What's wrong with MGs ??

[Scarlet Fever]

...still working on it (i've started so i'll finish, maybe it'll come in handy for others).

Covered the first 4 theories in depth now, working on #5, #6 and #7 - won't be long...

[stu@rt]

I think all this talk of HGF is very interesting…. Not from a technical point of view, because clearly it isn't really much of an issue… but I'm thinking about it in terms of the psychology of the modern consumer.

….people these days seem to think about cars the same way as they think about Micro Wave Ovens…. Clean, fast, 100% reliable,….. Instant gratification every time. Cars aren't like this I'm afraid, and they never have been

….I think its partially the fault of the manufacturers (think about modern engines, many are completely enclosed in plastic…. Look at an A class Merc… you can't even find the engine !!), they have in fact sold people on the concept that a car is like a microwave, and you tinker with it at your peril. The problem is of course a microwave has possibly 3 or 4 moving parts… a car has 1000's of parts. A car is a mass of complex engineered components linked to onboard computer systems (remember your TF has x1000 more computer power on board than what the Americans put a man on the moon with !!!!), which then has to be mass produced, mass assembled and bought in at cost to enable almost universal access whilst remaining profitable. Cars at present are cheaper (in real terms) than they have ever been, they are also 10 times more reliable than they were even 10 years ago. I had a Spitfire 15 years ago which had something like 6 grease points which had to be greased every 3,000 miles !!!! The car in my opinion is a miracle of modern industry, no other industry even comes close to being able to achieve this level of mass produced complexity.

….Considering all this it seems only fair that we have to accept some degree of critical failure. Unfortunate when its your car that goes bang, but such is life and such is cars.

[Scarlet Fever]

Head Gasket Failures – why is there a prevalence in MGFs?

First up, a bit of a history lesson…

The K series engine is a thoroughly modern powerplant with a small coolant capacity. It has been designed like this to speed up warm up times and thus deliver exceptional fuel economy and reduce emissions to meet ever-tightening standards. An engine produces most of its’ airborne pollutants during the warm up cycle when the choke is fully open. The K4 engine, being fuel injected doesn’t have a ‘choke’ in the traditional sense, rather the Modular Engine Management System (MEMS) controls the fuel : air ratio making it rich during the start up cycle and backing off the fuel as the engine reaches normal running temperatures. The primary signal that the MEMS uses to determine the temperature of the engine is sited in the cooling system, thus small coolant capacity = faster warm up time = rapidly climbing coolant temperature = MEMS thinks engine is warm sooner = fuel : air mixture is adjusted sooner = less pollutants & greater fuel efficiency.

So, the plus side to this type of engine design is greater fuel economy and less pollutants. What about the down side?

Theory #1 – thermal cycling

The K4 engine is basically a sandwich, it has an alloy block and head with a steel gasket. It is also thin walled (to save weight and increase the power to weight ratio) and uses a revolutionary through bolt construction where there are long bolts that travel top to bottom bolting the entire sandwich together. The bolts have very closely controlled metallurgy so that when they are torqued correctly they stretch in unison with the alloy engine. This is state of the art and even today, there are few engines on the market that are this advanced. However, it does make the engine susceptible to differing temperature gradients, if one part of the engine is thermally expanding / contracting at a different rate to the others and / or the stretch bolts then movement isn’t contained and it will occur at the join where the temperatures differ.

In the case of the F, as the head is coolant cooled and the block is oil cooled, there is potential here for differential thermal expansion – clearly the joint between these two parts needs to be controlled, as does the temperature of both the cooling mediums. So lets look at the engine under normal conditions…

Early K4 engines had steel locating dowels between the head and block, this is how the original engine designers intended it. At this time the 1.8 variant of the K4 engine didn’t exist yet and the engine was primarily used in the Rover 100, 200 and 400 in 1.4 and 1.6 capacities. There were a few incidents of HGF on these cars and then lots of repeat HGFs because the people repairing the engines were damaging the head gasket on the steel dowels! Yep, one of the prevention measures was causing repeat occurrences.

Rover, in their infinite wisdom, decided to cure the problem by replacing the steel dowels with plastic ones, thus all the new 1.8K4 (designed for the MGF and rolled out onto other platforms later) had engines with plastic dowels. What happened next is that BMW took over and overnight cancelled all funding to the MGF project meaning the car was effectively in limbo from a feedback to development point of view for months / years.

