1KD pistons cracking, the rebuild solution.

[punjehl]

Ok Im starting a new thread about pistons cracking as there are no in depth investigations of this phenomena regarding possible solutions and im presently reboulding a mates 1kd that had three cracked, one all the way through to the cooling gallery , it was blowing down in to the case. The driver report was loosing power suddenly and had to go for the next gear on the flat then thought about it, cant pull constant speed on the flat there is a problem and they stopped the car. thankfully before too much damage was done. in this post i will include some failure analysis and a proposed solution and a discussion about the many different models of piston that are out there for this car toyota have at least 6 distinctly different pistons for the 1kd in their parts books, I can find 5 after market types of piston that are available here in australia.

So many pistons in one engine, some of them fail if you have pics of any failed pistons please upload them to this discussion as i have a suspicion that only one or two of the types of piston fail easily and as a matter of course(toy part 13101-30090 this on e has the deepest and bowl and is the structurally weakest piston), the other ones being more robust. i have only seen pictures of this type broken on the internet and i find this interesting seeing as the type of piston fitted can be found by ringing toyota with a vin number i would always check a potential buy for these pistons and if that part number comes up go look at a different one.

bear with me there will be lots of pictures soon as i have time to put up this information

I will also cite some articles about piston fatigue and identification of modes of failure

The evolution of the faulty (weak ) piston from the first 1kd piston :

the first piston that went in the 1kd was what apears to be a copy of a dyna truck piston it looks nearly identical, toyota had evolved a reliable piston on the dyna for direct injection that didn't have problems. then they had to chase euro 4/5 EC and to do this and bring out a new model with a bit more power every couple of years, they dropped the compression ratio with a different bowl geometry in each instance they deepened the bowl first in the next piston then in the one after they made it wider . They did this to keep combustion temps down below euro cert levels. In my information i have only bowl geometry data from one after aftermarket piston manufacturer. Please post more measurements from others pistons thew are bowl diameter at the top bowl depth at deepest point and distance between crown face and top of the dome in the bowl, if you can be bothered to cc the bowl then we can work out what compression ratio each one gives.

The fix according to toyota is to replace faulty injectors and replace the pistons with the 13101-30200 I don't know what they look like but one after market piston is a copy i think, please correct me if i am wrong the bowl dia. is 62.3mm up from the previous bowls of 53mm 51.8 and 51.9 and 49 . The crown is thicker the cooling channel has shrunk to increase the strength and the bowl is shallower at the deepest point 16.5mm, the deepest on is 17.5 deep and 53mm dia. this one is the commonly breaking one

Toyota technical service bullitins say the new piston is stronger and has overcome the problem of cracking due to bad injectors. The early pistons don't fail from bad injectors this says something about their strength.

see this post for details of someone elses experience with cracked pistons
http://toyota-club.net/files/faq/15-10- ... on_eng.htm

Mode of failure in this car :

Cracking caused by thermal cycling stress in the bowl rim exacerbated by flexural weakness across the gudgeon pin bushes, this is clearly evidenced by the starting point of the main cracking in the bowl rim and the fact that there was a third smaller purely thermal crack originating 90 degrees away from the gudgeon pin bushes. the crack originated in the highly thermally stressed bowl rim from metal compaction due to thermal expansion in the bowl rim against the steel ring carrier and metal containing the cooling channel staying relatively cooler and expanding less thanthe bowl rim. on cooldown this compaction from expansion becomes tension from thermal contraction. this causes the cracks around the bowl the cracks happen to start where there is exacerbating flexural loads on teh piston crown above the gudgeon pin bushes (photos to come) scientific article about piston fatigue ( https://levilentz.com/work/Classes/MFG/stress.pdf )

The newer wider bowl piston looks more like a scania truck piston so the evolution is finally going in the right direction

while its apart the engine Im rebuilding is getting ported, nothing radical just improve the swirl dams and radius the valve pockets in the combustion chamber to allow more gas flow at smaller openings and improve the swirl further. and clean out the bowls behind the exhaust valves, photos to come. The rods will be balanced and the EGR removed with a bypass that feeds from the clean side of the filter in the air box. This engine was so gunked with carbon/oil/tar that it had lost more than 50% of the port area in all cylinders and had giant stalictites in the valve swirl chambers and huge growths on the valve stems the level of gunking had to be seen to be believed .

