Does anybody have any advice on how to find an elusive crack in an aluminium cylinder head? When at operating temperature with the injectors out and looking directly into cylinder number 1 you can see a tiny drop of water falling into the cylinder approx every minute on running the starter. The head has 4 valves per cylinder and we (mechanics, engineer and pressure tester) believe the crack is in the walls around a valve. The engine has been pressure tested at 50 psi with heat and the crack could not be found. A Dye test has been carried out without success. Their are no suitable X Ray facilities local to me. Apparently it is hard to X Ray aluminium anyway. I had the valves removed for the dye test (but not the pressure test since it is not required). I am at a loss with how to find the crack. The pressure tester ran the test all day and said that he has only had one cylinder head defeat him before. Any advice? (An Alfa cylinder head, not Ferrari, and this engine has blown its head gasket 4 times in 4 years)
I think we'll need some more info to try to narrow it down: 1) What model of Alfa is it? 2) You mention you pulled the injectors out, but you see water droplets into the "cylinder". Didn't you mean to say the intake throats? 3) Has the head been tested separately, or was it tested while on the block, still in the car? If the intake plenum has any coolant passageways, especially running near the intake throats, there could be a split in it's gasket, or a crack in the casting near that area. Is it possible to take photos of this?
Not 100 % certain, but the independent engineer I engaged can't see how the water would enter the cylinder in the way it is doing so if it was the block.
1) It is the last of the 166. It is 2 litre four cylinder bought new in July 06. It has showing this fault since approx 18 months old. 2) I may have got what was being said to me wrong. It is the hole directly above the cylinder we were looking down. 3) The head was removed and tested by itself submerged. I will put some photos up of the cylinder head with the valves removed. Thanks so far.
To me, that would be the sparkplug hole. It could very well be a crack between the spark plug hole to either the exhaust ports, or the intake ports. Please do. If the head was submerged in a hot tank, try doing it in cold water.
The red and White marks are from the dye test. I had cylinder 4 tested with that method as well because the tester saw some oxidation (?) marks on the head above cylinder 4. But the engineer and mechanics are convinced cylinder 1 is where the problem lies. Image Unavailable, Please Login Image Unavailable, Please Login Image Unavailable, Please Login Image Unavailable, Please Login Image Unavailable, Please Login Image Unavailable, Please Login Image Unavailable, Please Login Image Unavailable, Please Login Image Unavailable, Please Login
You said you can see a drop falling onto the piston (or into the cylinder when running on the starter. How about "dropping" a camera (bore scope) thru the spark plug hole pointing at the head's valve area and see better where the H20 is coming from? Could that be done?
I was thinking about having tested again. The independent engineer was impressed with the testing machine used though. Can I ask why you suggest doing it cold when the drop only appears when the car is warm?
Funny. I have such a 166 as my daily transport. Very good car, really satisfied with it - sorry to hear about your troubles, but don't give up on it just yet, it's worth the effort to fix it. It could well be the crack is under one of the valve seats and since the aluminum expands more than the steel, that it only opens up when the aluminum is really hot. What were the symptoms you had/have with this car? Just coolant usage, or also pressurisation of the cooling system, water or steam from the exhaust, poor starting, etc? Tell us a little more, that also helps in determining where to look for issues. I'm a bit of an Alfa fan and I can say the Twin Sparks are not known for head gasket issues. Are Alfa's rare where you live? If I was this much in doubt as you are about the condition of a cylinder head I'd find a good used one for a few hundred, but I live in Europe where they are quite plentiful. Four headgaskets aren't exactly cheap or fun to replace. Just putting it back together again without exchanging anything except the headgasket for the fifth time would be well.... illogical.
Has the head been pressure tested? Tony Pimental at AP Cylinder Head has been pressure checking Alfa heads for me for years (mostly old rotten stuff from the 50's). He also builds all the heads for Centerline Alfa and has all the right fixtures for pressure testing Alfa heads. Pressure test the coolant passages will show you exactly where the coolant is coming from.
same engine as my sons Spider - very smooth & gutsy- nice engine. In Oz they have a reputation of being pretty bulletproof, so long as the cambelt is changed regularly & don't get run out of oil. Anyway some suggestions: X rays are useless for this problem, you have to know the orientation of the crack to align the source & plate to get an image. are you sure that the guys doing the dye pen inspection know how? - there seems to be an awful lot of red dye around. If they cleaned the dye off properly, you should only see the developer (the white powder) with very little red showing, except at the defect. Other NDT methods are higher resolution dye pen (Zyglo for eg) or ultrasound, if you can find someone who really knows their stuff. Someone mentioned cracking UNDER the valve seat - I'd also have a look inside the ports, the casting can be quite thin there (moved cores etc). The obvious place to look is between the valve seats, valve seat/plug(s) and any other adjacent changes of casting thickness. Know anyone who builds racing Alfa engines?