Summary

• All early 1.8K4 engines fitted with plastic dowels

• No development funding for MGF, ergo, no development funding for 1.8K4 as it was an MGF only engine back then.

So, potentially we have a problem here with the lack of strength in the connection between the head and block, meaning the stretch bolts are now in the front line to prevent differential expansion and we also have a problem that although will be picked up quickly, will not be solved for months / years due to lack of funding (are you beginning to see why I hate BMW? ).

Now let’s look at the 1.8K4 engine insitu in the MGF…

As you all know, the MGF is mid-engined. What this means is that unlike normal cars, the radiator is a long way from the engine, the engine bay itself gets very little through flow of (cooling) air and the cooling system (plumbing) is pretty complicated by normal car standards. The thermostat is also located within the engine bay and is quite vicious in its’ operation. Here’s what happens as the engine runs: The engine temperature builds at a reasonably constant rate, then the thermostat closes and the MEMS adjusts the fuel : air ratio as it is now up to temperature. Meanwhile the coolant flows from the engine, under the car (where it is cooled) to the radiator (where it is cooled a great deal more), then back under the car (where it is cooled again) to the engine bay. The thermostat, being in the engine bay, doesn’t register the now supercooled coolant waiting to enter the engine and so when the temperature threshold is breached, MEMS opens the thermostat (which is vicious so opens fully) and a huge slug of super cold coolant floods into the head cooling it rapidly (small coolant capacity head remember? All those quick warm up times, low emissions and fuel economy benefits don't seem so good now do they? ) and inducing thermal contraction. The thermostat now registers the head is cold so closes (fully) again allowing the heat to build, and build, and build, before woosh! Another cold slug. So the head suffers a great deal from thermal cycling, and the block, which is oil cooled is a great deal more temperature stable.

You have an expanding and contracting head, sitting on a stable block, with (by now) fractured plastic locating dowels and the only real mechanical fixing is the long bolts which are torqued so that they can stretch.

Now lets look at the gasket…

The gasket is a steel plate with rubber beading glued to the edges (above and below). Once the dowels break, movement in the head is transmitted to the rubber bead which is now being rubbed off of the gasket. It is only a matter of time until the bead separates and the gasket fails.

Theory #2 – electrolytic corrosion

We know the head and block are alloy and the gasket is steel. These are dissimilar metals and over time electrolytic corrosion can take place. The oil prevents this below the gasket, and the coolant above. Early Fs used ethylene glycol based coolants (alcohol), and although the cooling properties of this type of coolant are excellent, as is the life expectancy of the anti-freeze compounds, the chemical inhibitors that prevent electrolytic corrosion break down over time, typically around 15-18 months. Early Fs have 12 month service intervals, and were also mainly bought by enthusiasts who kept it in the garage all week and the car only saw light use of a weekend. Thus 12 months would go by and it is very tempting to skip a service as the car had only done say 6000 miles. This is fatal – half way through the next year and corrosion has started, this can only get worse and cannot be reversed insitu resulting in a weakening of the bond between the rubber bead and the gasket.

Bingo – HGF.

Theory #3 – Air bubbles

We have seen that the MGF’s layout has resulted in a complex cooling system. Bleeding air out of the system takes time and patience and these two commodities are very scarce in a dealership where every job has a time target to be met to avoid costing the dealership money. This is particularly relevant as bleeding the system is done at the end of the coolant change, and if the technician is up against it time wise, or is under pressure to maximise profits, is likely to be done in a cursory manner. A significant number of HGFs occur shortly after a coolant change – the theory is that a poorly bled cooling system results in air bubbles which equals hot spots on the gasket and HGF.

A variation of this is a sheared radiator bleed screw. This is the bleed point on the top right of the radiator and is a plastic bolt. Over tightening the bolt results in the head shearing from the shaft and a loss of clamping pressure on the rubber washer – this equals a small coolant leak which is unlikely to be noticed by the owner (rad is hidden, coolant leaks onto hot rad and evaporates) unless it becomes a major leak. Water out = air in = air bubbles = hot spots = HGF. QED. Bear in mind the only person who ever uses this screw is a service technician bleeding the cooling system…

The formula: Water out = air in = air bubbles = hot spots = HGF is a major player in the HGF debate and will only become more so as the cars age. Corrosion of the radiator and under body coolant pipes (typically by year 6) results in leaks and ‘water out’.

