turbo is much harder to manage than a DI motor.

      With DI the fuel can be injected very close to ignition time, lowering the probability of the fuel heating to ignition before it’s time for ignition to occur.

      Many small gasoline engines have no doubt suffered very quick and violent deaths from over supercharging used in high continuous load applications.

      Race car engines are incredibly robust, and yet are usually only worth running one race with before being overhauled, so it’s not really a good comparison.

    • 0 avatar
      suspekt

      Yes. Toyota has. On the new Lexus GS hybrid… D4S injection + Atkinson cycle

  • avatar
    millmech

    I remember something from years ago- BMW had trouble with deposits on the back of intake valves. The valve cleaning procedure was to run ground walnut shells through the intake.

  • avatar
    highdesertcat

    Excellent article. Thank you. This could also explain the sentiment of more and more drivers/owners of new cars and new trucks to trade their new vehicles before the warranty period expires.

    Let subsequent owners get stuck with the repair bill! Get the most out of your trade-in before you have to pay to get it repaired.

    I am actively following the unceremonious aging of my grand daughter’s 2011 Elantra which we helped buy for her last year. I fill that car up from the stash of Shell Premium gas I keep around the house for our vehicles and AC-generators (in case of power failure).

    So far, so good. With 35K on the odo, using 15% Ethanol-spiked Shell Premium gas, the Elantra runs good and continues to drive well. Will put new tires on it this coming weekend.

    Gas mileage is 28mpg at best because the girl travels with three other chubby teenage girls to college and has a lead foot like her grand mother.

    Articles like the above convince me that my resolution to trade our vehicles every 3-5 years makes more sense than to try to keep them running after the warranty expires like I did in the past.

    All these new-fangled ‘improvements’ come at a cost, and durability, reliability, clogged injectors, etc, are just some of them. Even the financial aspects of professionally repairing GDI and TGDI engines is depressing.

    • 0 avatar
      CJinSD

      Not that I’m trying to suggest that Hyundai builds any keepers, but didn’t the Elantra get spared the GDI system used in Sonatas and Accents?

      • 0 avatar
        highdesertcat

        Indeed! My grand-daughter’s 2011 Elantra has the 1.8L VVT engine hooked up to an automatic slushbox. Right now it appears to be a keeper, at least for the 100,000-mile warranty period.

        The reason I am actively following the aging process of this Elantra is because it gets around 600 miles per week added to the odo commuting to college and all the Elantra haters seem to think that the Elantra is a rough-riding, trouble-prone car.

        In my experience so far, that is not the case. That little car just runs and runs. Starts up in the morning and goes where it is pointed.

        Here is the cue I’m looking for: if the 2011 Elantra proves to be a problem child for whatever reason, then the GDI version with its valve carbonization, high-pressure fuel injector failure and related higher maintenance costs would be one to avoid.

        But the whole industry is gravitating towards more complicated fuel injector systems as a matter of fuel economy and power management.

        Knowing which brands to avoid could be very helpful if I were to be faced with that predicament the next time I go shopping for a new car or truck.

  • avatar
    tjh8402

    What’s the additive Hyundai recommends for their engines? Makes one hesitant about buying one used since how many typical Hyundai owners do you actually expect to use it?

    Besides the good ole Italian tune up, what can be done to help a DI engine and fuel system out, especially since E10 fuel is pretty much de riguer now? Although my current car is tradition port injection, I’d say that a 35 series Bimmer, TDI VW/Audi, or Focus/Skyactive3 could always be a potential future ride and I’d wanna keep any of them for a while.

    • 0 avatar
      Mad_Hun

      Italian tune-ups don’t do much, as the fuel never touches the valves. Some say oil catch cans are beneficial, although I’ve seen plenty of buildup in VAG products even with catch cans.
      If you’re running some of these motors, it’s a no-win situation.

    • 0 avatar
      ppxhbqt

      Hyundai recommends Top Tier gasoline, for cars even without GDI, as many makers do. They only recommend a special additive if Top Tier gas isn’t used and starting or other problems develop. Not a GDI issue.

  • avatar
    Volt 230

    I’ll stick to Toyota or Honda, thank you very much!

