The Kavachi (Ultimate) Head Gasket
The Kavachi (Ultimate) Head Gasket
For the last 9 months the coolant in my ZS120 has been gradually loosing its clarity, it turned from crystal clear orange to muddy orange, then recently the expansion tank started to collect a clear, slightly orange, jelly which occasionally gained some dirty colouring, normally after the engine had got hot enough to turn on the fan. I came to the conclusion that the fire rings were probably leaking combustion gasses into the coolant. In the last couple of days and after a bit of hard exercise it became a lot more obvious:
Time for a new head gasket and since I plan to keep the car I decided on XParts Ultimate gasket kit part number ZUA000530 from any Xpart dealer.
The kit contains:
* Head Gasket: LVB90025
* Bolts: 10 x WAM2293SLP
* Oil Rail
The head gasket is now available on its own from one source for those that want to save on cost, but it should not then be considered Ultimate!
This head gasket is often referred to as the N-Series head gasket but it is not, it is the SAIC developed head gasket fitted to the Kavachi engine in the Roewe 750 and since used in the Roewe 550, MG6, MG7 post NAC/SAIC merger and Roewe W5. I think the N-Series head gasket used in the NAC MG TF, first generation MG3 and early MG7 is a version of the Land Rover head gasket and very different to this. Both are MLS head gaskets unlike the original K-Series single layer elastomer bead head gasket, but there are obvious differences.
It is a 6 layer MLS head gasket with no shims and has elastomer coatings on most surfaces but the main difference between this head gasket and any other K-Series head gasket is the lack of any fire rings or crimped end plates which could be a huge advantage if, as in my case, that is where your problem lies!
I wont go into how to change a head gasket, MGJohn has some excellent threads on that, eg: http://ift.tt/1nutbZq
There is also a guide to fitting the XPart Ultimate Kit: http://ift.tt/1nutcfW
I will just cover the things I didnt find in those sources and include a few photos
Inspection
The oil is OK, no mayonnaise there! Valves and spark plugs look good too.
Fire rings
At first sight there is nothing obvious wrong but after a bit of a clean-up and under the microscope:
I assume that the silver line is the part of the head which is always in contact with the head gasket fire ring and the grey area either side shows the amount of movement as the gasket/head/bolts expand and contract with temperature. The head is aluminium alloy and the gasket fire rings are stainless steel so there will be some movement and the steel long bolts will change the pressure on the fire rings as they expand with temperature. There is coolant on the right of the silver line and combustion chamber on the left.
The black spots are indentations in the head and have corresponding bumps in the stainless steel fire ring. Some of the indents appear to go the full width of the silver ring which I assume is the cause of the leak. There are no obvious indents away from the fire rings and none are wider than the silver ring so presumably these have been created by the fire rings and have grown over time to the point that the seal now doesnt seal when things get a little hotter than normal and the bumps and indents dont quite line up due to different rates of expansion of the different metals.
The Kavachi gasket doesnt have stainless steel fire rings and end plates, instead it uses a layer of elastomer over the entire surface which may give enough flexibility for there to be no movement between the surfaces so maybe this issue doesnt occur? Even if there is movement I guess that elastomer against alloy is less likely to cause damage than steel against alloy.
Apart from the little round indents, the fire rings have not indented the head at all so I assume that there has been no overheating problem and the head has not softened and is fine for reuse.
Waterways
I think the hole on the right is supposed to be the same size that the hole on the left would be if it was properly cleaned out. This is obviously not good for coolant flow through the head! This should have been spotted at the factory!
Inlet manifold
That blockage is left over from the moulding and plastic welding of the inlet manifold pieces, air would flow into the cylinders a lot more smoothly if it wasnt there!
There is a mark next to the coolant passage where the plastic of the inlet manifold has rubbed against the alloy head. If the plastic is touching the head then the inlet manifold gasket wont be compressed against the head as firmly as it should be which could potentially cause a leak.
