Rover 200 BRM - 1.8 K-Series turbo project
Discussion
It depends. as the metal is deformed it got rather hot. The alloy anneals past 400C and turns soft. Once the thin section is in some distance from the liner it heats up even further as now two sides are exposed to combustion gases.
400°C is well within the range of fuel auto ignition temperatures. From there it takes a single pre ignition event to brake things.
Any knock (autoignition) induced at these hot spots would not help either as it tends to heat things further.
400°C is well within the range of fuel auto ignition temperatures. From there it takes a single pre ignition event to brake things.
Any knock (autoignition) induced at these hot spots would not help either as it tends to heat things further.
Hugh Jarse said:
Dont know much about these things but doesnt that really thin bit on the piston crown edge get red hot and cause preignition?
No. By the time aluminium is red it's liquid and of no use in an engine, it's impossible to have aluminium glowing red, be solid and causing a problem. It's the high temps in the combustion chamber which have caused the piston to melt, it hasn't been enough to melt the main body, just picked on the thinnest easy bits. It's a bit like lighting a fire too near some trees, as it gets hotter and hotter it burns the thin extremities (the leaves) first, if it is left it will do the twigs, then the branches and finally the trunk.
Cast iron heads, valves and carbon can glow red though.
Ive said:
Points of improvement of this piston could be machining of the thin section of the valve cut-out and leaving more metal on the ramp to the piston crown between the valve pockets. It will add a few cc of volume, but make the pistons much stronger.
Given that the OEM Rover K Turbo 1.8l in the Rover 75 uses 2mm shorter rods to reduce compression, that factor is important, but not instrumental.
This leaves us with a OEM design that has a quench/squish distance of more than 3mm. This will have close to zero effect, especially on quench. Still Lee managed to extract 300HP out of that configuration.
I know every little helps. I like the idea of adding this strengthening to add extra squish at the same time.
Thing is that in a Rover K with OEM rod length, there is very little space to lose as the pin is already very high up.
Adding 2cc of metal in this area would already increase CR by 0.3.
This is the first know failure of such a piston. Running things at full boost on a dyno at constant rpm over extended periods of time introduces tremendous amounts of heat. This is only comparable to hard track use or a top speed run on the Autobahn.
I once cracked a OEM liner due to pre ignition. There was nothing more than a slight hesitation before it blew all its coolant out of the exhaust at 6000 rpm. No audible misfire or anything. Happened because of fuel pressure fluctuations. The OEM pump was at its limits at that point.
My current engine and also stuballs' engine use Westwood liners that are much stronger. So strong that in this case it might well be that a piston gave in first.
That the piston got very hot is indicated by the bend piston crown pieces and the lack of carbon deposits IMHO.
There is always something that goes first. Happens. :-)
Do you consider a 1.9l block from VGK racing or from Scholar?
there are three other ways to get this going again:
QEDs shorted rods with forged low compression Accralite pistons (http://qedmotorsport.co.uk/qed-shop/rover-k-series/pistons/low-compression-pistons-suitable-for-supercharged-engines) Maxspeeding can also make you 2mm shorter rods to take these pistons.
Rover turbo K rods with their trophy pistons
Fit Omega forged pistons with a machined 1mm dish.
For a road car I also recommend the methanol water injection. Keeps critical components cool :-)
This leaves us with a OEM design that has a quench/squish distance of more than 3mm. This will have close to zero effect, especially on quench. Still Lee managed to extract 300HP out of that configuration.
I know every little helps. I like the idea of adding this strengthening to add extra squish at the same time.
Thing is that in a Rover K with OEM rod length, there is very little space to lose as the pin is already very high up.
Adding 2cc of metal in this area would already increase CR by 0.3.
This is the first know failure of such a piston. Running things at full boost on a dyno at constant rpm over extended periods of time introduces tremendous amounts of heat. This is only comparable to hard track use or a top speed run on the Autobahn.
I once cracked a OEM liner due to pre ignition. There was nothing more than a slight hesitation before it blew all its coolant out of the exhaust at 6000 rpm. No audible misfire or anything. Happened because of fuel pressure fluctuations. The OEM pump was at its limits at that point.
My current engine and also stuballs' engine use Westwood liners that are much stronger. So strong that in this case it might well be that a piston gave in first.
That the piston got very hot is indicated by the bend piston crown pieces and the lack of carbon deposits IMHO.
There is always something that goes first. Happens. :-)
Do you consider a 1.9l block from VGK racing or from Scholar?
there are three other ways to get this going again:
QEDs shorted rods with forged low compression Accralite pistons (http://qedmotorsport.co.uk/qed-shop/rover-k-series/pistons/low-compression-pistons-suitable-for-supercharged-engines) Maxspeeding can also make you 2mm shorter rods to take these pistons.
Rover turbo K rods with their trophy pistons
Fit Omega forged pistons with a machined 1mm dish.
For a road car I also recommend the methanol water injection. Keeps critical components cool :-)
IVE thanks for all your input here. This is fascinating!
I'm going to check the fuel system thoroughly as that's an obvious potential culprit for suddenly running lean and melting things. I had recently switched to a walbro 255 and my fuel pressure needle did develop a very small fluctuation after at idle (a few psi). It didn't fluctuate under load. Mapper said it was normal and due to slight fluctuation in vacuum at idle. Now I'm not ruling anything out. The rest of the fuel system is stock (return-based fuel rail, pressure regulator, lines) so maybe that's room for improvement there. I'll text it all first though.
I haven't decided where to go for the 1.9 conversion yet. Scholar is an option. The supplier of my new 82mm Arias has a tame engineer who he uses to supply as a package but they are more than twice the cost of scholar.
I'm also going to start looking into water/methanol injection. Currently I know nothing about it!
I'm going to check the fuel system thoroughly as that's an obvious potential culprit for suddenly running lean and melting things. I had recently switched to a walbro 255 and my fuel pressure needle did develop a very small fluctuation after at idle (a few psi). It didn't fluctuate under load. Mapper said it was normal and due to slight fluctuation in vacuum at idle. Now I'm not ruling anything out. The rest of the fuel system is stock (return-based fuel rail, pressure regulator, lines) so maybe that's room for improvement there. I'll text it all first though.
I haven't decided where to go for the 1.9 conversion yet. Scholar is an option. The supplier of my new 82mm Arias has a tame engineer who he uses to supply as a package but they are more than twice the cost of scholar.