    • 0 avatar
      supersleuth

      Except Honda is about to jump on the bandwagon with its “Earth Dreams” engines. (Though I don’t know whether they will feature the dual port / direct injector scheme.)

      • 0 avatar
        CJinSD

        This seems like a reaction to the media panning of Honda’s durable and efficient engines. Stupidity has hit critical mass and there isn’t a sufficient market to justify cars for people that think. We just added a third new K-series powered Honda and are prepared to wait out the teething issues before buying any more.

      • 0 avatar
        Volt 230

        In that case, I’ll stick with Toyota and if they go that route, then I won’t buy anything built this Century. Besides I prefer the mid 90′s Camry overall.

  • avatar
    ismail

    again please read my previous entry, VW/AUDI seems to be adopting the
    same method as toyota with the hybrid injection…

  • avatar
    Pastor Glenn

    When I had a deer accident and had to rent a car, I got lucky enough to obtain a new ’11 Hyundai Sonata which has GDI. I liked the power, but not the tappety clacky noise of the GDI, which I’d heard on virtually every other GDI car also.

    So when it came time to buy a Sonata, I selected the Hybrid car which does not have GDI.

    The other reason was 0% money for 5 years as well as having owned 3 prior Sonatas over the years (we’re now a 2 Sonata family) and having owned 2 Prius cars before.

    Hybrids are excellent cars for Pastors because we do drive a lot and good stewardship of money is something worth demonstrating to others.

    Payback time on the hybrid Sonata is supposedly just under 6 years (as opposed to 28 years for the Chevy Volt), but this is not so much different than diesel Mercedes and VW cars of years gone by – with the added reliability of NOT being a VW.

  • avatar
    Pastor Glenn

    In response to the Top Tier gasoline discussion, I would add that if you are concerned about getting “slugged” with E15 (despite the fact that the pumps are SUPPOSED to be marked as having 15% ethanol) you should use Top Tier gasolines since the maximum amount of ethanol “allowed” is 10%, minimum 8%, according to toptiergas dot com.

    This is important because when I checked with Subaru (prior car) and Hyundai, I got IN WRITING that if E15 is used and there is engine problems, the warrantee is VOID.

    Even Top Tier gasoline won’t clear off the valve stems in GDI engines for the aformentioned reasons; the fuel never touches the valve stems, only oil vapors from the PCV system do. I have to wonder if this is why Hyundai are recommending a cleaning agent for the engines periodically.

    And yes, my Hyundai Sonata Hybrid owner’s manual (which is vastly different from the regular Sonata owner’s manual since it is built in a different factory, for one reason – not to mention it is a vastly different car) states clearly that Hyundai recommend Top Tier fuel.

    I exclusively use Top Tier fuel AND I keep my receipts for fuel in case of engine trouble for both of our cars. I do find it keeps the engines in top shape (and putting on over 20,000 miles a year means I have some experience to prove my point). I’ve had no internal engine problems at all since switching to Shell or Mobil (the only 2 Top Tier fuels in my area), and I prefer Shell.

    A friend of mine races stock cars and states unequivocally that Shell fuel is best; he can use one ‘grade’ of fuel lower than other fuel manufacturers and get the same results, and uses premium Shell in his race car with excellent results.

  • avatar
    wmba

    Great article on a timely topic. Judging by some responses, a lot of folks don’t really know how engines work, so perhaps further articles using more graphics would help out.

    My reading suggests that Ford aims a bit of the DI fuel spray on the back of at least one open intake valve on Ecoboosts. That should help a bit. What GM does seems to be a bit of a secret, but they seem to have little trouble with DI. I never count out GM engines – while not a fan of their vehicles, I have massive respect for their engineering, particularly engines.

    Then there are the DI disasters. VW/Audi leads the way, just google it and learn what that company faces for intake manifold sludging issues, let alone coked up intake valves. They deny it as usual, but it’s a real problem, both 4 cylinder and V8.

    How about the Mazdaspeed3. Not good. The Toyota IS250 engine is bad, but the IS350 with D4S is good. Almost as if Toyota is proving the problem with DI, and letting a small number of owners of the cheapie Lexus IS250 act as guinea pigs in an experiment.