Fixing it
I started with peening of the fire ring area. The idea is to squash the surface of the head so that any cavities both on and below the surface get filled up and disappear, each hit with the hammer will move some alloy material either into a cavity or sideways which will push the neighbouring material upwards, the next hit should squash that material back down so that overall, apart from filling all the cavities you end up with a surface that is the same level as the original, the only differences being a complete lack of cavities and the surface being a little rougher than it was.
This is the end result under the microscope after the head had been polished with 1200 grit sandpaper. The white spots are cavities which were not visible until the head was polished and still only show up with favourable lighting. Presumably it was some of these that the original head gasket enlarged to form the indents, there is the odd one that is almost big enough to span the original fire ring but none are big enough to span the seal on the new gasket. Apart from being quite small they appear to be very thin and dont go down into the head, Im not sure why they would be flat. The main thing to notice in this image is the absence of any cavities along the curve of the fire ring, the peening has removed them giving the head gasket a perfect surface to seal against. Also note the scale of the image, the cavities are not noticeable without magnification and when sanded with courser grit were not noticeable even with magnification. The factory cutter marks were many times larger and deeper.
This image was taken part way through removal of the peening marks. There are still faint traces of the peening marks on the fire ring area and also traces of the factory cutter marks indicating that the peening marks were no deeper than the factory cutter marks! Actually that is not quite true, the peening marks are slightly deeper but not much. The remaining factory cutter marks are there because for some reason that area of the head is slightly lower than the rest. Presumably the factory cutter cut it flat so it must have sunk slightly in this area. I dont understand why since this is the inlet, cooler side of the head and the low part is a fairly small area which doesnt extend as far as the inlet manifold face and isnt even centred on the head bolt hole although it does run up to it. There is a similar area in the diagonally opposite corner:
In this corner, diagonally opposite the last image, the lowered area is somewhat bigger and slightly deeper and extends as far as the fire ring, although it still doesnt reach the exhaust manifold face. That coolant passageway is the one that was nearly blocked in the earlier image and which I have now opened out but the surface around that is OK. The problem is centred on an indent (diagonal line) caused by the crimped stainless steel end plate of the gasket which has pressed the surface of the head inwards. It has not worn the surface as the factory cutter marks are also pressed into the indent. The original head gasket is the same thickness here as everywhere else, so again I cant see any cause for this, maybe the two diagonally opposite corner indents are due to twisting forces in the engine under load? But you would still expect them to extend to the front/back face of the head and they dont
The diagonal indent is a problem because the Kavachi gasket has a coolant seal that runs exactly along the indent. Unlike the original gasket, it allows coolant around the outside of the oilways which surround the locating dowels at the ends of the head. The original gasket keeps the coolant to the inside of the oilways so it never gets near this area and there was no need for a seal here. As a result of the diagonal indent I had to remove nearly twice the material I would otherwise have needed to remove. If you are fitting a Kavachi gasket to a head without a full skim then check for this indent carefully, otherwise you may end up with coolant weeping from the corner of the block when it warms up.
I have no specification for how smooth this gasket wants the head to be but most MLS gasket manufacturers specify a much smoother surface than for non-MLS gaskets. The specification is normally for the surface to be somewhere between the smoothness of 800 grit sandpaper and 2500 grit sandpaper with smoother being better but it shouldnt go below the minimum as the gasket needs to be able to get some grip on the surface. You shouldnt sand it with 2500 grit sandpaper, that would produce too fine a surface, just try to match the smoothness of the paper. I went as far as sanding with 1200 grit sandpaper. There are still obvious sanding marks, much finer than the factory cutter marks, but probably within specification while I suspect the factory cutter was nowhere near. One reason for going as far as 1200 grit is that you can then use it as a mirror if you look from a really low angle and the reflection of straight lines such as the edge of buildings shows up any imperfection in the flatness of the head, however small the imperfection. I couldnt see or measure any imperfection in the flatness of the head.
All 16 coolant holes opened up to their full extent.
Exhaust manifold surface cleaned up, flatted and factory cutter marks removed.
Same for the inlet manifold surface, and also flat the surface of the inlet manifold itself so that it will sit flat on the head and allow the gasket to work properly remember the indent mark next to the coolant way. Then clean out the excess plastic in the inlet manifold runners.