Theory #4 – The radiator fan fuse

The radiator fan only comes on intermittently when demand calls for it (i.e. in traffic on a hot day). It is also a long way from the cabin in its’ own enclosure and very difficult to hear. If you can hear a fan, it is likely to be the engine bay fan (not the rad fan) – in fact the only way I’ve been able to tell it’s on from the cabin is by feeling for vibrations through the pedals. Why is this important? Well, the fuse that controls the fan is rated at 15 amps. The fan is only used rarely and can partially seize – causing a power spike when the MEMS tries to switch it on and this can very easily blow the fuse. A blown fuse stops it coming on and as you, the driver, don’t know when it is on anyway, you don’t realise there is a problem until the car overheats. A severe overheating will blow the head gasket (and do other damage too). Moral, change the rad fan fuse to a 20 amp one – go on, do it now!!!

Theory #5 – Expansion tank cap

The expansion tank cap contains a pressure release valve as a safety feature (to stop the system being over pressurised). The cooling system is pressurised as pressurised fluids have a higher temperature boiling point, so a loss of pressure is likely to result in localised boiling in the expansion tank - causing expansion of the fluid until it overspills. When the coolant cools down (after switch off for instance) it contracts in volume resulting in an empty tank. Empty tank = ‘air in’… QED HGF

The caps are very simple items, being a spring loaded plate on a rubber membrane within a plastic housing, but they are prone to failure over time resulting in less clamping pressure on the membrane, greater fluid expansion and loss of coolant.

Good news is they are really cheap. I carry a spare with me and tend to change mine annually.

Theory #6 – Water pump

It appears the water pump can fail without warning when it gets on a bit. Specialists (Mike Satur, B+G et al) are now changing these as part of the 5 year / 60K service as once failed, no coolant circulation, overheating and HGF.

Theory #7 – High liners

As I started off saying that the K4 engine is revolutionary and highly advanced, one of the features of this engine are the middle hung (damp) cylinder liners. These are steel tubes that the pistons travel up and down inside and have to be totally flush with the top of the block to make a seal on the firing rings on the head gasket. If they are microns too low you will get gases escaping = hot spots = HGF, if they are microns too high you get distortion of the firing ring = gases escaping = hot spots = HGF. The K4 is made in one of the most advanced engine production facilities in the world (to ensure these accurate tolerances), but there have been some reports of high, or dropped liners. Fortunately this is very rare as a cause.

Theory #8 – Inlet Manifold Gasket

This seems to be prevallent on early Fs mainly, and most people will tell you that it only happens on MPi engines (although i'm not convinced, see later). Between the inlet manifold and the head is a gasket (Inlet Manifold Gasket, or IMG), the manifold not only has the inlet throats cast (alloy VVC style plenum) or moulded (plastic MPi style plenum) into it, it also has a coolant pipe to the expansion tank from the head. What has happened in the past is the gasket has failed and allowed coolant to be sucked out of the pipe across the gasket into the intake for cylinder 1 where it is burned and expelled with the exhaust gases. This tends to happen when the engine is cold, as it warms up the gasket expands slightly and seals the gap. The upshot of this is a persistant coolant leak that is very difficult to diagnose and if it is unnoticed then you can empty the expansion tank resulting in 'air in'...

And why am i not convinced? Well, Scarlet, although being an MPi, has a VVC inlet plenum and manifold (including VVC IMG). I personally have experienced a failed VVC IMG and this is why i am not convinced it is an MPi only problem.

---------------------

So, HGF is the end result of a whole host of potential causes in the MGF. It is true that a large number of the above causes apply to any car with a head gasket, but there are also a few additional ones that are K4 specific and a few more that are MGF (and Elise!) specific.

So, we’re screwed right?

Well not really. There are cures and partial cures for most of the above, a lot are snake oil (like drilling the thermostat – at best this will open the thermostat earlier, but doesn’t address the cause which is the long pipe runs). Some are very effective (like steel dowels, uprated gasket (bead is pinned through the gasket as well as being bonded on), remote thermostat (as fitted to TF) and combined oil water coolers). And some are merely service and inspection items (bleed screw, expansion tank cap, 20 amp rad fuse etc).

Additionally, despite all of the above, the K4 engine has proved to be very resilient. Even if you have an early MGF there is still no reason to assume that you will get HGF. Indeed I personally know of 2 high mileage 1996 MPi Fs still with their original owners and no HGF yet, one of them is regularly sprinted on circuits up and down the country and the other does regular trips across Europe touring. Personally, I had HGF on my Feb 1996 F at 91,000 miles – not a bad showing really, I mean most normal cars would be onto their third owner by the time it gets to this kind of mileage.