I'm also going to start looking into water/methanol injection. Currently I know nothing about it!
Edited by Stuballs on Monday 25th April 12:18
Ive said:
Given that the OEM Rover K Turbo 1.8l in the Rover 75 uses 2mm shorter rods to reduce compression, that factor is important, but not instrumental.
This leaves us with a OEM design that has a quench/squish distance of more than 3mm. This will have close to zero effect, especially on quench. Still Lee managed to extract 300HP out of that configuration.
I know every little helps. I like the idea of adding this strengthening to add extra squish at the same time.
Thing is that in a Rover K with OEM rod length, there is very little space to lose as the pin is already very high up.
Adding 2cc of metal in this area would already increase CR by 0.3.
For a road car I also recommend the methanol water injection. Keeps critical components cool :-)
I think the fact this lash up of an engine produced only 160bhp speaks volumes about it, it's no surprise Rover met its demise some time after using tricks like that.This leaves us with a OEM design that has a quench/squish distance of more than 3mm. This will have close to zero effect, especially on quench. Still Lee managed to extract 300HP out of that configuration.
I know every little helps. I like the idea of adding this strengthening to add extra squish at the same time.
Thing is that in a Rover K with OEM rod length, there is very little space to lose as the pin is already very high up.
Adding 2cc of metal in this area would already increase CR by 0.3.
For a road car I also recommend the methanol water injection. Keeps critical components cool :-)
Correctly shaped squish pads with minimum clearance are very important, that will be one of the reasons you have to use the band-aid of WI to make an engine run properly at even these low power levels. The burn has to be made to happen as much towards the centre and within the bowl as possible otherwise detonation will occur much more easily, that is fact and is 'instrumental'.
Below is a properly re-designed piston, (OE on left), not only does it have matching squish pads and a completely different crown design to manipulate the mixture to where it should be, but you can also see the lowered 'ears' as I call them, ring pack pushed further up and if you think your pin is close to the bottom ring think again
More static CR than yours, over 500bhp from 2 ltrs on pump fuel, no VVTI, no WI, no other crutches and they don't break.
I know of squish, quench, the pads, turbulence vs. flame propagation speed etc. and would have installed appropriate parts right away. :-) I just don't have them installed as they were not available on the market and I had to work with other boundary conditions.
The reason I "need" WI is not that much related to the piston design IMHO, but to an other reason specific to my build.
In order to protect my cylinder head against indentation from a MLS gasket, a known issue, I have installed a 1mm copper shim between head gasket and head. Normally it comes with a 83mm bore size.
Due to me being overly smart, I had the bore reduced to 80mm, the nominal cylinder bore size. This way forces from the sealing edge of the head gasket would distribute uniformly. That part works perfectly fine BTW.
I was also ill informed that the N-series gasket, it does not indent heads as it has a wider fire ring, it not suitable for FI applications. This was a misunderstanding. I even have the gasket sitting in a box. I works perfectly fine, just not in box, but in a engine.
As usual, you need more to to go wrong before you have issues. I did not chamfer the lower edge of the 80mm bore. This has the result that there is a sharp edge kind of on the outer perimeter of my combustion chamber. These sharp edges create a BIG round hot spot and reflect sound waves. Edges induce knock. Wit the 83mm bore this edge is near the fire ring in a cool zone and less exposed. So "bad" choices made. Thing is that happens. No Evo 9 & STI mega power catalog for forced induction Rover Ks unfortunalty.
Those pistons run in a chap's turbo Elise with 280HP at my boost level without any WI. OK, he does not track the car.
Despite this my engine runs well and lasts so far. I am not the one ending his trackday prematurely due to technical issues.
Solution: remove head, pull shim, install N-series gasket, retime cams. It is a day worth of work.
Why not done yet: Car is plenty fast as is, I have little spare time available due to family and I am not to keen of popping it open again.
Yes, I could have 30HP more. But who cares at this level? 250HP or 280HP for a 1650 pound Rover S1 Elise?
I recently had my first ever 4th gear wheel spin in the wet hitting peak torque running a Torsen Typ B LSD.
Spa is scary as is, 30HP more won't improve that.
After all I am no engineer, nor a profession engine builder. This is just a hobby and I do most things DIY excluding machining work, e.g. Head porting. I left this to the excellent Roger Fabry.
No tuner has touched this car it's a system in the 12 yrs I own it. I had a "tuner" prepare parts for me resulting in a catastrophic failure as gudgeon pins moved out of the rods. They were not refitted properly. Ever since it is all DIY and all my choices. This puts me in the fortunate position to exacly know who I have to call and blame shall thing fail. ;-)
That is how one learns. But it is also time consuming. I share my experiences and this is how things progress.
With every iteration those boosted Ks get better. :-)
The reason I "need" WI is not that much related to the piston design IMHO, but to an other reason specific to my build.
In order to protect my cylinder head against indentation from a MLS gasket, a known issue, I have installed a 1mm copper shim between head gasket and head. Normally it comes with a 83mm bore size.
Due to me being overly smart, I had the bore reduced to 80mm, the nominal cylinder bore size. This way forces from the sealing edge of the head gasket would distribute uniformly. That part works perfectly fine BTW.
I was also ill informed that the N-series gasket, it does not indent heads as it has a wider fire ring, it not suitable for FI applications. This was a misunderstanding. I even have the gasket sitting in a box. I works perfectly fine, just not in box, but in a engine.
As usual, you need more to to go wrong before you have issues. I did not chamfer the lower edge of the 80mm bore. This has the result that there is a sharp edge kind of on the outer perimeter of my combustion chamber. These sharp edges create a BIG round hot spot and reflect sound waves. Edges induce knock. Wit the 83mm bore this edge is near the fire ring in a cool zone and less exposed. So "bad" choices made. Thing is that happens. No Evo 9 & STI mega power catalog for forced induction Rover Ks unfortunalty.
Those pistons run in a chap's turbo Elise with 280HP at my boost level without any WI. OK, he does not track the car.
Despite this my engine runs well and lasts so far. I am not the one ending his trackday prematurely due to technical issues.
Solution: remove head, pull shim, install N-series gasket, retime cams. It is a day worth of work.
Why not done yet: Car is plenty fast as is, I have little spare time available due to family and I am not to keen of popping it open again.