And clean the sealing surface of the sump which needs to come off to fit the new oil rail.
Thats the new improved oil rail in place.
Ready to fit the head, two blocks of wood tied on so that it can be plonked down on top of the locating dowels, the dowels will indent the wood instead of the head. Next time I will also fit the inlet manifold at this point! Also note that diagonal sanding marks are still visible, its not as smooth as the previous mirror image made it look, need something for the elastomer to grip.
New gasket in place on the block, note the coolant way around the outside of the oilways, an area which is sealed off on the original gasket.
New water pump came with a cast alloy impeller instead of the original pressed steel, much nicer. Also changed the thermostat while there was easy access to it.
So that is how you lock the power steering belt tensioner open so much easier to fit when its locked!
All back together and ready for testing
Radiator
In addition to opening out the coolant ways in the head, Ive also modified the radiator cooling. Ive never liked the fact that it goes all the way to 105 degrees C before turning on the fan when in normal use the coolant temperature seen on the Scangauge never goes above 95. If the fan turned on at 95 then it would never reach 100 and so the coolant would never build up pressure and without pressure, even if you do have a small leak the coolant will not squirt out, in fact if any comes out it will form a vacuum which will hold the rest in. So I added a variable speed motor controller to drive the fan and connected a second coolant temperature sensor to it. Now the fan starts to turn when under bonnet air temperature seen on the AIT (air intake temperature) sensor goes above 30 and speeds up with increasing temperature so the coolant never reaches 100, in fact Ive not seen it go above 95. The fan doesnt do much until the engine gets hotter than normal, and then stops it getting any hotter. If youre stuck in a traffic jam then the engine coolant temperature tends to gradually drop down to 88 and stay there instead of rise to 105 and then suddenly drop to 88 before rising again. This system has the advantage that you can then block off the lower grill and rely on the fan to bring air through the radiator when needed. Blocking the lower grill significantly improves the aerodynamics, both reducing drag and reducing lift at higher speeds. It is very noticeable that when turning a sharp corner at 70mph that it sticks to the road like you were only doing 40 while previously the steering tended to lose its accuracy and feel at higher speeds.
Testing
I didnt have time for a proper test so for the first drive I took it for its Easter holidays, starting by filling the tank with Tesco Unleaded, then from South Devon up the MG5 past Gloucester, across to Hereford and up through Mid Wales to Rhyader, then up the Elan Valley and over the Aberystwyth mountain road, then back via Tregaron, over the Cambrian mountains via the Abergwesyn Mountain Road, across the Brecon Beacons and down to the M4 via Merthyr Tydfil then over the Severn bridge and back down the M5. Then I refilled the tank with just 35.69 litres after 408 miles. That works out at 52.0 miles per gallon (629 miles per full tank) so I guess it must be working OK. Would probably have got better MPG if it hadnt been raining for half the return journey, and the Abergwesyn Mountain Road is definitely not ideal for high MPG! Still, its not bad for a car that will do 0-60 in about 7.4 seconds.
The engine sounds a bit different like the difference you might imagine but probably not expect between a metal fire ring and an elastomer seal! Runs very nicely too.
An acceleration test through the rev range in 3rd gear showed it performing just about as well as the best run Id had before the change despite being done on a full tank of normal Tesco unleaded instead of ¾ tank of Tesco Momentum. Maybe removing the rubbish from the inlet manifold did help a little, although it was in the midrange that it beat the previous best.
The only problem is that it sounds like there is a loose cable or pipe at certain rpms, but only one so still much better than when it left the factory and that will soon be sorted.
One last thing
If you buy an ultimate gasket then get the magnifying glass out and check that it is indeed ultimate before you fit it.
I sent the first one back due to this:
That was not obvious without magnification. I have read of a similar issue on another, only found after installation due to it leaking!
For the last 9 months the coolant in my ZS120 has been gradually loosing its clarity, it turned from crystal clear orange to muddy orange, then recently the expansion tank started to collect a clear, slightly orange, jelly which occasionally gained some dirty colouring, normally after the engine had got hot enough to turn on the fan. I came to the conclusion that the fire rings were probably leaking combustion gasses into the coolant. In the last couple of days and after a bit of hard exercise it became a lot more obvious:
Time for a new head gasket and since I plan to keep the car I decided on XParts Ultimate gasket kit part number ZUA000530 from any Xpart dealer.