The rebuild mods are mainly about improving efficency and longevity and allowing the engine to make its full power over a wider rev range without changing the tune of the computer .

So which piston to choose ?
One of the early shallow bowl versions, highest compression 51.9mm dia. bowl 16.1mm deep 3.7mm dome top to crown.
Arguments about: (the russians like this one because of the cold weather starting performance). higher compression is good for very cold weather starts and better thermal efficency but increases strain and friction of rings. there is a forged all alloy version of this one with out the cooling channel and alfin ring carrier. This is a race spec part but the ring life will suffer due to the lack of a hardened steel top ring carrier .

The later model piston with the much wider bowl, lower compression
Arguments about: lower compression, trade a loss in thermal efficency for less friction from the rings, lower peak cylinder pressures and temperatures, and a lower stress build that should live longer and the ability to handle increased boost .
This one looks more like the widest bowl pistons in the 1vd

The 1GD pistons in later models of hilux/hiace/prado are the result of toyota useing their customers to perform on road testing of unproven components and also look more like a heavy duty late model scania truck piston than any others previously.

[hiluxxury]

Thanks for that. I did a similar search when I got worried about my pistons. I have a 2005 1kd so fingers crossed its ok. Given how many updated 1kd pistons there out there being "road tested" I'm sure the failure rate is still relatively small.

With all of the build-up in your engine, did you check to see if your injectors were leaking?

[Gipsy]

I would be interested in when did they find the right piston for d4d? What year is reliable?

Edit... actually I would like to know if they ever did?

[apSlain]

Awesome that you'll document the process and attempt to provide more information.

Potentially related videos that I always see pop up:

[punjehl]

Hi again, yes i have seen the above videos posted by apSlain. In these videos the guy failed to identify the fact that his pistons have been hitting the head, as evidenced by the bruise marks on the crown and the mashed dimensions i have talked to him but he refuses to admit his mistake in the diagnosis his cracks are so minor that the engine would have run for a long time longer if it had not pounded the slugs against the head. Further evidence for this is that they are shiny on top.

In response to hiluxxury, the car had new injectors 20,000km ago after the originals failed , this may have been a contributing factor in the piston failure although i have seen one other before that had preventative replacement injectors every 100,000km and had blowdown after the crack widened to the sump at 245,000km and its injectors never were allowed to fail.

and now for some photos

[punjehl]

Here are some photos,
http://imgur.com/U2aPA71
original failed piston (see daylight through hole)
http://imgur.com/GvwCd3H
new after market non toyota item (dont know protocol about posting brands and rules about product promotion, if you want to know what it is contact me )
http://imgur.com/uFXhjqc
side by side
http://imgur.com/PdJlWZ6
detail of oil cooling gallery that is 7mm dia down from 10mm in the original and a heavier pin boss area
http://imgur.com/rTxp6aJ
compression thrust face and slipper detail the new one is narrower for better alignment and less friction
http://imgur.com/W2gVbHw
power stroke thrust face slipper detail, the new one has a wider thrust face slipper on the power stroke side
http://imgur.com/rzZTUVO
intake port, cleaned a little showing the original shape of the swirl dam
http://imgur.com/yMhrIhT
intake port after porting work to improve the swirl and open up the passage a bit, smoke flow through the port showed much greater swirl and better flow after the work.

presently have the bottom end together and the rest of the assembly is on going will post more photos later .

sorry the pics are too large to display ill resize them later

[Rocket55]

Check this thread out for posting pics. Note the forum will only display 640 pixels in width.

viewtopic.php?t=9833

[punjehl]