Yes, I could have 30HP more. But who cares at this level? 250HP or 280HP for a 1650 pound Rover S1 Elise?
I recently had my first ever 4th gear wheel spin in the wet hitting peak torque running a Torsen Typ B LSD.
Spa is scary as is, 30HP more won't improve that.
After all I am no engineer, nor a profession engine builder. This is just a hobby and I do most things DIY excluding machining work, e.g. Head porting. I left this to the excellent Roger Fabry.
No tuner has touched this car it's a system in the 12 yrs I own it. I had a "tuner" prepare parts for me resulting in a catastrophic failure as gudgeon pins moved out of the rods. They were not refitted properly. Ever since it is all DIY and all my choices. This puts me in the fortunate position to exacly know who I have to call and blame shall thing fail. ;-)
That is how one learns. But it is also time consuming. I share my experiences and this is how things progress.
With every iteration those boosted Ks get better. :-)
Edited by Ive on Monday 25th April 16:44
Ive said:
I know of squish, quench, the pads, turbulence vs. flame propagation speed etc. and would have installed appropriate parts right away. :-) I just don't have them installed as they were not available on the market and I had to work with other boundary conditions.
The reason I "need" WI is not that much related to the piston design IMHO, but to an other reason specific to my build.
In order to protect my cylinder head against indentation from a MLS gasket, a known issue, I have installed a 1mm copper shim between head gasket and head. Normally it comes with a 83mm bore size.
Due to me being overly smart, I had the bore reduced to 80mm, the nominal cylinder bore size. This way forces from the sealing edge of the head gasket would distribute uniformly. That part works perfectly fine BTW.
I was also ill informed that the N-series gasket, it does not indent heads as it has a wider fire ring, it not suitable for FI applications. This was a misunderstanding. I even have the gasket sitting in a box. I works perfectly fine, just not in box, but in a engine.
As usual, you need more to to go wrong before you have issues. I did not chamfer the lower edge of the 80mm bore. This has the result that there is a sharp edge kind of on the outer perimeter of my combustion chamber. These sharp edges create a BIG round hot spot and reflect sound waves. Edges induce knock. Wit the 83mm bore this edge is near the fire ring in a cool zone and less exposed. So "bad" choices made. Thing is that happens. No Evo 9 & STI mega power catalog for forced induction Rover Ks unfortunalty.
Those pistons run in a chap's turbo Elise with 280HP at my boost level without any WI. OK, he does not track the car.
Despite this my engine runs well and lasts so far. I am not the one ending his trackday prematurely due to technical issues.
Solution: remove head, pull shim, install N-series gasket, retime cams. It is a day worth of work.
Why not done yet: Car is plenty fast as is, I have little spare time available due to family and I am not to keen of popping it open again.
Yes, I could have 30HP more. But who cares at this level? 250HP or 280HP for a 1650 pound Rover S1 Elise?
I recently had my first ever 4th gear wheel spin in the wet hitting peak torque running a Torsen Typ B LSD.
Spa is scary as is, 30HP more won't improve that.
After all I am no engineer, nor a profession engine builder. This is just a hobby and I do most things DIY excluding machining work, e.g. Head porting. I left this to the excellent Roger Fabry.
No tuner has touched this car it's a system in the 12 yrs I own it. I had a "tuner" prepare parts for me resulting in a catastrophic failure as gudgeon pins moved out of the rods. They were not refitted properly. Ever since it is all DIY and all my choices. This puts me in the fortunate position to exacly know who I have to call and blame shall thing fail. ;-)
That is how one learns. But it is also time consuming. I share my experiences and this is how things progress.
With every iteration those boosted Ks get better. :-)
What is your squish clearance?The reason I "need" WI is not that much related to the piston design IMHO, but to an other reason specific to my build.
In order to protect my cylinder head against indentation from a MLS gasket, a known issue, I have installed a 1mm copper shim between head gasket and head. Normally it comes with a 83mm bore size.
Due to me being overly smart, I had the bore reduced to 80mm, the nominal cylinder bore size. This way forces from the sealing edge of the head gasket would distribute uniformly. That part works perfectly fine BTW.
I was also ill informed that the N-series gasket, it does not indent heads as it has a wider fire ring, it not suitable for FI applications. This was a misunderstanding. I even have the gasket sitting in a box. I works perfectly fine, just not in box, but in a engine.
As usual, you need more to to go wrong before you have issues. I did not chamfer the lower edge of the 80mm bore. This has the result that there is a sharp edge kind of on the outer perimeter of my combustion chamber. These sharp edges create a BIG round hot spot and reflect sound waves. Edges induce knock. Wit the 83mm bore this edge is near the fire ring in a cool zone and less exposed. So "bad" choices made. Thing is that happens. No Evo 9 & STI mega power catalog for forced induction Rover Ks unfortunalty.
Those pistons run in a chap's turbo Elise with 280HP at my boost level without any WI. OK, he does not track the car.
Despite this my engine runs well and lasts so far. I am not the one ending his trackday prematurely due to technical issues.
Solution: remove head, pull shim, install N-series gasket, retime cams. It is a day worth of work.
Why not done yet: Car is plenty fast as is, I have little spare time available due to family and I am not to keen of popping it open again.
Yes, I could have 30HP more. But who cares at this level? 250HP or 280HP for a 1650 pound Rover S1 Elise?
I recently had my first ever 4th gear wheel spin in the wet hitting peak torque running a Torsen Typ B LSD.
Spa is scary as is, 30HP more won't improve that.
After all I am no engineer, nor a profession engine builder. This is just a hobby and I do most things DIY excluding machining work, e.g. Head porting. I left this to the excellent Roger Fabry.
No tuner has touched this car it's a system in the 12 yrs I own it. I had a "tuner" prepare parts for me resulting in a catastrophic failure as gudgeon pins moved out of the rods. They were not refitted properly. Ever since it is all DIY and all my choices. This puts me in the fortunate position to exacly know who I have to call and blame shall thing fail. ;-)
That is how one learns. But it is also time consuming. I share my experiences and this is how things progress.
With every iteration those boosted Ks get better. :-)
Edited by Ive on Monday 25th April 16:44
Ive said:
I know of squish, quench, the pads, turbulence vs. flame propagation speed etc. and would have installed appropriate parts right away. :-) I just don't have them installed as they were not available on the market and I had to work with other boundary conditions.
