MG-Cars.info
MG MGF Technical Replacing radiator and pipes BBS discussion at MG-Cars.info
MG-Cars.info
Welcome to our Site for MG, Triumph and Austin-Healey Car Information.
Parts
MG parts spares and accessories are available for MG T Series (TA, MG TB, MG TC, MG TD, MG TF), Magnette, MGA, Twin cam, MGB, MGBGT, MGC, MGC GT, MG Midget, Sprite and other MG models from British car spares company LBCarCo.
|
|
MG MGF Technical - Replacing radiator and pipes
| The time has come to get the radiator and coolant pipes replaced. A morning's research through the archives has given me much food for thought, so here goes with the list of questions: 1) I'm sticking with standard pipes for cost reasons, but see no disadvantage in fitting steel to a 1995 car that I hope to keep for say 5 years. Any thoughts? 2) Puzzling my way through the thermal shocks/ hole in the thermostat discussions, I conclude that there is probably no advantage and possibly some disadvantage in fitting a larger capacity radiator (price close to the same as a standard one). I reason that the larger capacity could make for larger slugs of cold coolant and as a road-only user, I doubt I need the greater cooling - correct? 3) What's the current status of the drill your thermostat debate? The second/bypass 'stat route seems to be preferred - has anyone done that on an MGF? 4) Could changing the radiator and pipes cause stresses on other parts of the coolant system? I remember doing this job on an old Golf only to then need to replace some rubber hoses that found the restored standard pressure too much. Anything obvious to look out for (though I assume that the pressure test will reveal problems)? 5) MS's site shows these ali heater connectors, but I am not sure where these go and if I should be fitting them? 6) I'm going to get the work done at Brown & Gammons (after the excellent service etc work they did this weekend) who use the vacuum method to replace the coolant. Should I still be checking for air in the weeks following? Thanks for your answers to any or all of the above... Regards Chaz |
| C Golvala |
| Hi Chaz, some great questions there, so here go my opinions on these issues: >> 1) I'm sticking with standard pipes for cost reasons, but see no disadvantage in fitting steel to a 1995 car that I hope to keep for say 5 years. Any thoughts? << No real disadvantage here. Of course it'll probably mean that you'll need to replace the replacement pipes in another 5 years - but that's hardly the end of the world! >> 2) Puzzling my way through the thermal shocks/ hole in the thermostat discussions, I conclude that there is probably no advantage and possibly some disadvantage in fitting a larger capacity radiator (price close to the same as a standard one). I reason that the larger capacity could make for larger slugs of cold coolant and as a road-only user, I doubt I need the greater cooling - correct? << Yes, you're right about the larger rad - it's only really needed for engines that have had a significant amount of tuning work performed. >> 3) What's the current status of the drill your thermostat debate? The second/bypass 'stat route seems to be preferred - has anyone done that on an MGF? << Bypass is preferred, and if you can obtain the pipework from an '04 TF, then this ought to be even better. Regarding remote thermostats, there are loads of Elises running around with them, and Carl has a similar mod on his F. >> 4) Could changing the radiator and pipes cause stresses on other parts of the coolant system? Anything obvious to look out for (though I assume that the pressure test will reveal problems)? << Yes, as the car gets older, the rubber hoses tend to get brittle. Both Dave and Ted have suffered from hose failures - Dave's probably more linked to engine work and movement of hoses... Anything obvious to look for? Not really, unless a leak is obvious. >> 5) MS's site shows these ali heater connectors, but I am not sure where these go and if I should be fitting them? << I think Mike would be better placed to answer this himself - but I can't see a burning reason to fit them. I'm sure that Mike's identified a problem that merits this solution however. >> 6) I'm going to get the work done at Brown & Gammons (after the excellent service etc work they did this weekend) who use the vacuum method to replace the coolant. Should I still be checking for air in the weeks following? << No, it shouldn't be necessary if the lads at B&G have done all the work correctly - plus if the auto-air-bleed valves (those Jiggle valves) work as designed (ie not blocked), any missed air will be harmlessly guided to the expansion tank. For peace of mind you could perform your own checks after a week. |
| Rob Bell |