Progress update
the engine is 90% together
had a minor hicup with fitting the main bearings, the after market ones were a bit on the loose side with clearances between 0.09 and 0.1mm, these are runout values and manual says regrind at 0.10 mm however because clever toyota for reasons of sloppy manufacturing tolerances make parts that are not to spec so they use their 5 sub sized micro clearance difference graded parts to go with the sloppy manufacturing , "ho hum crank main pin number one is a bit larger than 2, 3, 4 and 5 so we will just put a number 3 bearing in instead of a number 4 " toyota supply bearing shells in 0.020mm increments with five sizes between each typical undersize bearing as you would find from an aftermarket maker. So i miked and plastigaged the originals and supprise they come up much tighter than the standard size after market ones, it was a PIA swapping in and out the crank and measuring everything twice that ate up too much time but the old shells were in what looks like nearly mint condition and gave a range of clearance between 0.045mm and 0.060 well within spec, a note on clearances race boys will usually build a little on the larger size to get more room for shock loading and warpage of components under extreme stress and allow the use of thicker oils 20w50 or higher and a bigger clearance doesn't seem to affect the engine lifetime. the exception is cars that race real hard like circle track nas cars that are rebuilt every race and have very tiny clearances so they can use watery oils like a 0w20 or 0w30. Not in y car please

Any way i digress. the engine went together smoothly the oil squirter nozzles needed to be bent by about 1mm to center on the new oil cooling passage holes in the piston and after the minor bend they center with about 1.5mm clearance from sides of the the hole in the piston assembly of all timing components and the head was finished leaving the water pump for which the wrong gasket was supplied in the kit. oh well i reach for the roll of orange clay packed high quality gasket paper and a razor blade.

engine is now painted kermit green because after the mods it should be greener than all the others out there by virtue of being more efficient

next update will be details of the egr bypass solution and on road testing .

[TOYZX]

Nice work mate. Keen to seen the egr bypass! Been looking to do this myself, just trying to get a few bit together.

Sent from my SM-N920I using Tapatalk

[punjehl]

My thoughts have changed and Im looking at an Australian made device that defeats the computer electronically in combination with a blanking plate that fully blocks the egr and offers + and - 15% fuel trim adjustment in the cabin with a simple knob, it is a very clever device that interrupts and modifies the signal from the maf sensor using egr sensor position to fool the computer into seeing the air mass flow readings change as it expects them to with egr operation and can vary the air mass reading to trim the fuel quantity as well, people using this device have no complaints so far and see some impressive fuel efficency and power gains . it is the manual mapper2 from ozbush electronics for the 1kd engine it is one of the cheapest tuning devices around and unlike the rail pressure fooling devices i can see no way that it will shorten the injection components life. One is ordered and will be fitted when it arrives from Sydney. I will post the results here after some testing . I think this is the best solution for fully disabling the egr system and keeping the computer happy

[hiluxxury]

It's an interesting bit of kit that Ozbush controller. The Prado folk seem to be all over it... but I've only seen it mentioned here once or twice in passing.

I'm keen to find out more, so please yes, post more info as you learn more.

[davo94]

Dont expect to notice the fuel trim +/- 15 it has proved ineffective in a 2012 hilux.

[martynvella]

Thanks for the detail in your thread.
I have been interested in finding out expected valve life, but none of the mechanics on here have experienced a 1KD fail because of valves. It wouldn't have any relevance to the piston failure but nobody had gone into the detail you have and I doubt after going into so much detail with the bottom end that you will just slap the old head back on without examining it.

Can I ask how many KM's was on that engine before it failed?

Were the valve clearances still in spec?

Do you know if the seats in these engines can be reground or are they disposable heads?

[fourby4diesel]

Good write up. Although I have not read it all yet.
The pistons to use are "110" more robust than the old "080".
Also keeping injectors fresh is the other side of the prevention, New denso. Also regular diagnostics.
IMO spray pattern, possibly affected by egr soot, carbonisation to nozzles could be a contributor.

[punjehl]

read on for the answers to your questions asked above:

The moment of truth arrived the engine fully assembled descended into the car and bolted in nicely after a fair amount of wiggles and jiggles to get the new clutch to line up and it eventually found the splines on the gearbox and slid home. after several attempts at starting remembered to tighten the high pressure pump to fuel rail connection engine started fine after that

The egr system upgrade this consisted of two blanking plates and a longer hose to bypass the egr cooler's coolant supply the cooler is in a spares box we took it off to give the air another way to get out from under the intercooler. both ends of the egr were fully blanked off one on the cylinder head egr supply port and another on the egr valve it self. the car was driven for a couple of days like this with no ill effects other than the MIL light coming on scan tool showed egr flow error but it did not affect engine performance. we then fitted the Oz bush manual mapper 2 and it clipped in easily the connectors all fit perfectly the only thing to do was to zip tie its loom into place mount the in cabin control and plug it into the egr position sensor and airflow meter .