    BMW has had DI sludge problems, which is why the new N20 turbo engine goes to heroic lengths to minimize oil splash which would cause oil droplets to get through the PCV valve and cause the problems Andrew Bell describes. The EGR system is similarly heroic:

    http://www.bimmerboost.com/content.php?1662-Full-BMW-N20-turbo-4-cylinder-technical-specifications-plus-history-overview-and-N52-comparison

    I think (only an opinion) that Toyota lucked into the D4S system, in that they were not looking to minimize intake valve deposits, but wanted certain characteristics. The fact the port injection cleans the valves is just a bonus. Reams of pdfs on the system and its development since 1998 at FT86 club.com.

    Also, the Society of Automotive Engineers has dozens of papers on DI sludging, so this is no secret, my friends.

    Ah, but today was a great day, as I had a Subaru BRZ all to myself for 15 minutes this morning to poke, prod, adjust the seats, telescope the wheel etc. etc in preparation for the grand opening of new Subaru premises in Halifax, NS today. Attended the gathering this evening and it was quite the event for a car dealer. The BRZ’s VIN ended in 00001, so had to be pre-production. Nice seats but low-mounted, shiny cheapo interior, sitting in a coal scuttle feeling, but good visibility, wide inside, needs a center armrest badly, cheapy headliner, boink, boink, finger-flicking light six speed gearlever.

    I think I want one, and it has D4S DI. Hooray!

  • avatar
    John

    I’d go for a high-tech solution if I had one of these fancy-dancy motors: fill the PCV valve with JB Weld, pop a valve cover, get out the old hole saw, and cut a hole for a $20 breather. Problem solved.

    • 0 avatar
      Aqua225

      I like this idea, would probably boost perf too, by keeping oil vapors out of the cylinders, and keeping the engine cooler.

      However, I am not sure the motor would pass inspection in most states after that change. Isn’t that considered to be emissions equipment nowadays?

      • 0 avatar
        John

        It would pass a tailpipe sniffer, but not an inspection. They stopped doing emissions testing where I live back in the nineties.
        You might want to put that breather in a place where it isn’t easy to see. By the way, that’s why I said fill the PCV valve with JB weld instead of just chucking it and filling the hole – looks stock. In fact, you’d probably get a lower hydrocarbon reading with a tailpipe sniffer, cause all that PCV sludge doesn’t really burn up in the combustion cycle.

  • avatar
    TW4

    Relevant engineering and ownership info. Thanks.

  • avatar
    icemilkcoffee

    CNG/LPG engines went through an analogous problem. CNG is a gas vapor, unlike gasoline, which enters the combustion chamber in liquid droplets. Converted CNG engines were plagued by intake valve wear. With CNG, soot is generally not a problem. It was postulated that the little droplets of gasoline deposited on the intake valve actually prevents valve wear to some degree. Another theory is that the intake valve just got too hot without the cooling action of the gasoline droplets. So what the manufacturers did in the end was to use hardened valve seats and left it at that.

  • avatar
    Princeps

    Minor niggle: EGR does not “lean-out” the combustion mixture, it dillutes it.

  • avatar
    acuraandy

    Well, I guess i’m screwed as I just bought an Accent with GDI. F.

  • avatar
    maxwell_2

    Nissan has a GDI engine in the Juke, pretty interesting design.

    1.6L direct injection turbocharged in-line 4
    Sodium filled exhaust valves
    Beehive valve springs
    Twin variable camshafts
    Designed in conjunction with Renault
    Nano finished camshafts
    Hydrogen free DLC coated lifter buckets
    Four counterweight crankshaft
    Primary and Secondary injection events
    Stratified combustion
    Electronic engine mount

    • 0 avatar
      John

      Now THAT is what I call the “Law of Diminishing Returns”! All that Formula One engineering to make a street car get 3% more mpg and a bit more power.
      Wonder how much added cost that came to compared to sequential port injection?
      The NEXT generation of engines will probably make a Corolla/Cruze cost over $100,000.