>> 3) What's the current status of the drill your >> thermostat debate? The second/bypass 'stat route >> seems to be preferred - has anyone done that on an MGF? >Bypass is preferred, and if you can obtain the pipework >from an '04 TF, then this ought to be even better. >Regarding remote thermostats, there are loads of Elises >running around with them, and Carl has a similar mod on >his F. TBQH I would be a lot more worried about 'thermal shock from the engine reaching 100 degrees and then suddenly being cooled to 88 in a hot car on a cold winters day when the fan kicks in in traffic - Really there is little to prove that messing with the cooling system in this way will give you any greater reliability (Carl? managed to cook his engine by having a remote thermostat installed incorrectly, so choose you poison wisely!) >plus if the auto-air-bleed valves (those Jiggle valves) work as designed. This is true for the engine end, but the rad end only has a manual screw, having said that there would have to be a lot of air bubbles in the system to cause you trouble here! |
| Will Munns |
| >> Really there is little to prove that messing with the cooling system in this way will give you any greater reliability << Actually, there IS evidence to show that remote thermostats do work - and this is both indirect and direct on K-series engines: Indirectly, remote thermostats are what are employed on the majority of mid-engine cars, ever since Ford had the same HGF problem on their mid-engined RS rally cars. Remote thermostats cured the problem. Directly, Simon Scutham (who was instrumental in developing the QED remote thermostat kit, apparently), performed temperature monitoring experiments on his Elise at Bruntingthorpe. He found that the remote thermostat radically reduced thermocycling. In practice, with the remote thermostat fitted, racing HGFs have become a good deal less frequent. Final trench of evidence comes from MG themselves fitting remote thermostats (and they're fitted to the troublesome Freelander too) - as this is effectively what the new thermostat represents. So, yes, they do work ;o) >> (Carl? managed to cook his engine by having a remote thermostat installed incorrectly, so choose you poison wisely!) << Not Carl - it was Thierry: his neurologically challenged dealership managed to fit his thermostat completely incorrectly. This isn't an obvious mistake to make as the instructions are clear. Well, clear enough for most of the human race... ;o) |
| Rob Bell |
| Wasn't aware of the other evidence, but much else has changed, like the gasket design over the same timeframe. >with the remote thermostat fitted, racing HGFs have >become a good deal less frequent. This is quite a lot diffrent from most(all?) road use >Final trench of evidence comes from MG themselves >fitting remote thermostats (and they're fitted to the >troublesome Freelander too) - as this is effectively >what the new thermostat represents. Isn't the 'remote' thermostat in the TF just 'not attached to the engine' rather than at the front of the car, and more to do with the redesigned thermostat not fitting in the standard position? >Not Carl - it was Thierry I had a feeling that Carl had done it long ago! >This isn't an obvious mistake to make as the instructions are clear As a very wise man once said "never underestimate the inginuity of idiots" ;-) |
| Will Munns |
| >> Isn't the 'remote' thermostat in the TF just 'not attached to the engine' rather than at the front of the car, and more to do with the redesigned thermostat not fitting in the standard position? << That's fair comment Will - but if the gasket doesn't fail under the stress of racing, then it is very likely to fair well under circumstances of normal road driving :o) >> Isn't the 'remote' thermostat in the TF just 'not attached to the engine' rather than at the front of the car, and more to do with the redesigned thermostat not fitting in the standard position? << No, there is a fair amount more to it than that - pressure sensing has an effect upon water flow to the radiator... there is quite a lot to get one's head around to understand the new thermostat installation - some more detail here: http://www.mgf.ultimatemg.com/hgf_pages/cooling_system_modifications.htm >> As a very wise man once said "never underestimate the inginuity of idiots" ;-) << ROFL! Indeed! ;o) |
| Rob Bell |
| What about the drilled hole ? Good / Bad |
| S Laithwaite |
| Wow Rob, didn't realise you'd written up the water pump stuff as well! BTW - goldfish 1 to goldfish 2 - hankie didn't work for me ;-) Stephen wrote: >>What about the drilled hole ? Good / Bad Tech-speed drilled mine as a matter of course when my first HGF went. However, the new thermostat is a no brainer IMHO, but I'm still waiting on sourcing the new parts :-( |
| Dave Livingstone |