If you knew these things you would have had 1mm squish clearance.Ive said:
The reason I "need" WI is not that much related to the piston design IMHO, but to an other reason specific to my build.
In order to protect my cylinder head against indentation from a MLS gasket, a known issue, I have installed a 1mm copper shim between head gasket and head. Normally it comes with a 83mm bore size.
Due to me being overly smart, I had the bore reduced to 80mm, the nominal cylinder bore size. This way forces from the sealing edge of the head gasket would distribute uniformly. That part works perfectly fine BTW.
I was also ill informed that the N-series gasket, it does not indent heads as it has a wider fire ring, it not suitable for FI applications. This was a misunderstanding. I even have the gasket sitting in a box. I works perfectly fine, just not in box, but in a engine.
As usual, you need more to to go wrong before you have issues. I did not chamfer the lower edge of the 80mm bore. This has the result that there is a sharp edge kind of on the outer perimeter of my combustion chamber. These sharp edges create a BIG round hot spot and reflect sound waves. Edges induce knock. Wit the 83mm bore this edge is near the fire ring in a cool zone and less exposed. So "bad" choices made.
Why did you have the bore reduced? So it's stronger presumably? Why is it an issue that a gasket indents into the head? Some gaskets are purposely made this way although I think outdated given the more modern MLS we use now. I wonder if anyone has taken off a straight head, done some work and put it back on with the same type of gasket and it been ok. I guess you don't like to take risks with this gasket sensitive engine.In order to protect my cylinder head against indentation from a MLS gasket, a known issue, I have installed a 1mm copper shim between head gasket and head. Normally it comes with a 83mm bore size.
Due to me being overly smart, I had the bore reduced to 80mm, the nominal cylinder bore size. This way forces from the sealing edge of the head gasket would distribute uniformly. That part works perfectly fine BTW.
I was also ill informed that the N-series gasket, it does not indent heads as it has a wider fire ring, it not suitable for FI applications. This was a misunderstanding. I even have the gasket sitting in a box. I works perfectly fine, just not in box, but in a engine.
As usual, you need more to to go wrong before you have issues. I did not chamfer the lower edge of the 80mm bore. This has the result that there is a sharp edge kind of on the outer perimeter of my combustion chamber. These sharp edges create a BIG round hot spot and reflect sound waves. Edges induce knock. Wit the 83mm bore this edge is near the fire ring in a cool zone and less exposed. So "bad" choices made.
Whether you are seeing any hot spot at the top inside edge of the liner is pure speculation, I wouldn't be at all surprised if there is none as the combustion process shouldn't be happening there. If there is any it's possibly because your squish is so large it's allowing it to happen. When the piston comes up to TDC on the firing stroke the mixture has (or should have) nowhere to go, it gets pushed out swirling (as it should) for a good gradual and complete burn. What you guys are doing by using this huge gap (and pistons with less squish pad) is completely ruining this effect, you are being left with stagnant areas which combust irregularly and contribute to detonation.
Consider what 'Quench' actually means in (olde) English, i'll let you Google it if you don't know.
Ive said:
Solution: remove head, pull shim, install N-series gasket, retime cams. It is a day worth of work.
Why not done yet: Car is plenty fast as is, I have little spare time available due to family and I am not to keen of popping it open again.
Yes, I could have 30HP more. But who cares at this level? 250HP or 280HP for a 1650 pound Rover S1 Elise?
I recently had my first ever 4th gear wheel spin in the wet hitting peak torque running a Torsen Typ B LSD.
Spa is scary as is, 30HP more won't improve that.
Why modify at all? It will be much safer as standard. Why not done yet: Car is plenty fast as is, I have little spare time available due to family and I am not to keen of popping it open again.
Yes, I could have 30HP more. But who cares at this level? 250HP or 280HP for a 1650 pound Rover S1 Elise?
I recently had my first ever 4th gear wheel spin in the wet hitting peak torque running a Torsen Typ B LSD.
Spa is scary as is, 30HP more won't improve that.
The gains in getting it right are much more than max BHP:No need to buy WI (it's also something else which can go wrong).
More torque (especially at lower rpm)
More reliability
Greater MPG
An engine which pulls a lot better and feels much crisper from lower rpms
Ive said:
After all I am no engineer, nor a profession engine builder.
You're doing a good job of trying to teach Granny to suck eggs 
Ive said:
This is just a hobby and I do most things DIY excluding machining work, e.g. Head porting. I left this to the excellent Roger Fabry.
No tuner has touched this car it's a system in the 12 yrs I own it. I had a "tuner" prepare parts for me resulting in a catastrophic failure as gudgeon pins moved out of the rods. They were not refitted properly. Ever since it is all DIY and all my choices. This puts me in the fortunate position to exacly know who I have to call and blame shall thing fail. ;-)
That is how one learns. But it is also time consuming. I share my experiences and this is how things progress.
With every iteration those boosted Ks get better. :-)
Well you've got plenty of improving to keep you busy for a few years No tuner has touched this car it's a system in the 12 yrs I own it. I had a "tuner" prepare parts for me resulting in a catastrophic failure as gudgeon pins moved out of the rods. They were not refitted properly. Ever since it is all DIY and all my choices. This puts me in the fortunate position to exacly know who I have to call and blame shall thing fail. ;-)
That is how one learns. But it is also time consuming. I share my experiences and this is how things progress.
With every iteration those boosted Ks get better. :-)
Edited by Evoluzione on Tuesday 26th April 10:50
If I was to offer you Turbo K series guys some pointers on what to research it would be:
Squish and mixture motion.
Dynamic CR.
Never underestimate the importance of these on a successful engine.
There is too much talk of static CR, from one running engine to another of a different spec (apart from the same static CR) it's irrelevant.
"I'm running 9:1 on pump fuel and it's detting its tits off!"
"I am too and have no issues!"
Hmmm, wonder why
Consider what it means: 'Static'.
How many engines create power when they are static?
I don't know many....
Squish and mixture motion.
Dynamic CR.
Never underestimate the importance of these on a successful engine.
There is too much talk of static CR, from one running engine to another of a different spec (apart from the same static CR) it's irrelevant.
"I'm running 9:1 on pump fuel and it's detting its tits off!"
"I am too and have no issues!"