The egr bypass box: it works great no more MIL light and no more engine codes after a couple of days of driving there are no error codes in the computer. it says + or - 15% on its knob this trims the airflow meters output two ways it subtracts various ammounts of air flow from the real reading depending on the position of the egr valve position sensor so that the computer sees the expected changes in airflow from activating the egr system, at the same time the knob lets you trim the airflow reading sent to the computer by a further percentage. i dont think it means that fuel delivery is varied by up to 15 % either side of normal. it is trimming the airflow meter signal this is a very safe way to fool the computer to inject a bit more or less fuel as it does not mess with rail pressure like those dangerous juice boxes that will shorten the lifetime of overloaded injection componentry. i have been talking to the guy who builds them about a three way fooling device that is still perfectly safe as the main computer still has control of fuel pressure and timing it would trim the MAF meter and the MAP sensor and the IAT sensor. boost trim would allow no more fueling but keep the engine burning cleaner and cooler in hot weather as the computer would have to see the lesser value to allow more boost to develop but i think big gains could be had by fooling the intake air temperature .

some claim this box does nothing but cause the computer to ignore the dead egr system but some report that it can increase smoke at near full trim on a stock motor . on this car there is a definite power improvement to be had but not enough to make it smoke even a tiny wiff this says that some other peoples cars don't breathe as well as this one and the difference is subtle probably only 4 or 5 % more fuel than standard we tested it on a big hill and the hill climbs were consistantly faster with it turned up to 15% from 0 and consistently slower at -15% so it works and so does the complete removal of the egr system



Martynvella the valve clearances were all on the tight side indicative of valve seat recession wear, none were riding on the cam yet but several exhaust valves were very closest to ride limit showing polishing of the cam further along the loading and un loading ramp and had worn off the coating on the back of the cam, some valves were less recessed and lacked the polish to the factory coating left on the back of the cam circles , the seats were badly pitted and worn by the harder valves that showed very little wear against the seats. the bores would have gone another 300,000km easily before loosing their aparently perfect honed surface it was rebored because the burned oil ring on teh blow through piston had scraped a ditch in the bore wall about 0.15mm deep easily felt with a finger is too deep to hone out.

the head was skimmed to make flat again as was the block which was 0.15 mm high at the front left corner of the block the piston tops were skimmed the same value as the block to avoid excessive protrusion ( skimmed block and pistons 0.20mm) the head had its seast recut with a careful 5 angle valve job and the valve pockets radiused into the head roof on three out of four valves see photos. The valves were refaced and reused the tops of the valves were ground down to make the new clearances correct as following toyota's recomendation to replace valve buckets with different value valve buckets is both uneconomical and unnecessary unless you plan to correct the clearances with the head on the car . if they are that far out it needs a seat and face job anyway.

The heads are not disposable unless you mill too much off the face even most cracks can be repaired but i have not seen any that are cracked they seem to be a well designed casting except for the shitty swirl port shape. the ports were all cleaned with a die grinder to get rid of casting flash and generally poorly dfesigned swirl generating shapes, all sharp edged and sharp angled transitions were removed from behind the valve seat recess machining and the swirl outer bend and roof opened up because gas flow likes to stick to the outside of the curve. on my primitive test rig that gives no real numbers but shows gains or loss very well ( a vacuum cleaner with a cup that you put on the head face and draw air out from around the open valve and a vacuum guage.) shows higher vacuum for lesser flow. head intake ports flowed 25% less pressure drop across fully open valves and 45 % more at with the valves open 1 mm than standard before porting

A note about porting it is all about increasing the time area for charge air to enter not just about making ports bigger fixing shrouded valves is often the biggest gain for the smallest effort. Have to be careful in unleaded petrols because that removal of metal from the combustion chamber ought to lower the compression ratio right ? wrong it lowers the static compression ratio but raises the dynamic compression ratio substantially as more air can get in in the same time it increases the VE and the car goes faster after but needs premium where it never did before.

this shows the biggest gains are to be had from the smallest work of opening the valve pockets in the head face around the valves changed around 3/4 of the circumference of 1 intake and both exhausts from a 45 degree cut to a 30 degree cut with a smooth radius into the head face. less than 1/4 of a cc of metal removed this should make a big difference at the top of the rev band as there is much increased flow capability at the end of the valve time area