| Thanks.I ask as I am too replacing the rad and have managed to source an uprated one, sorry Mike don't have enough poundo's at the mo. Anyway I am concerned with the effect of a larger rad. In addition to this, with the drilled thermostat it does take a long time to get the temp up and I mean a long time I can drive at least 8 miles before I see normal temp and that is with a normal rad. God knows what it will be like with a higher capacity one. I must admit that when I got the bumper off I was gutted at the pathetic sight that greeted me of a so called radiator. Now I am definitley no expert but it was sooooo small in contrast to what I had on my 620 si which is obviously only 200cc larger. Wonder what the water capacity of that block was. |
| S Laithwaite |
| Yup, definitely goldfish: the pump was (and still is!) sitting in the boot of the Rover ready to give to you. D'oh! >> Stephen wrote: >>What about the drilled hole ? Good / Bad Tech-speed drilled mine as a matter of course when my first HGF went. However, the new thermostat is a no brainer IMHO, but I'm still waiting on sourcing the new parts :-( << Completely agree with you there Dave. Drill the thermostat is good - but doesn't solve the problem - merely reduces the amplitude of the oscillation. |
| Rob Bell |
| Stephen, I never noticed a great deal of change in the warm up period after the thermostat was drilled, but I did notice a significant warm up change after I had the TF oil cooler fitted - I now get oil and water up to temperature much quicker. The water temp runs up to its normal one notch below halfway in about the normal time/distance. The oil temp now moves off the bottom of the gauge between 3 and 5 miles depending on outside temp. However, retro-fitting the TF oil cooler to a '98 VVC is probably something Tech-speed will never want to do again! Rob, which boot? R75 or F. If the F, then leave it there and I'll tie the knot in the handkie for North Weald at the end of the month, unless you happen to be over my way at any time. Thanks. Dave |
| Dave Livingstone |
| It's in the R75 Dave - so I'll transfer it back to the MG before I forget! I'm also still thinking about an oil cooler: I can't make a convincing case on the oil-cooling aspect as I don't see excessive oil temperatures even on track (if the temperature gauge is to be believed), but on the basis of oil warming and reduced engine wear? Yes, this would seem to be worth it to me... ... would it reduce HGF? Who knows??? It would simply be too difficult for an individual ever to find out. If you had the resources of a car manufacturer, then the story would be different. |
| Rob Bell |
| Thanks all for the responses and to Rob for the new info on his site. So far, it still sounds as though no one has done the remote thermostat mod on an MGF. As with all things, there is a cost/risk ratio here. My assessment of all the comments here and in the archive is that the standard radiator is fine for standard engines driven on road (Rover 620 notwithstanding) - if anything it is sufficiently efficient (by virtue of its distance from the engine) to send too cold coolant to the head and block. It must be right to say (as Will does) that the HGF risk is greatly reduced with a properly bled, well maintained system with new-type gaskets and dowels. Although the thermal shock theory has grown in favour as the root cause for weakening the head gasket and as Rob says the evidence is there to back it up, I get the feeling that it is the biggest issue left after the more obvious matters have been resolved. In other words you're going to get an HGF faster from leaky pipes or old gaskets than from thermal shocks. Is that a fair summary? So, drilling the 'stat reduces the thermal gradient and costs next to nothing. Remote 'stat plus pipework eliminates the problem, but costs something - anyone know what? If anyone can even estimate the cost, we could then assess whether it is worth it to eliminate the residual risk posed by the thermal shock issue. I've asked B&G to comment and I'll let you know if they do. Regards Chaz |
| C Golvala |
| The other thing in the larger radiator's favour is the bleed screw - which is metal and not plastic as on the OE one. SF |
| Scarlet Fever |
| >>In other words you're going to get an HGF faster from leaky pipes or old gaskets than from thermal shocks. Is that a fair summary? I think it is a fair summary. I'd guess that more HGFs arise from unoticed leaks and subsequent overheating than slow cumulative issues like thermal shock, liner heights and localised boiling of coolant in the head after hot shutdown. However, a guess is all it will ever be, as we can never know the reality. |
| Dave Livingstone |