Hmmm, wonder why
Consider what it means: 'Static'.
How many engines create power when they are static?
I don't know many....
Edited by Evoluzione on Tuesday 26th April 11:42
That is correct.
Cylinder head indentation is a Rover K specific issue.
The cylinder head is manufactured from a heat treated alloy called LM25 hardened to more than 110 Brinnel.
IThe block is made the same way.
Should the cylinder head for whatever reason overheat and be exposed to temperatures above 200C, the alloy anneals and losens its hardness. The fire ring, a very narrow design, something on the order of half a millimeter, can then deform the metal and lose its seal. As gases enter the cylinder head cooling paths, things get worse. It is a run away process.
This is one of the failure modes of the head gasket of a Rover K.
There have been a few things done to prevent this. The latest measure is a new style of head gasket, actually very similar to most head gaskets today. It employs a wide flat seal, and not a wire wrapped in a folded aluminium film.
It is combined with increased stiffness long bolts and a revised oil rail. Overall this design reduces the pressure the head surface sees and does therefore not rely on the high hardness to maintain a seal.
Overall there is very little public knowledge on how to build a reliable 250 to 300HP FI Rover K.
We are getting closer, but we are still not 100% there.
Turbotechnics failed a decade ago with their 260HP supercharged VHPD versions despite major efforts.
This is why most folks just convert to Audi 1.8l, Audi 2l, Duratec or Honda K20A2 engines if they want to much go beyond 200HP. These engines, especially the Honda and there Audi offer proven paths for 250 to 350HP without to much fuzz.
Past 350HP they need additional work. Thing is that 350HP in a S1 Elise makes a friggin fast car.
We talk about 500HP per ton here in a short wheelbase mid engined RWD car without any electronic aids. You won't need anymore. Still folks have build 500HP+ monster Elises or Exiges weighting 800kg. It starts getting scary and not that enjoyable at some point.
I personally have installed a electric water pump system in line with the mechanical pump to maintain reasonable coolant flow at low rpm and idle. Remember, this is a mid engine installation with a radiator at the front. Coolant flow almost comes to halt at idle indicated by rising coolant temperatures and with the fan runing cold water exiting the radiator.
With a EWP this is much improved. This phenomenon is only ever critical at low revs or idle after a sustained high load.
Now, why do I not have 1mm squish? Very easy. I would need a custom piston with a crown that protrudes the liner as the head gasket itself is 1.5mm thick. I thought the latest piston as used in my and Stuballs engine would offer that solution as a COTS product. I does not. Now I also know that.
The next step would be indeed yet an other custom piston. I am currently not bothered with this as it works OK as is. Shall I need new pistons I would be more than interested in a proper design that takes advantage of the squish pads on the head. I am fully with you on that. But again, if you dig a little deeper into the Rover K, you'll find that this is not trivial to do either due to the rod length. Now you can use shorter rods, but then the rod to stroke ratio gets even worse than it already is.
Next issue is that Cr get rather high should that squish be used to its full extend. You canot take out much more material from the chamber. The usual solution is a deeper dish. Again, we have a problem with the deck height and rod length.
Ok, now extend the block and use longer rods. VGKracing is currently trying this. Maybe it works, maybe it has other issues.
You may now get a first idea why things are as these are.
Knock usually starts at and near the edges of a combustion chamber.
Yes, WI can fail. I had issues, but I also employ three fail safety features.
What happens if it stops properly working under full boost. I have tried it. I get misfires and my knock control system pulls 10 degrees of timing. Fueling stays Ok as I run closed loop and fuel is automatically added to maintain the AFR. Misfires lead to more fuel being added than necessary. Not a bad thing in this moment.
Nothing brakes. :-). Would I install a system with significant flow in a customer car (I don't do anything commercially)? Probalby no unless they really take care about the system and know what to do if there are issues. If they just press on and later tell me:" Yeah, there were a few misfires etc, but I really needed to maintain top speed on the autobahn for a little longer and then came the flames from under the hood....". WI is not idiot proof.
Additionally, shall the WI rail pressure drop, I use a pulse width modulated injection valve fed by a pressured line, the ECU automatically switches to a non-WI map. No big deal either.
Is the shim causing problems at that edge? Maybe. We'll find out shall the time come to pull the head and take it off.
Maybe this fall. Was it a good idea. On one side yes, on an other, no. That is experimentation.
Let's focus again on Stuballs engine here. I'll benefit from that, too.
I can only dance with the girls I have. If that ideal piston does not exist, I am more or less out of options. This partcular piston came pretty close. The Accralite low compression piston in combination with shorter rods is the only really other forged COTS option.
If it would have been easy, I would have done it. And afterwards everybody is smarter.
BTW, what does that teach granny sucking eggs thing mean?
Cylinder head indentation is a Rover K specific issue.
The cylinder head is manufactured from a heat treated alloy called LM25 hardened to more than 110 Brinnel.
IThe block is made the same way.
Should the cylinder head for whatever reason overheat and be exposed to temperatures above 200C, the alloy anneals and losens its hardness. The fire ring, a very narrow design, something on the order of half a millimeter, can then deform the metal and lose its seal. As gases enter the cylinder head cooling paths, things get worse. It is a run away process.
This is one of the failure modes of the head gasket of a Rover K.
There have been a few things done to prevent this. The latest measure is a new style of head gasket, actually very similar to most head gaskets today. It employs a wide flat seal, and not a wire wrapped in a folded aluminium film.
It is combined with increased stiffness long bolts and a revised oil rail. Overall this design reduces the pressure the head surface sees and does therefore not rely on the high hardness to maintain a seal.
Overall there is very little public knowledge on how to build a reliable 250 to 300HP FI Rover K.
We are getting closer, but we are still not 100% there.
Turbotechnics failed a decade ago with their 260HP supercharged VHPD versions despite major efforts.
This is why most folks just convert to Audi 1.8l, Audi 2l, Duratec or Honda K20A2 engines if they want to much go beyond 200HP. These engines, especially the Honda and there Audi offer proven paths for 250 to 350HP without to much fuzz.
Past 350HP they need additional work. Thing is that 350HP in a S1 Elise makes a friggin fast car.
We talk about 500HP per ton here in a short wheelbase mid engined RWD car without any electronic aids. You won't need anymore. Still folks have build 500HP+ monster Elises or Exiges weighting 800kg. It starts getting scary and not that enjoyable at some point.