The injector spray pattern on the tops of the old pistons was very even and symetrical like a perfect 8 pettaled lotus. and egr crud IMO would have little or no effect on spray pattern only hole erosion from cavitation changes amount of fuel metered out each nozzle hole the pressure is too high for anything from in the cylinder to cause blockage of the nozzle holes unless there are depositing contaminants in the fuel but even then zinc doped fuel tests show build up around the nozzle exits on the outer surface of the injector tip with very little build up in the nozzle holes . stuff comes out at kilobars nothing is going to stay in there that can get in there from the cylinder . egr soot kills valve seats fairly fast and rings very slowly

for recon injectors get the DLC coated ones from baileys they have been building ships injectors for years and have just recently started selling flow matched DLC coated injectors for a bit more than half what they are from denso but wilt a much longer lifetime

i used the hypertec 13101-30200 equavalent pistons that are IMO the best ones available stronger wider superiour geometry bowl and lower compression ratio

the problem of scale and tiny direct injection engines,
getting the fuel to burn before it hits the wall is the main difficulty with bigger bowls its easier simple physics if the bowl is too small it will hit the wall and cause huge temperature gradients . these new bowls generate a torroidal vortex and move all the air through the spray a problem subaru found building their diesel boxer was swirl aids combustion but as the air rotates about the bowl axis the burned gasses from one injector spray nozzle hole get pushed infront of the next injector hole and it sprays into a zone of low oxygen and burns less cleanly and efficently than if it was spraying in to clean air . The overcame this problem by aiming the nozzle holes at different angles to the bowl axis so that they were nolonger spraying in the same plane . there is another way to overcome this gas flow problem. looking at the cross section through one side of the spinning donut of air in the bowl. if the piston bowl has a cone in the center that ends in a drop off to the deepest part of the outer bowl bottom. the initial pilot injection creates a wave of hot expanding gases to move outward and hit the bowl wall where they are forced downward and under and inward where they enter the the injector jets from below at 90 degrees to the jet. this generates a two phase rotation in the donut of air one about the bowl axis and another about the donut cross sectional axis causing gases to generate a lot more turbulence at the injector spray mixing zone and because the bowl is wider more of the fuel is burned and less impinges on the piston surface . better efficency and less friction from the lowered compression


The car had done 209,000 km since new and had injectors changed at 140,000 due one of them leaking fuel continuously ( white smoke at start up and a nasty knocking ) ironically the failed injector belonged to the piston with the least damage. incedentially the injector malfunction was not allowed to continue for a long time they were changed within 100km of the manifestation of the nasty knocking noise. this points to failure related to other causes the 13101-30090 piston is geometrically the most structurally weak of the designs i have seen as well as lacking the ceramic fiber reinforced bowl rim found in the earliest piston 13101-30030 the thickness between the bowl wall and the cooling channel is less than 3mm and this is where the crack travels to once initiated in the bowl rim. the three damaged pistons are a good study of the crack propagation each one showing a greater length of crack.

Piston one showed the only one small crack running radially from the bowl rim edge toward the edge of the piston about 1 cm long above the rear gudgeon pin bush.

number two that suffered a bad injector for about 100km and was driven very conservatively to the shop for repair. this piston had no visible cracking. piston three is the one in the above pictures can anyone view them please tell me and i will correct them if unreadable through the link.

piston three was the most damaged with a crack extending from the edge of the crown to the read and across the bowl stopping one cm short of the crown edge the crack extended below all the way through the gudgeon pin bush to the rear and stopping at the piston pin circlip groove in the front gudgeon pin bust. Fire had burned a plasma channel through the side of the bowl where the material eas thinnest into the 10mm dia oil cooling channel and on through the side of the piston and out into the oil ring groove. the oil cooling gallery had a river of slag coming out its drain hole in the underside of the piston you have probably used a gas torch before . the driver reported a slow loss of pawer over about 20 seconds and then stopped the car as they were cruising on the flat and the car could nolonger maintain speed at full pedal. think of oxy torch running at about 100 bar it burns alloy very fast once the burn through begins.