| HGF in the MGF is almost certainly multifactorial - and the cause of early failure (under 20k miles) is likely to be very different to failure that occurs at much higher mileages... ie build issues are likely to be a much more prevalent cause of early failure, with problems relating to cylinder liner height, material problems with the gasket (that may render it more vulnerable to thermal cycling) and so on. Later failure could result by the accumilation of thermal cycling and possibly even localised boiling over a long period of time. But without doubt, if there is a problem causing leakage of coolant (poor coolant cap, leaking pipe work, corroded radiator etc), then you are at a radically increased risk of HGF. All of the above makes for a somewhat confused message - which is part an parcel of the problem that we face with HGF. However, keeping the cooling system in optimal condition, with regular changes of antifreeze, is probably the best way an owner can keep out of trouble. Other mods do cost money - which is fine if you are an enthusiast owner who intends to keep the car for a prolonged period of time. Costs on a remote thermostat installation? Best clues on this are available from the QED website (and also Elise Parts I think) - but costing the TF thermostat is a problem at the moment simply because the pipe work is not readily available. The thermostat itself is however cheap: I seem to recall a figure of 12 quid? You could probably add a 'zero' to that figure for all the brand new pipes and unions through MGR though :o( |
| Rob Bell |
| I join this thread as another owner about to change the radiator due to corrosion. I was shocked when I saw how knackered it is after only 6 years. Im interested to know the outcome of the Galvanic corrosion investigation. It must be something other than poor quality parts?? In my case for the records, the majority of the corrosion appears to be at the top of the radiator behind the grill (and hence the strongest wind blast). There is little left of the fins, just the cores suspended rather worryingly. This is the second problem I have had with the coolant system in as many weeks, the first being the split coolant pipe saga just before Christmas. Anyway is there any chance that someone could tell me where I can find the instructions for changing the radiator, as I can not see them on this thread anywhere. Also, I don't know if the radiator mounts are metal or not (yet) but if they are then is it worth trying to insulate them from the rest of the car when re-fitting the radiator using fibre washers, insulation tape etc? This should prevent any Galvanic corrosion occurring? Thanks. |
| R Longmore |
| Am I right to assume that fitting the TF thermostat in place of the standard F one is a no-go, otherwise we'd all be coughing up our £12 plus VAT? The QED solution costs £125 plus VAT for the thermostat and a kit of necessary hoses. It comes in 74, 82 and 88 degree C variations. QED recommend the 82 deg one even for road use and say it will go in an MGF, though it may be a tight fit. I'm still trying to understand what causes the cool coolant hitting the engine (I refuse to call 40 deg C cold!). Reading further from the links on Rob Bell's site, I see that Simon Scuffam and QED both refer to the problem in the context of uprated engines (although the QED site does suggest that the distance of the radiator from the mid-mounted engine exacerbates the problem). Though age must be a factor, Simon's explanation includes this: "All works fine for the std car that just trolls around, spending very little time on full throttle at speeds over 70Mph. The problems start when the engine is up-rated, driven at 100+ mph, full throttle for 20+ seconds. What happens is that the water back from the radiator gets very cold because of the huge amounts of air forced though it at speed and the relatively slow water flow rate... So all of a sudden, the cylinder head water jacket goes from 95+C to 40C..." The more I read this the more I think that how the car is driven has a bearing on the problem. On my journey to Baldock last weekend I certainly spent over an hour at motorway speeds, but it takes me 15-30 minutes of B and A roads to reach those speeds. Surely there should be less sudden temperature changes from accelerating a well-warmed car? Or would the cooling effects of the high speeds, still cause the thermal shock effect? A couple of points about the suggested modifications. Losing the heater until the thermostat opens would be a real pain on cold mornings when you want to defrost the windows and drive off quickly and safely. Not to mention the effect on emissions. Secondly, could you insulate the thermostat from colder water coming in from the radiator? I am not clear from the explanations whether the 'stat closes because of indirect cooling (see QED site) or because the cooler radiator liquid actually enters the 'stat. If the former could some insulation help? Rob, I assume that you're waiting for the results of the Techspeed trials of the Davies Craig/MAW unit before deciding what to do, or are you debarred due to racing regs? The water pump seems to address a separate problem from the thermal shock issue, though. PS: R Longmore - the radiator is rubber mounted, I think. See Rob's site under "Why do Head Gaskets fail" for a description of the standard system. 'pologies for the long post, but this is interesting stuff. Best Chaz |
| C Golvala |