The kit contains:
* Head Gasket: LVB90025
* Bolts: 10 x WAM2293SLP
* Oil Rail
The head gasket is now available on its own from one source for those that want to save on cost, but it should not then be considered Ultimate!
This head gasket is often referred to as the N-Series head gasket but it is not, it is the SAIC developed head gasket fitted to the Kavachi engine in the Roewe 750 and since used in the Roewe 550, MG6, MG7 post NAC/SAIC merger and Roewe W5. I think the N-Series head gasket used in the NAC MG TF, first generation MG3 and early MG7 is a version of the Land Rover head gasket and very different to this. Both are MLS head gaskets unlike the original K-Series single layer elastomer bead head gasket, but there are obvious differences.
It is a 6 layer MLS head gasket with no shims and has elastomer coatings on most surfaces but the main difference between this head gasket and any other K-Series head gasket is the lack of any fire rings or crimped end plates which could be a huge advantage if, as in my case, that is where your problem lies!
I wont go into how to change a head gasket, MGJohn has some excellent threads on that, eg: http://ift.tt/1nutbZq
There is also a guide to fitting the XPart Ultimate Kit: http://ift.tt/1nutcfW
I will just cover the things I didnt find in those sources and include a few photos
Inspection
The oil is OK, no mayonnaise there! Valves and spark plugs look good too.
Fire rings
At first sight there is nothing obvious wrong but after a bit of a clean-up and under the microscope:
I assume that the silver line is the part of the head which is always in contact with the head gasket fire ring and the grey area either side shows the amount of movement as the gasket/head/bolts expand and contract with temperature. The head is aluminium alloy and the gasket fire rings are stainless steel so there will be some movement and the steel long bolts will change the pressure on the fire rings as they expand with temperature. There is coolant on the right of the silver line and combustion chamber on the left.
The black spots are indentations in the head and have corresponding bumps in the stainless steel fire ring. Some of the indents appear to go the full width of the silver ring which I assume is the cause of the leak. There are no obvious indents away from the fire rings and none are wider than the silver ring so presumably these have been created by the fire rings and have grown over time to the point that the seal now doesnt seal when things get a little hotter than normal and the bumps and indents dont quite line up due to different rates of expansion of the different metals.
The Kavachi gasket doesnt have stainless steel fire rings and end plates, instead it uses a layer of elastomer over the entire surface which may give enough flexibility for there to be no movement between the surfaces so maybe this issue doesnt occur? Even if there is movement I guess that elastomer against alloy is less likely to cause damage than steel against alloy.
Apart from the little round indents, the fire rings have not indented the head at all so I assume that there has been no overheating problem and the head has not softened and is fine for reuse.
Waterways
I think the hole on the right is supposed to be the same size that the hole on the left would be if it was properly cleaned out. This is obviously not good for coolant flow through the head! This should have been spotted at the factory!
Inlet manifold
That blockage is left over from the moulding and plastic welding of the inlet manifold pieces, air would flow into the cylinders a lot more smoothly if it wasnt there!
There is a mark next to the coolant passage where the plastic of the inlet manifold has rubbed against the alloy head. If the plastic is touching the head then the inlet manifold gasket wont be compressed against the head as firmly as it should be which could potentially cause a leak.
Fixing it
I started with peening of the fire ring area. The idea is to squash the surface of the head so that any cavities both on and below the surface get filled up and disappear, each hit with the hammer will move some alloy material either into a cavity or sideways which will push the neighbouring material upwards, the next hit should squash that material back down so that overall, apart from filling all the cavities you end up with a surface that is the same level as the original, the only differences being a complete lack of cavities and the surface being a little rougher than it was.