piston four was much like piston two interesting that they crack more on the rear side gudgeon pin this matches the worst area for weakness and flex loads and the area of the highest thermal gradient in the piston the relatively cool oil is supplied to this cooling gallery entrance on the compression slipper face side of the piston pin to the rear of The engine .

thermal shock from repeated thermal cycling exacerbated by tension loads in the crown from flexation across the gudgeon pin would be the cause of failure in my opinion exacerbated by airflow asymetry between individual cylinders. The computer only sees total air mas flow and delivers a balanced fuel load to all cylinders. cylinders 1,3,4, had severely obstructed valves from egr gunk we removed 4 coffee mugs of it from the intake manifold alone and more from the cylinder head ports. for those of you who have attempted to clean a badly gunked intake tract on the car in my opinion doing it without removing the valves from the cylinder head is almost a waste of time as it is impossible to clean out the helical swirl port design by accessing only the intake ports on the engine. With the factory AFR's the computer tune uses that IMO fuel the engine too close to smoke limit becuase they smoke a bit under full power when new is burning too hot and needs some more boost to cool it but they cant tune like this because of nox emissions. more boost = a cooler burn with more free oxygen to complete combustion and more inert gasses to work as expansion mass and dampen the temperature peak. more boost never killed a diesel engine unless it was a weak design or had too much fuel stuffed in.

will keep posted with fuel economy data as car is run in

so far another success story and the 07 hilux does talk obd2 makes it easy with a cheap scan tool

[punjehl]

today chris reported lost power and engine light was using the mapper device at +15% airmass trim up a big hill the ambient tempreature is low today about 17 degrees c have yet to read the error but he said he stopped the engine and started it again with the trim set to 0 and it drove normally and the light went away. i suspect the computer saw airmass values above its set safe limit that it assumes means burst intercooler or piping and goes into no boost limp mode and continues to run using the hot maximum air temperature at atmoshperic pressure tables to control injection and opens the turbo vanes to limit boost.

it will be interesting to see what the code is and if there is a freeze frame or if after the restart the computer detected normal parameters within the safe limits and stopped ignoring the sensor that may have been malfunctioning. and turned off the MIL light. some one else with a large efficient front mount intercooler reported this problem with this device before.

Motors with a freer flowing intake system, ported and or with larger high efficiency intercoolers. will likely see max airflows sooner than those that are un modified with the trim device. Because they are already exceeding the factory airflows by a good margin and the trim device adds airflow to far toward the safety limit.

so either way it sees sensor malfunction and cuts boost and fuel to safe conservative values or sees it as a boost vent and ignores the MAP and MAF and uses conservative safe values. not entirely sure what the limp mode would consist of but if you use the conservative worst case scenario for all parameters you are safe that is revert to a fueling map that assumes maximum intake air temperature 3000m altitude air pressure, and no boost. at the same time open the turbo vanes to lowest boost position. thats how i would do it if designing a computer system.
call it survival mode. or limp going to plug in the code reader to see whats up

up date later

[fourby4diesel]

Interesting write up, Thankx for going to the effort.

[punjehl]

IMHO the DLC coated baileys injectors are superiour to new denso ones as they last longer and are more tolerant of fuel contamination

[Jacko9]

IMHO the DLC coated baileys injectors are superiour to new denso ones as they last longer and are more tolerant of fuel contamination

Genuinely interested why you believe the DLC baileys injectors are better than the new DLC denso injectors? I was under the assumption that they were both quite good now, although the baileys was a remanufactured item, whereas the denso was a new item.

[punjehl]

I was not aware that the new denso wones are dlc coated, if so there is probably no real difference between them than the price the remanufactured units from baileys are cheaper than denso, the only other difference is that the baileys ones use proven and reused injector bodies and come as sets that are picked to more closely match in flow. They all have a correction code to put into the computer because no two injectors have quite the same flow rates at the same supply pressure but baileys pick from a large number of them and divide them into sets that are the closest match. The computer using the correction code can vary the open time to obtain the same quantities from each injector but to do this the computer varies the pulse time length, shorter for a higher flowing injector or longer pulse for a slower flowing one. this is a source of asymetry in engine function that while it can be lived with in a functioning engine and does no damage it is not ideal from an efficency perspective.