| I agree Chaz - interesting and confusing at the same time! You raise some very relevant points regarding driving style - this is something that is hugely variable, and it's effects upon thermal cycling are unknown. You are also right that Simon's testing was over some very er, testing (!) miles at Bruntingthorpe. This was both most relevant to what he was trying to find out and also the most likely way that thermal cycling would be uncovered in a short period of time. Basically, what we need to do is do some testing of our own in MGFs being driven 'normally' over a longer period of time. Tim was working on a laptop based telemetry system that we were going to use for testing of the water pump, and also could be used for testing of the thermostat ;o) I think that the status of the kit is that it is pretty much ready, except that there is some problems getting all the channels collecting data in parallel... If we could do this, we could get a much better handle on the problems that we face. That there is a problem here, I am confident of; someone who I can't name, but is involved in the MGF racing programme, told me about the thermostat drilling solution to counter thermal cycling... Is 40C cold? It is against a block who's mean temperature is the same as the oil's temperature (circa 110C)... Can you insulate the thermostat from the cold radiator return? No. Unless you redesign the cooling system (I guess that this is effectively what one is trying to achieve with the remote thermostat). TF thermostat is a totally different design, and comes in its own plastic moulded housing. It certainly can't be fitted inside the existing thermostat housing! In fact, the existing thermostat housing is retained on TF, albeit with a spacer with a hole through it's centre (presumably to maintain the designed flow resistance) Ross - I'm not 100% sure how the rad is fitted in the F (ie rubber insulated or not - I thought not?) - anyone else know? |
| Rob Bell |
| Chaz, > >I assume that you're waiting for the results of the Techspeed trials of the Davies Craig/MAW unit before deciding what to do, or are you debarred due to racing regs? The water pump seems to address a separate problem from the thermal shock issue, though. Yes, we're still in the middle of the EWP project and yes it is a separate issue from Thermal shock. The water pump is only intended to reduce the potential for localised boiling of coolant in the head after the engine is shutdown. This would be pertinent to the motorway driving situation. The car gets hot at 70(+) on the motorway - you then pull in to the services and switch off. Heat soak in the head, especially on the hot exhaust side of the engine, causes the small amount of coolant left in the head (and no static) to boil. Over a long period of time this can help lift the gasket or soften the head. The test equipment Rob mentions is how we plan to try to measure the effectiveness of the EWP installation. |
| Dave Livingstone |
| the issue with racing and thermostats is this, engine gets to temp, rad is very cold. Water running thru the thermostat via the bypass hose (thru the heater or not) gets very hot, thermostat snaps open wide, but the engine is at high revs and pushing water around the system very fast, cold rush of very cold water shoots past the thermostat before it has chance to close, thermostat snaps shut, block is cooled, water reaches temp, thermostat snaps open, repeat. When not ragging the engine, the water comes to temp much slower and the thermostat opens gently, mixing the very cold water from the rad with the hot from the bypass to create a nice even mixture at 88deg C, also the water is not rushing thru the system at such speed, so the thermostat has more chance to react. The TF thermostat has a pressure releif, so when at high RPM the engine gets extra cooling regardless of temp. The original remote thermostat was another thermostat at the front of the car which did some mixing of hot and cold, leaving the thermostat on the engine with less work to do. Drilling a hole in the thermostat will have little effect, it was a mod designed when cars did not have bypass circuits, and the thermostat was on the outlet side of the engine. The F thermostat is on the inlet side. Drilling a hole will only let cold water circulate when the engine is cold, it will do nothing to stop the thermostat snapping open when the engine is hot. |
| Will Munns |
| Drat, Dave! Now I'm going to worry everytime I stop for a leak (ouch!) at the Watford Gap! This reminds me of the old advice to leave the car idling for a little time after a fast run to allow the fan to cut in and cool the system. Would that help? Rob, we'd all be interested in the results of the tests and wish you, Dave and Tim etc best of luck. As for whether 40C cold? Well, to the man who can use the word "multifactorial" without blushing, I'd say all things are relative. thanks Chaz |
| C Golvala |