I personally have installed a electric water pump system in line with the mechanical pump to maintain reasonable coolant flow at low rpm and idle. Remember, this is a mid engine installation with a radiator at the front. Coolant flow almost comes to halt at idle indicated by rising coolant temperatures and with the fan runing cold water exiting the radiator.
With a EWP this is much improved. This phenomenon is only ever critical at low revs or idle after a sustained high load.
Now, why do I not have 1mm squish? Very easy. I would need a custom piston with a crown that protrudes the liner as the head gasket itself is 1.5mm thick. I thought the latest piston as used in my and Stuballs engine would offer that solution as a COTS product. I does not. Now I also know that.
The next step would be indeed yet an other custom piston. I am currently not bothered with this as it works OK as is. Shall I need new pistons I would be more than interested in a proper design that takes advantage of the squish pads on the head. I am fully with you on that. But again, if you dig a little deeper into the Rover K, you'll find that this is not trivial to do either due to the rod length. Now you can use shorter rods, but then the rod to stroke ratio gets even worse than it already is.
Next issue is that Cr get rather high should that squish be used to its full extend. You canot take out much more material from the chamber. The usual solution is a deeper dish. Again, we have a problem with the deck height and rod length.
Ok, now extend the block and use longer rods. VGKracing is currently trying this. Maybe it works, maybe it has other issues.
You may now get a first idea why things are as these are.
Knock usually starts at and near the edges of a combustion chamber.
Yes, WI can fail. I had issues, but I also employ three fail safety features.
What happens if it stops properly working under full boost. I have tried it. I get misfires and my knock control system pulls 10 degrees of timing. Fueling stays Ok as I run closed loop and fuel is automatically added to maintain the AFR. Misfires lead to more fuel being added than necessary. Not a bad thing in this moment.
Nothing brakes. :-). Would I install a system with significant flow in a customer car (I don't do anything commercially)? Probalby no unless they really take care about the system and know what to do if there are issues. If they just press on and later tell me:" Yeah, there were a few misfires etc, but I really needed to maintain top speed on the autobahn for a little longer and then came the flames from under the hood....". WI is not idiot proof.
Additionally, shall the WI rail pressure drop, I use a pulse width modulated injection valve fed by a pressured line, the ECU automatically switches to a non-WI map. No big deal either.
Is the shim causing problems at that edge? Maybe. We'll find out shall the time come to pull the head and take it off.
Maybe this fall. Was it a good idea. On one side yes, on an other, no. That is experimentation.
Let's focus again on Stuballs engine here. I'll benefit from that, too.
I can only dance with the girls I have. If that ideal piston does not exist, I am more or less out of options. This partcular piston came pretty close. The Accralite low compression piston in combination with shorter rods is the only really other forged COTS option.
If it would have been easy, I would have done it. And afterwards everybody is smarter.
BTW, what does that teach granny sucking eggs thing mean?
Edited by Ive on Tuesday 26th April 11:47
http://www.pistonheads.com/gassing/topic.asp?h=0&a...
Relevant? Worth thinking about, not my area of expertise, but you never know it might help long term
Relevant? Worth thinking about, not my area of expertise, but you never know it might help long term
Engine back out
And pistons removed for a closer look.
No signs of det damage anywhere else on the piston (and there also was nothing on the head or valves) Other than where that front edge had melted off. You can see where the inlet side has started to fail and crack.
The lower compression ring snapped when I tried to move it as it's wedged tight in the groove, presumably where that side of the piston deformed from heat. The liner is scored badly all around.
The other pistons look completely fine apart from cyl 2 that has a very slight bending in on the inlet side. No visible det though.
So, it looks like something went wrong specifically with cyl 1 causing it to overheat. It's got to be either the fuel system or ring related.
Rings: Did that ring bind in its groove first or as a result of overheating caused by something else? If it happened first this would have caused it to overheat and the gap to close up, further increasing friction and resulting in what happened.
Fuel: seeing as only cyl 1 was damaged, injector would be the likely culprit. Although cyl 1 is the last on the fuel rail before the fpr. I was never happy with the walbro install. Whilst it worked, it felt like a bit of a bodge up. Today's job is to see if I can rig something up to test the system as it sits on the car. Although even if it all works, I'll overhaul the system for the rebuild.
So what happens now?
- Arias pistons (82mm) are on order from Grant at gbe. Ceramic coated crowns and moly coated skirts.
- The block will be modified and fit with 82mm liners, increasing capacity to 1.9
Stock image:
- The head will also be modified to make the most of the added diameter and to help lower the cr a bit.
- I'll have to review the cams as I'm not sure they will clear these pistons, especially with the thinner elastomer gasket I'll have to use (the n-series is unsuitable for the 82mm bore).
- Exhaust manifold will be ditched in favour of an equal length tubular header. This may be overkill but having longer equal-length runners should eliminate exhaust reversion and help reduce cylinder temperature. Every little helps. It should also spool the turbo quicker.
- Fuel system will be upgraded. I'm going to go back to a standard in-tank pump and use that to feed a swirl pot and external pump and filter in the boot (in a vented enclosure to keep fumes out) or underneath if it will fit. From there, I'll have my own lines running to and from the stock fuel rail. I sourced an adapter allowing me to fit an-6 lines to the end of the stock rail. I'm also going to add a fuel pressure gauge in the cabin.
Adapter:
The waiting times alone for the pistons and block mean I'm not likely to see her running again until July at the earliest. Hopefully I can be up and running in time for the respray which is still booked for August.
And pistons removed for a closer look.
No signs of det damage anywhere else on the piston (and there also was nothing on the head or valves) Other than where that front edge had melted off. You can see where the inlet side has started to fail and crack.
The lower compression ring snapped when I tried to move it as it's wedged tight in the groove, presumably where that side of the piston deformed from heat. The liner is scored badly all around.
The other pistons look completely fine apart from cyl 2 that has a very slight bending in on the inlet side. No visible det though.
So, it looks like something went wrong specifically with cyl 1 causing it to overheat. It's got to be either the fuel system or ring related.
Rings: Did that ring bind in its groove first or as a result of overheating caused by something else? If it happened first this would have caused it to overheat and the gap to close up, further increasing friction and resulting in what happened.