      • 0 avatar
        wmba

        Formula One engines use Port Fuel Injection.

        Direct injection is far better than port injection for minimizing emissions on start-up, and throughout the rev range. Formula One couldn’t care less about emissions.

        The car companies aren’t spending the money on DI just for the sake of it. Emission regulations are driving the change. Blame government. This discussion is about implementation, and how some companies are much better at it than others.

        As for the above poster epoxying his PCV valve shut with JB Weld and installing a breather. What an effing great idea … not.

  • avatar
    Bill Wade

    I have a 2000 Mitsubishi Pajero IO with the 1.8 GDI engine with just over 175,000 miles. While the thing is over all a wretched car barely worthy of GM’s build quality I’ve had no trouble with the injectors or fuel pump. Since new though it tends to miss slightly at steady throttle position regardless of speed.

    I’d like to meet the engineer that designed the tappets. They start to clatter about 1000 miles after an oil change.

  • avatar
    RobW

    Well that does it.

    I was just about to head to the Mazda dealer tomorrow to buy a new 2012 Mazda 3 with GDI here in Raleigh NC – after narrowing the field down to it and the “new” Honda Civic.

    I preferred the Civic’s appearance (it’s overall shape and lighter colored interior) over the Mazda’s dreary all-black cave-like darkness. I know the all black interior (and Mazda products in general) cater to the Driving “enthusiast” but while the dashboard is cute – that interior color scheme sucks to no end and just depresses me.

    So much so that I planned to buy the sunroof option just to get some light in there – and re-do the upholstery to a lighter beige or something ($$$).

    But all this talk of GDI’s dark side has got me losing sleep. The Civic’s lower tech (non GDI) engine starts to sound like the best thing for me – as I simply must keep my next car for + 250K miles – without a bunch of repairs. I just retired – and future cash flow is a huge concern for me.

    I’m thinking that the GDI’s 12:1 compression ratio could add strain on the crankshaft connecting rod bearings – especially later in the car’s life when all those other problems you fellows are discussing start to add up.

    Engine ping may not be an issue under +50K miles (which can shorten bearing life) but who know’s what will happen when I get past +150K miles.

    The Civic’s engine performance was pretty impressive – I repeatedly test drove it on highway entrance ramps at 70 mph – and it still pulled strongly past that speed – with very nice AT transmission gear changes.

    But that psychotic Civic two-tier dashboard was the sole reason I chose the Mazda 3 (even with it’s cave-like interior darkness). Mazda’s only option to all-black was a pukey yellow-tan leather.

    I don’t understand why Mazda is so chincy on color options – and why Honda is so screwed up on dashboard design. These guys are really starting to P*** me off.

    I test drove the Kia Forte (same all-black interiors and a crappy as hell looking dashboard). The Ford Focus & Fusion – while good on paper – just left me cold. I’m considering just looking for a good low-mileage 2007 Accord and call it a day.

    Or maybe find a 2011 Mazda 3 with no GDI – now there’s a thought!

    Whether perceived or real – these GDI reliability issues really screwed up my day. And they seem to be real for many manufacturers.

    Rob

  • avatar
    DRJJJ

    Toyota’s 2.5 and 3.5 liter in the Camry and others don’t have DI, they’re still top tier for MPGs, less noise, long term intake issues, dramatically lower fuel pressure, etc-smart!

  • avatar
    vinti

    i want to know what if we use pfi injector for gasoline direct injection does any reverse flow due to high pressure of cylinder occur . actually i am working on two stroke gasoline direct injection and using pfi injector instead of gdi injector at a compression ratio of 7.4:1 but air is coming through the injector to the fuelo line when we are not injecting or the injector is closed.


Back to TopLeave a Reply

You must be logged in to post a comment.

Subscribe without commenting

Recent Comments

New Car Research

Get a Free Dealer Quote

Staff

  • Authors

  • Brendan McAleer, Canada
  • Marcelo De Vasconcellos, Brazil
  • Matthias Gasnier, Australia
  • W. Christian 'Mental' Ward, Abu Dhabi
  • Mark Stevenson, Canada
  • Faisal Ali Khan, India