How much is a replacement? A lot of times cracks are not weldable without a ton of work and it might not be worth it.
They are good quality cars. However, while any brand including those most associated with quality can inadvertently sell a car with a serious fault, the difference is that they are much more likely to help solve the customers problem. This is not the case with Alfa Romeo, and I bought the car knowing this full well. It was bought for its quality, its charismatic looks and feel though it is not a great drivers car, and took a risk that I wouldn't have to rely on the warranty for a serious fault. I believe the car has a manufacturing fault in the cylinder head. There are a few other things it could be but I have believed for a long time it is the cylinder head. Yes, we could only see the water drop when the car was hot and it could be there. However, the engineer and pressure tester are confident that the pressure could not be contained under test if the fault was there. (It passes the test) I have had the valves removed from cylinder 1 since that test but they don't think that matters in terms of retesting. The file of papers is 3" thick. I would check the oil weekly as these cars use a bit and I would always glance at the coolant. After about 18 mths it started using a small amount of coolant. It was taken to the dealer who said it was normal. I have never owned a car that used coolant significantly. Over the next few months coolant use increased dramatically. I am not the daily driver but it was never allowed to overheat. It got to the stage where it was using approx a litre and half of coolant per week. We got Alfa UK involved and they agreed to take it in. Prior to it leaving our local non franchised garage looked at it and found the visual water leak in cylinder one and no leak in the cooling system. . The car was taken to the franchised dealer who had the cylinder head pressure tested and machine shop tested. No fault was found so the car was reassembled with a new head gasket. They claimed they found a fault in a cooling system pipe and charged us to repair that. The car was picked up and delivered home. We discovered it would not rev beyond 4000 rpm. On complaint the service manager stated in writing that although they road tested the car after their work they had no comparison to make as 166's are very rare and therefore found no fault in that the car would not rev. (That is really what he wrote). We had that adjusted and solved. After approx 4 months the car started using ever increasing amounts of coolant again. It overheated one evening in freezing conditions in an airport carpark after idiling 10 minutes. I happened to be driving it and noticed the gauge rise about 15 degrees and shut the engine off before it went into the red. We have been watching that gauge more than the driving gauges for 2 1/2 years. Alfa had been kept aware of the increasing coolant use. After waiting until 3 in the morning the car was taken away on a lorry. The Alfa dealer said that the fault was electrical in relation to the fan and carried out a "repair" which they charged us for. They said there was no engine fault despite the coolant usage. This made me cross and we carefully documented everything. After a short while the car was using up to 2 litres of coolant a week. It was taken in to the Alfa dealer (arranged via Alfa UK again) and the same repair carried out. Again they said that there was no fault in the cylinder head and it was re used with a new gasket. They said that it re-occurred because we had allowed an electrical fault in the fan to allow the car to slightly overheat regularly. The temp gauge never strayed from the centre line other than the airport episode and the coolant was checked day and night with 5 bottles in the car at all times. Alfa UK paid for the repair. On return I had the cooling system checked by my local mechanic. No faults found. After a similar coolant use problem happened again we went through the whole thing again. I purchased a new cylinder head, head gasket, timing belt kit, water pump etc and arranged to have the car repaired myself outside of Alfa Romeo. We told them that we would attempt to recover our losses from Alfa UK later. They asked to be allowed to repair the car depsite our no confidence in the dealer. We foolishly let them. This time Alfa UK employed an independent engineer at our insistence, who could find no fault in the cylinder head. He said that the coolant being used was too weak a mixture and this likely resulted in permanent slight overheating as evidenced by small "witness" marks in the head gasket. I resented this since I attended to the coolant guzzling car extremely carefully, while until it used very large amounts the dealer said it was normal and would not be convinced otherwise. The car was repaired as before without using my parts and without the new cylinder head. After another short period the car started using large amounts of coolant again .......... We took it to the Alfa dealer who said that there was no fault in the engine and that coolant was being lost in the interior heating system. The owner of the dealership stated that the car was to be taken off his premises and he would not deal with it again. The warranty only had months to run. We took it to our garage who could NOT find not this fault and inspected the rest of the cooling system with no faults being found. We told Alfa UK this. Over a short period the coolant use increased to about 2 litres a week and all of a sudden started producing huge white clouds of smoke from the exhaust. We took it to our garage who said the head gasket had failed. They reproduced the cylinder leak in cylinder number 1 as they had each time and which the Alfa dealer has never accepted. Alfa UK said they would deal with the dealer. For over a month we tried to get an answer while the car was stored at my local mechanic. I employed an independent engineer