Fuel: seeing as only cyl 1 was damaged, injector would be the likely culprit. Although cyl 1 is the last on the fuel rail before the fpr. I was never happy with the walbro install. Whilst it worked, it felt like a bit of a bodge up. Today's job is to see if I can rig something up to test the system as it sits on the car. Although even if it all works, I'll overhaul the system for the rebuild.
So what happens now?
- Arias pistons (82mm) are on order from Grant at gbe. Ceramic coated crowns and moly coated skirts.
- The block will be modified and fit with 82mm liners, increasing capacity to 1.9
Stock image:
- The head will also be modified to make the most of the added diameter and to help lower the cr a bit.
- I'll have to review the cams as I'm not sure they will clear these pistons, especially with the thinner elastomer gasket I'll have to use (the n-series is unsuitable for the 82mm bore).
- Exhaust manifold will be ditched in favour of an equal length tubular header. This may be overkill but having longer equal-length runners should eliminate exhaust reversion and help reduce cylinder temperature. Every little helps. It should also spool the turbo quicker.
- Fuel system will be upgraded. I'm going to go back to a standard in-tank pump and use that to feed a swirl pot and external pump and filter in the boot (in a vented enclosure to keep fumes out) or underneath if it will fit. From there, I'll have my own lines running to and from the stock fuel rail. I sourced an adapter allowing me to fit an-6 lines to the end of the stock rail. I'm also going to add a fuel pressure gauge in the cabin.
Adapter:
The waiting times alone for the pistons and block mean I'm not likely to see her running again until July at the earliest. Hopefully I can be up and running in time for the respray which is still booked for August.
The n-series gasket can be machined to accommodate a 82mm bore. On Seloc a chap uses it on his 2l VGK block. He even reused it three times now as he is in the US.
The K's plenum is not that suitable to forced induction as its cross section remains constant. For FI it should drop by 25% per cylinder to maintain equal air velocity before each runner. The first cylinder otherwise gets the most air the the furthest away the least. This gets worse as total airflow, boost or HP increase.
The pictures still make be believe the Pistons overheated, lost strength due to that, started cracking/melting on the thin spots and then the entire bit of the crown fell off. Was there any black carbon on the crowns?
The K's plenum is not that suitable to forced induction as its cross section remains constant. For FI it should drop by 25% per cylinder to maintain equal air velocity before each runner. The first cylinder otherwise gets the most air the the furthest away the least. This gets worse as total airflow, boost or HP increase.
The pictures still make be believe the Pistons overheated, lost strength due to that, started cracking/melting on the thin spots and then the entire bit of the crown fell off. Was there any black carbon on the crowns?
Edited by Ive on Monday 9th May 17:22
Ive said:
The n-series gasket can be machined to accommodate a 82mm bore. On Seloc a chap uses it on his 2l VGK block. He even reused it three times now as he is in the US.
The K's plenum is not that suitable to forced induction as its cross section remains constant. For FI it should drop by 25% per cylinder to maintain equal air velocity before each runner. The first cylinder otherwise gets the most air the the furthest away the least. This gets worse as total airflow, boost or HP increase.
The pictures still make be believe the Pistons overheated, lost strength due to that, started cracking/melting on the thin spots and then the entire bit of the crown fell off. Was there any black carbon on the crowns?
Never really thought about plenum design as an issue. Not really much I can do about that, unfortunately. Unless I have a custom plenum made up. Doable, but is it necessary for sub 300bhp and 12psi?The K's plenum is not that suitable to forced induction as its cross section remains constant. For FI it should drop by 25% per cylinder to maintain equal air velocity before each runner. The first cylinder otherwise gets the most air the the furthest away the least. This gets worse as total airflow, boost or HP increase.
The pictures still make be believe the Pistons overheated, lost strength due to that, started cracking/melting on the thin spots and then the entire bit of the crown fell off. Was there any black carbon on the crowns?
Edited by Ive on Monday 9th May 17:22
Piston did have carbon deposits I just cleaned it off for a better look before I took the pictures.
Having tested my fuel system and found the injectors to be flowing pretty much the same across the board, I'm leaning more toward friction from a ring issue starting off the overheating process. Anything other than a bad injector would have affected all 4 cylinders.
I'm seriously considering a water/meth injection system for the rebuild. But it gets expensive when you want to build in suitable failsafes based on fluid flow. Aem do some decent looking kits.
I'm trying to cover all the bases for the rebuild to stop this happening again, basically.
Regarding soot I was thinking about the piston pictures when they were stil in the engine. As they have already seen a few miles I'd expect them to be covered with a pitch black carbon layer. Yours show metallic blank with a yellowish brown hint.
Did you clean the Pistons before taking the pictures with the Pistons still in the block?
If not and they are as is they were way too hot /andor you were running way to lean at WOT.
Here are my pistons with their carbon deposits with the intake section, the one that broke off on your's, and the opposite side showing bare metal due to the water/meth injection and fuel wash.
I only run 12 to 12.5 to one, so not that rich for a FI engine.
On the comparison shop I posted before of the old high compresion pistos vs the JDM pnes you see a similar colour. These were not cleaned and also used with up to 10 PSI of boost..
your piston uncleaned I expect as still in the cylinder.
Do you have a wideband lambda sensor install with a gauge in view and connected to the ECU?
If not, I'd strongly recommend to do so, especially during development. I you can quickly spot if there is something wrong and if not, you KNOW your fueling is right. STACK makes a nice and very reliable unit with the electronics integrated into the gauge. This is what I and 2 friends use. No calibration need or any other issues.
If the mappers oxygen probe had issues you don't know what the actual lambda/AFR was.
If you run rich enough or water meth, a little air flow difference won't matter that much. Plently of folks have successfully used a VVC plenum to turbocharge or supercharge the K on this boost level.
One thing I have found that helped me was my J&S vampire safeguard knock control system. As it reacts within one revolution on individual cylinders, it prevents heat building due to knock spectacularly well. Then fix those map cells with too much advance and continue. If you run out of fuel, pressure peaks etc, it will react in an instant. The only thing you get is a little hesitation from the retard and some blinking LEDs
Though for the price of this system you might consider getting a ECU with integrated knock control and probalby even integrated wide band control.
Did you clean the Pistons before taking the pictures with the Pistons still in the block?
If not and they are as is they were way too hot /andor you were running way to lean at WOT.