This is the end result under the microscope after the head had been polished with 1200 grit sandpaper. The white spots are cavities which were not visible until the head was polished and still only show up with favourable lighting. Presumably it was some of these that the original head gasket enlarged to form the indents, there is the odd one that is almost big enough to span the original fire ring but none are big enough to span the seal on the new gasket. Apart from being quite small they appear to be very thin and dont go down into the head, Im not sure why they would be flat. The main thing to notice in this image is the absence of any cavities along the curve of the fire ring, the peening has removed them giving the head gasket a perfect surface to seal against. Also note the scale of the image, the cavities are not noticeable without magnification and when sanded with courser grit were not noticeable even with magnification. The factory cutter marks were many times larger and deeper.
This image was taken part way through removal of the peening marks. There are still faint traces of the peening marks on the fire ring area and also traces of the factory cutter marks indicating that the peening marks were no deeper than the factory cutter marks! Actually that is not quite true, the peening marks are slightly deeper but not much. The remaining factory cutter marks are there because for some reason that area of the head is slightly lower than the rest. Presumably the factory cutter cut it flat so it must have sunk slightly in this area. I dont understand why since this is the inlet, cooler side of the head and the low part is a fairly small area which doesnt extend as far as the inlet manifold face and isnt even centred on the head bolt hole although it does run up to it. There is a similar area in the diagonally opposite corner:
In this corner, diagonally opposite the last image, the lowered area is somewhat bigger and slightly deeper and extends as far as the fire ring, although it still doesnt reach the exhaust manifold face. That coolant passageway is the one that was nearly blocked in the earlier image and which I have now opened out but the surface around that is OK. The problem is centred on an indent (diagonal line) caused by the crimped stainless steel end plate of the gasket which has pressed the surface of the head inwards. It has not worn the surface as the factory cutter marks are also pressed into the indent. The original head gasket is the same thickness here as everywhere else, so again I cant see any cause for this, maybe the two diagonally opposite corner indents are due to twisting forces in the engine under load? But you would still expect them to extend to the front/back face of the head and they dont
The diagonal indent is a problem because the Kavachi gasket has a coolant seal that runs exactly along the indent. Unlike the original gasket, it allows coolant around the outside of the oilways which surround the locating dowels at the ends of the head. The original gasket keeps the coolant to the inside of the oilways so it never gets near this area and there was no need for a seal here. As a result of the diagonal indent I had to remove nearly twice the material I would otherwise have needed to remove. If you are fitting a Kavachi gasket to a head without a full skim then check for this indent carefully, otherwise you may end up with coolant weeping from the corner of the block when it warms up.
I have no specification for how smooth this gasket wants the head to be but most MLS gasket manufacturers specify a much smoother surface than for non-MLS gaskets. The specification is normally for the surface to be somewhere between the smoothness of 800 grit sandpaper and 2500 grit sandpaper with smoother being better but it shouldnt go below the minimum as the gasket needs to be able to get some grip on the surface. You shouldnt sand it with 2500 grit sandpaper, that would produce too fine a surface, just try to match the smoothness of the paper. I went as far as sanding with 1200 grit sandpaper. There are still obvious sanding marks, much finer than the factory cutter marks, but probably within specification while I suspect the factory cutter was nowhere near. One reason for going as far as 1200 grit is that you can then use it as a mirror if you look from a really low angle and the reflection of straight lines such as the edge of buildings shows up any imperfection in the flatness of the head, however small the imperfection. I couldnt see or measure any imperfection in the flatness of the head.
All 16 coolant holes opened up to their full extent.
Exhaust manifold surface cleaned up, flatted and factory cutter marks removed.
Same for the inlet manifold surface, and also flat the surface of the inlet manifold itself so that it will sit flat on the head and allow the gasket to work properly remember the indent mark next to the coolant way. Then clean out the excess plastic in the inlet manifold runners.
And clean the sealing surface of the sump which needs to come off to fit the new oil rail.
Thats the new improved oil rail in place.
Ready to fit the head, two blocks of wood tied on so that it can be plonked down on top of the locating dowels, the dowels will indent the wood instead of the head. Next time I will also fit the inlet manifold at this point! Also note that diagonal sanding marks are still visible, its not as smooth as the previous mirror image made it look, need something for the elastomer to grip.