at this stage. Eventually Alfa UK said that the dealer would not repair the car because the dealer communicated in writing that the owner and several employees had smelt coolant loss through the heater when it was brought to them months before. We told Alfa UK this was unacceptable and asked them to stand over the repair. They wrote a one line response stating that because their dealer had diagnosed the fault which had not been repaired they were not responsible for it. I had the car repaired myself using the new cylinder head and new gasket, timing belt kit and water pump. It has since traveled 1500 miles without coolant use. I am a car enthusiast and I have owned Italian cars including a Maserati and this car was bought for a family member on the basis that it was an nice Italian car that could be used daily without too much risk of trouble. I am in the UK. I was an Alfa fan and was willing to buy a car that I knew was not as good as the equivalent BMW's basically because of its charisma. Though I accept that this car has a one off fault and is not representative of average quality, I think the standard of assistance By Alfa Italy and UK was representative. With Alfa you will not find systematic standards throughout the dealer network. There are good dealers and bad. The good ones are good because they are capable business people or enthusiasts. The bad ones are just bad and you cannot expect Alfa UK or Alfa Italy to help you. Fiat have been talking about improving the customer experience for decades but they have never achieved it. This car has been off road approx 1/4 of its life. You are quite correct to say that the experience hasn't been fun. At this stage I really want to know what caused all this. Thank You for your illogical comment. This is exactly what we tried to tell Alfa whose dealer always insisted that there was yet another fault in the external cooling system.
Yes, the head was pressure tested each time, and the engineer and mechanics do not believe the block could be at fault given the location of the leak into cylinder 1.
I don't think I will have much luck finding an Ultrasound test here but I will make some calls. I had never seen the dye test done before. Maybe I should pressure test it again with the valves out though I am told it doesn't make any difference. No specialists here unfortunately. Thank You
The new parts were just under £1000. The labour was just over £1000. The engineer was £500. The car was off road 3 months on the last problem. Cylinder head was around £600 by itself new from Alfa Romeo.
Its already fitted. Those costs are for the last repair I paid for. All 3 previous "repairs" were by Alfa. I want to find out the real cause.
Sounds like no one tested the block but said it was OK. There are precise specs for the block and deck height and lip of liner height relative to deck height. If this is not right then every new head you put on this car will fail. With the history of the 2nd head it makes sense to pull the motor and have the bottom end inspected properly. You can't do much more than an eyeball with the block in the car.
You are correct the block was never tested. Unfortunately I have just had the new cylinder head installed and the engine built up so it would be difficult to test it. My Mechanic (not the Alfa dealer) was 100% sure that it was the cylinder head and the engineer couldn't see how the water would appear where it did if the block was at fault. The New Cylinder head has only been fitted this time i.e. the 4th time. The original head was used in all other repairs. I hope it is not the block and I really don't want to take the engine out after all this. Are Cylinder Head pressure tests the best way to test a head for a crack? The tester said he has had only other cylinder head he could not get to bottom of and the only reason he accepted their was a fault in it was because it was eventually replaced and the problem stopped. That was a Peugeot. I am at a loss with how to proceed but given the disaster this car has been I would like to get to the bottom of it.
Well I would have said to try magnafluxing the head, but if it's aluminum it won't work. The metal has to be magnetic.
When you have weird problems you start with basics. [ QUOTE=davidoloan;139934451]You are correct the block was never tested. Unfortunately I have just had the new cylinder head installed and the engine built up so it would be difficult to test it. My Mechanic (not the Alfa dealer) was 100% sure that it was the cylinder head and the engineer couldn't see how the water would appear where it did if the block was at fault. The New Cylinder head has only been fitted this time i.e. the 4th time. The original head was used in all other repairs. I hope it is not the block and I really don't want to take the engine out after all this. Are Cylinder Head pressure tests the best way to test a head for a crack? The tester said he has had only other cylinder head he could not get to bottom of and the only reason he accepted their was a fault in it was because it was eventually replaced and the problem stopped. That was a Peugeot. I am at a loss with how to proceed but given the disaster this car has been I would like to get to the bottom of it.[/QUOTE]
David, Having chased problems like this on BMW's - let me give you a plan. First, clean ALL the dye from that head - and I mean - ALL. Verify that w/ UV light. Next, reassemble the engine and place ONE type of dye in the oil, and another type in the coolant - when I say "type" I mean "color". Run the motor - see your drop - and stop. Disassemble - use UV light to follow back to source. If you have a borescope and a similar UV light, you don't have to disassemble, just scope each runner and then the bore itself, looking for the leak. Jim PS: Of course, most coolant already HAS a UV dye in it, so your tech should have been able to follow the leak back to it's source. The crack should be filled with dye.