Additionally, despite the ‘cash freeze’ from BMW, work was going on behind the scenes to solve this. When the money was finally allocated to sort this out, Rover moved swiftly and began a roll out through 2000 / 2001 of engines with the uprated head gasket (beads pinned through gasket in addition to being bonded on) and the steel dowels returned. If you are buying an F and are worried about HGF, then you should aim to buy as new as possible, preferably Model Year 2000 (MY2K) onwards. The improvements continued when MGR took over and the TF also has a remote thermostat and on the 160 model, an oil cooler too. All of these things help with reducing the likelihood of HGF. The other main modification is the switch to Organic Acid Technology (OAT) based coolant. OAT is longer lasting than the ethylene glycol (alcohol) based coolants from a corrosion inhibitor standpoint. This not only has helped eradicate the dissimilar metals issue, but it has also had the added benefit of contributing to the extension of the service interval to 15000 miles (or 12 months) making the car appear more reliable from a PR standpoint (and helping to fight some of the negative press).

The true scale of the problem will never be known by Joe public I suspect, indeed I reckon there are only a very few people who actually have a true handle on it, and they work within MGR. Warrantee companies may also have a fair idea, but they would need to pool their data to get anything reasonably accurate (and then they won’t know about out of warrantee incidents). The best we can do is here:

>> http://www.shame.4mg.com <<

This is an online register of HGF incidents started in the mid 1990’s. It is not representative for a number of reasons, mainly the demographics of the people adding to the database is very small (people suffering HGF who are IT literate and use the internet forums) and also from the other perspective, when you suffer HGF you tend to look around for advice, thus data from online forum users are massively skewed. There is no flip side to this in so far as people who have no problems (and there are tens of thousands of them) don’t tend to post on forums.

What the data, flawed though it is, does show are general trends. The analysis page with the bar graphs shows a definite and significant drop in incidents when the steel dowels and uprated gasket was introduced – and this is the best we have as regards proof MGR are on the case and the changes made are effective. HGF still happens, but it is mercifully a great deal rarer than it used to be. Long term forum pundits such as I put the current incident level at around the average for any other engine, so if you have one on a late model F or a TF, it is simply bad luck these days, and you could have equally experienced it if you had bought any other car.

Food for thought.

[stu@rt]

...is that the best posting you've ever read or what

[Scarlet Fever]

Cheers - but i'm knackered now!

Ideally it could do with a bit more about the changes made over time and the resulting drop in reported incedents, i might do some more later on today (i've a drawing i need to finish for tonight's post! Work intrudes... ).

[TREX]

ANDREW YOU ARE simply a PHENOMENON. I bet MGR doesnt know or get into this detail as you do. A+ grading.....you managed to pass with flying colors

Andy....you are amazing!!!! You explain everything so clearly that even I understand it! I want to give you masses of good rep for that post...but....unfortunately....tap tap tap tap tap......compooter says nooooooooooo.................

Think I may have given you too many points already!!!! But you do deserve them!!!!

I worship at the altar of Andy......!!!!!!

[mgzr24]

Hgf ???? Youve Lost Me There Ill Kick My Self When I Find Out What It Is

Thanks
From A Dumb Ass

[matho]

Well done Andy - concise and impartial.


Perhaps we should have it as a sticky in this Forum and hopefully cut down the number of threads like these?

[Sprinter]

Originally Posted by mgzr24

Hgf ???? Youve Lost Me There Ill Kick My Self When I Find Out What It Is

Thanks
From A Dumb Ass

Head Gasket Failure - speak to Clarkson if youwant more details

[Sprinter]

Originally Posted by matho

Well done Andy - concise and impartial.


Perhaps we should have it as a sticky in this Forum and hopefully cut down the number of threads like these?

Sticky sounds a good idea. Possibly with the thoughts of Roger attached?

[Scarlet Fever]

Maybe i need to add the other little bits as well then - hold on a mo...

[Sprinter]

Originally Posted by Scarlet Fever

Maybe i need to add the other little bits as well then - hold on a mo...

Andy, don't rush it, spend a bit of time on work. We will all appreciate the final product.