Here are my pistons with their carbon deposits with the intake section, the one that broke off on your's, and the opposite side showing bare metal due to the water/meth injection and fuel wash.
I only run 12 to 12.5 to one, so not that rich for a FI engine.
On the comparison shop I posted before of the old high compresion pistos vs the JDM pnes you see a similar colour. These were not cleaned and also used with up to 10 PSI of boost..
your piston uncleaned I expect as still in the cylinder.
Do you have a wideband lambda sensor install with a gauge in view and connected to the ECU?
If not, I'd strongly recommend to do so, especially during development. I you can quickly spot if there is something wrong and if not, you KNOW your fueling is right. STACK makes a nice and very reliable unit with the electronics integrated into the gauge. This is what I and 2 friends use. No calibration need or any other issues.
If the mappers oxygen probe had issues you don't know what the actual lambda/AFR was.
If you run rich enough or water meth, a little air flow difference won't matter that much. Plently of folks have successfully used a VVC plenum to turbocharge or supercharge the K on this boost level.
One thing I have found that helped me was my J&S vampire safeguard knock control system. As it reacts within one revolution on individual cylinders, it prevents heat building due to knock spectacularly well. Then fix those map cells with too much advance and continue. If you run out of fuel, pressure peaks etc, it will react in an instant. The only thing you get is a little hesitation from the retard and some blinking LEDs
Though for the price of this system you might consider getting a ECU with integrated knock control and probalby even integrated wide band control.
Edited by Ive on Tuesday 10th May 10:36
Edited by Ive on Tuesday 10th May 10:39
Ive said:
Regarding soot I was thinking about the piston pictures when they were stil in the engine. As they have already seen a few miles I'd expect them to be covered with a pitch black carbon layer. Yours show metallic blank with a yellowish brown hint.
Did you clean the Pistons before taking the pictures with the Pistons still in the block?
If not and they are as is they were way too hot /andor you were running way to lean at WOT.
Here are my pistons with their carbon deposits with the intake section, the one that broke off on your's, and the opposite side showing bare metal due to the water/meth injection and fuel wash.
I only run 12 to 12.5 to one, so not that rich for a FI engine.
On the comparison shop I posted before of the old high compresion pistos vs the JDM pnes you see a similar colour. These were not cleaned and also used with up to 10 PSI of boost..
your piston uncleaned I expect as still in the cylinder.
Do you have a wideband lambda sensor install with a gauge in view and connected to the ECU?
If not, I'd strongly recommend to do so, especially during development. I you can quickly spot if there is something wrong and if not, you KNOW your fueling is right. STACK makes a nice and very reliable unit with the electronics integrated into the gauge. This is what I and 2 friends use. No calibration need or any other issues.
If the mappers oxygen probe had issues you don't know what the actual lambda/AFR was.
If you run rich enough or water meth, a little air flow difference won't matter that much. Plently of folks have successfully used a VVC plenum to turbocharge or supercharge the K on this boost level.
One thing I have found that helped me was my J&S vampire safeguard knock control system. As it reacts within one revolution on individual cylinders, it prevents heat building due to knock spectacularly well. Then fix those map cells with too much advance and continue. If you run out of fuel, pressure peaks etc, it will react in an instant. The only thing you get is a little hesitation from the retard and some blinking LEDs
Though for the price of this system you might consider getting a ECU with integrated knock control and probalby even integrated wide band control.
Thanks for that. Bear in mind the engine only had 500 miles on at that stage. And actually I'd cleaned the piston tops at 150 miles when I had to take the head off to overhaul the valve train. So I think there may hasn't been enough time for carbon to build up .Did you clean the Pistons before taking the pictures with the Pistons still in the block?
If not and they are as is they were way too hot /andor you were running way to lean at WOT.
Here are my pistons with their carbon deposits with the intake section, the one that broke off on your's, and the opposite side showing bare metal due to the water/meth injection and fuel wash.
I only run 12 to 12.5 to one, so not that rich for a FI engine.
On the comparison shop I posted before of the old high compresion pistos vs the JDM pnes you see a similar colour. These were not cleaned and also used with up to 10 PSI of boost..
your piston uncleaned I expect as still in the cylinder.
Do you have a wideband lambda sensor install with a gauge in view and connected to the ECU?
If not, I'd strongly recommend to do so, especially during development. I you can quickly spot if there is something wrong and if not, you KNOW your fueling is right. STACK makes a nice and very reliable unit with the electronics integrated into the gauge. This is what I and 2 friends use. No calibration need or any other issues.
If the mappers oxygen probe had issues you don't know what the actual lambda/AFR was.
If you run rich enough or water meth, a little air flow difference won't matter that much. Plently of folks have successfully used a VVC plenum to turbocharge or supercharge the K on this boost level.
One thing I have found that helped me was my J&S vampire safeguard knock control system. As it reacts within one revolution on individual cylinders, it prevents heat building due to knock spectacularly well. Then fix those map cells with too much advance and continue. If you run out of fuel, pressure peaks etc, it will react in an instant. The only thing you get is a little hesitation from the retard and some blinking LEDs
Though for the price of this system you might consider getting a ECU with integrated knock control and probalby even integrated wide band control.
Edited by Ive on Tuesday 10th May 10:36
Edited by Ive on Tuesday 10th May 10:39
I have a plx wideband plumbed info the Emerald ecu. The mapper also used a plx wideband with their own sensor screwed into a spare bung in the downpipe. Both systems advised the same afr. At wot with 12psi afrs were high 11s. The plugs were a chocolate/tan colour and showed no signs of excessive temps. Nor did the exhaust valves. I really wish it was that simple. Currently, I still have not pinned down a definitive cause for the failure.
The Emerald has the ability to run closed loop based on wideband input but I could never find settings that worked that well. I don't think the Emerald has knock control. But I like the sound of the J&S vampire system. I will look into that.
I'm hesitant to throw money at safeguards and preventative measures when I don't know what caused this problem in the first place. But really, what choice do I have? I may have no choice but to rebuild meticulously, upgrading and "future-proofing" things at I go (fuel system/intake/exhaust) and put those safeguards in place so I know I've done everything in my power to prevent a reoccurrence.
IVE really appreciate your input. And everyone else's too.
Gassing Station | Readers' Cars | Top of Page | What's New | My Stuff