New gasket in place on the block, note the coolant way around the outside of the oilways, an area which is sealed off on the original gasket.
New water pump came with a cast alloy impeller instead of the original pressed steel, much nicer. Also changed the thermostat while there was easy access to it.
So that is how you lock the power steering belt tensioner open so much easier to fit when its locked!
All back together and ready for testing
Radiator
In addition to opening out the coolant ways in the head, Ive also modified the radiator cooling. Ive never liked the fact that it goes all the way to 105 degrees C before turning on the fan when in normal use the coolant temperature seen on the Scangauge never goes above 95. If the fan turned on at 95 then it would never reach 100 and so the coolant would never build up pressure and without pressure, even if you do have a small leak the coolant will not squirt out, in fact if any comes out it will form a vacuum which will hold the rest in. So I added a variable speed motor controller to drive the fan and connected a second coolant temperature sensor to it. Now the fan starts to turn when under bonnet air temperature seen on the AIT (air intake temperature) sensor goes above 30 and speeds up with increasing temperature so the coolant never reaches 100, in fact Ive not seen it go above 95. The fan doesnt do much until the engine gets hotter than normal, and then stops it getting any hotter. If youre stuck in a traffic jam then the engine coolant temperature tends to gradually drop down to 88 and stay there instead of rise to 105 and then suddenly drop to 88 before rising again. This system has the advantage that you can then block off the lower grill and rely on the fan to bring air through the radiator when needed. Blocking the lower grill significantly improves the aerodynamics, both reducing drag and reducing lift at higher speeds. It is very noticeable that when turning a sharp corner at 70mph that it sticks to the road like you were only doing 40 while previously the steering tended to lose its accuracy and feel at higher speeds.
Testing
I didnt have time for a proper test so for the first drive I took it for its Easter holidays, starting by filling the tank with Tesco Unleaded, then from South Devon up the MG5 past Gloucester, across to Hereford and up through Mid Wales to Rhyader, then up the Elan Valley and over the Aberystwyth mountain road, then back via Tregaron, over the Cambrian mountains via the Abergwesyn Mountain Road, across the Brecon Beacons and down to the M4 via Merthyr Tydfil then over the Severn bridge and back down the M5. Then I refilled the tank with just 35.69 litres after 408 miles. That works out at 52.0 miles per gallon (629 miles per full tank) so I guess it must be working OK. Would probably have got better MPG if it hadnt been raining for half the return journey, and the Abergwesyn Mountain Road is definitely not ideal for high MPG! Still, its not bad for a car that will do 0-60 in about 7.4 seconds.
Quote:
The Mountain Road from Abergwesyn to Tregaron, over the Devil's Staircase, has for seventy years or more been one of the most celebrated trials and rally sections in Britain, of which a major hazard was along the triple fords at Llannerch-yrfa, where the road crosses and recrosses the infant Irfon. one of the most epic driving roads in the country. Hell, Id even go as far as saying it could be amongst the best in Europe - http://ift.tt/1fBb1xf ) |
The engine sounds a bit different like the difference you might imagine but probably not expect between a metal fire ring and an elastomer seal! Runs very nicely too.
An acceleration test through the rev range in 3rd gear showed it performing just about as well as the best run Id had before the change despite being done on a full tank of normal Tesco unleaded instead of ¾ tank of Tesco Momentum. Maybe removing the rubbish from the inlet manifold did help a little, although it was in the midrange that it beat the previous best.
The only problem is that it sounds like there is a loose cable or pipe at certain rpms, but only one so still much better than when it left the factory and that will soon be sorted.
One last thing
If you buy an ultimate gasket then get the magnifying glass out and check that it is indeed ultimate before you fit it.
I sent the first one back due to this:
That was not obvious without magnification. I have read of a similar issue on another, only found after installation due to it leaking!
from the last news http://ift.tt/1nutfbw
via IFTTT
Libellés : IFTTT, the last news
publié par vb @ 14:30
0 commentaires
0 commentaires:
Enregistrer un commentaire
Abonnement Publier les commentaires [Atom]
<< Accueil