[DarrenB]

Ask whats right with an MG it is a shorter list and save's Andys finger tips (poor sod has no finger prints now )


Must admit i am back into the F again after a mail from the big man (God aka Andy P) yesterday. When dosen't know about an f isnt worth knowing.... On the other hand what i know about an F you could write on a stamp and still have room to lick it


D

[mgtf nut]

What a post - mucho respecto!!!!! Quality techno stuff.

I saw a post a while back, mine I think, asking about remote QED thermoststs on the outlet leg of the system from the head.
I totally understand what you were saying, and as I read it, the QED thermostat may partially help this remote, thermal "lag" effect. Do you think this may be one answer, of possibly several?
On another silly idea, my R 25 TDI sits on a PG1 gearbox, anyone fitted a diesel to a F/TF??

[Scarlet Fever]

OK, edited now - more conclusion type stuff and a summing up.

QED thermostats are a good idea, although i only know of one MGF fitment (loads of Elise boys have done it) and they had problems with the install. The theory is sound though. Personally i am after a remote thermostat from a TF, i have not had a chance to inspect one yet, but the installation issues should have been resolved (as it is now an OE part). I am also keen on the TF 160 oil cooler, but feel a combined oil / coolant cooler that will 'clamp' the temperatures together will be a lot more effective.

[mgtf nut]

More stupid questions I haven't worked on a "K" series since my son's GTA Metro years ago!!!

There was a thread about remote thermostats a short time ago, I saved the pictures for reference!! Do I infer that this is more an "F" problem than a "TF" one? When did the remote come in exactly, without taking the top off to have a look? Is the current siting of the thermostat on the incoming cooled water? I only ask as I believe the QED's are on the head exit pipe, which would be a better reference temperature to control. Is the thermostat drilled to allow a small bypass, if not, would it help?
Put Nomex on now!!

[mgtf nut]

Home > K > Components > Cooling > Top hose stat
Top hose thermostat holder
The thermostat arrangement on the standard K series engine is designed to give a rapid warm up and then to maintain a fairly high running temperature. For an unmodified engine in a standard road car this is quite reasonable but for a race engine, where there is much more heat being generated, the standard system can lead to problems. This situation is exaggerated further still when the system incorporates a remote radiator as in, for example, the Lotus Elise.
The standard thermostat installation can give rise to a flow where the coolant is cycling through the pump and the engine but is not being directed towards the radiator. This happens as a result of the thermostat reacting to the temperature of the cooled water from the radiator rather than the heated water from the engine.
Removing the original thermostat, and fitting our thermostat holder in the top hose can eliminate these effects. Our thermostat holder is available with a range of pre-fitted thermostats to give opening temperatures more suited for competition use.

[SHARKYMGF]

Originally Posted by Scarlet Fever

OK, edited now - more conclusion type stuff and a summing up.

QED thermostats are a good idea, although i only know of one MGF fitment (loads of Elise boys have done it) and they had problems with the install. The theory is sound though. Personally i am after a remote thermostat from a TF, i have not had a chance to inspect one yet, but the installation issues should have been resolved (as it is now an OE part). I am also keen on the TF 160 oil cooler, but feel a combined oil / coolant cooler that will 'clamp' the temperatures together will be a lot more effective.

Have a word with Techspeed, I know they have fitted at least 3 remote thermostats now (IIRC they are TF ones, not QED). I believe the converted cars belong to Rob Bell, Tom Randle, and (TimW?).

[Scarlet Fever]

No stupid questions, only stupid answers

Originally Posted by mgtf nut

There was a thread about remote thermostats a short time ago, I saved the pictures for reference!! Do I infer that this is more an "F" problem than a "TF" one?

True

Originally Posted by mgtf nut

When did the remote come in exactly, without taking the top off to have a look? Is the current siting of the thermostat on the incoming cooled water?

Also true

Originally Posted by mgtf nut

I only ask as I believe the QED's are on the head exit pipe, which would be a better reference temperature to control.

True again

Originally Posted by mgtf nut

Is the thermostat drilled to allow a small bypass, if not, would it help?

False.

The modification i referred to is to drill a small hole in the thermostat ring. I am not sure 100% how it is supposed to work, but it is fairly common in motorsport and is supposed to lower the operating temperature of the thermostat. This means it opens earlier aloowing more coolant to flow through the engine and thus reducing thermal cycling. It is the subject of hot debate amongst the Elise fraternity, some people swear by it and others (like me) are a bit sceptical.