| Ahh! Will's voice of reason was posted whilst I was spelling "multifactorial". What you say about normal driving resulting in temperature-homogeneous coolant makes sense to me. Does anyone know or can estimate the correlation between the coolant temp sensor, the oil temp sensor and the opening of the thermostat? I usually get the coolant gauge to one notch under half-way in about a mile and the oil gauge moving off the marker in 5 miles (speeds from zero to 70mph on the station run morning and evening). It would be interesting to know if the stat is likely to be open by either of these two points. Another thought...has anyone thought of or found a more accurate replacement sensor and gauge for coolant temperature? Cheers Chaz |
| C Golvala |
| >> Well, to the man who can use the word "multifactorial" without blushing, I'd say all things are relative. << LOL! BTW, the boss' LR Freelander has suffered - surprise surprise - a HGF. He's getting the gasket replaced, along with all the pipework and the latest 'TF' type pressure sensing thermostat fitted (looks as though this is now standard practice at LR these days). Parts are being paid for by LR, but it transpires that the parts bill is somewhere in the region of 280 quid. Given that the gasket costs about 15 quid, and the thermostat 12 - that means that the pipe work is costing a staggering 200 quid!!!! 8-O I think that this is going to prove a costly option for anyone considering retro-engineering the TF PRT option to their Fs - unless second hand parts become readily available :o( |
| Rob Bell |
| >> the issue with racing and thermostats is this, engine gets to temp, rad is very cold. << This is certainly true Will - and this certainly explains why people don't get HGFs 30 metres down the road having just bought the car off the forecourt. The problem is likely to a 'low-grade' problem that manesfests as a problem only after long term exposure. Like many things, a low intensity nasty can be as bad, if not worse than a high intensity nasty over a short period of time. It's the incidiousness that is part of the HGF problem - and probably also part of the reason why it has taken Power Train so long to sort the problem out (that and an osterich mentality!) Chaz, you can't rely on the temperature gauges: they're both highly damped, so their position only gives you a general idea of what the temperature is up to. The only way do this properly is with some proper instrumentation - hence the need for telemetry... ;o) |
| Rob Bell |
| >>This reminds me of the old advice to leave the car idling for a little time after a fast run to allow the fan to cut in and cool the system. Would that help? Yes, and this is of course what we do after blatting around the track. It's just a pain to have to keep doing, even on track days, let alone in service stations when you're desperate for a leak! >>has anyone thought of or found a more accurate replacement sensor and gauge for coolant temperature? As Ron says, it seems to be the gauge that is heavily damped. The Elise guys have the same temp sensor, yet have digital readouts that freak them out as they rise and fall :-) It would be nice if someone came up with some alternative to the one notch under half problem, but I've never heard any talk of it. |
| Dave Livingstone |
| For 'Ron' read 'Rob' !! |
| Dave Livingstone |
| Dave, if you wanted a digital temperature read out, it would be quite easy to install one (not especially cheap though) - Demon Tweeks retail a range of digital instruments that are dual display: you can have water temperature and oil temperature. The good thing about them is that they are the same 32mm(?) fitting as the existing circular oil temp and clock gauges in the centre console. Replace one of these, and you can have the new digital gauge without messing around with the car's interior :o) Only question is are they compatible with the MG's existing temperature sensors? That I don't know... but it would be possible to find out. See http://www.demon-tweeks.co.uk/catalogue/category_list.asp?cls=MSPORT&grp=MC056&pgrp=MP051&tlgrp=M004 for more details. At over 134 quid, I don't think I fancy it... :o( |
| Rob Bell |
| >>At over 134 quid, I don't think I fancy it... :o( Agreed, better to put that sort of money towards the new thermostat pipework! |
| Dave Livingstone |
| Gulp - even going from SPA Direct, the gauge (model number DG209W) costs 141.38 (including VAT) - plus another 4.40 for delivery. Erm, no thanks. :o( Pitty. |
| Rob Bell |
| Hi I have followed this thread with concern as I have a 1 year old TF 160. There seems to be many theories as to why this engine suffers from HGF's But why does it only seem to affect this engine. Other engines in other cars are cooled in a similar way, their thermostats must operate in a similar way so what is so different.....Are you all looking for a cure for a badly designed block or head or have I missed an earlier thead that has already eliminated this....... Ron |
| R PRESS |
| I don't think that the 1.8 litre K-series is fundamentally weak - although there will be issues arising from bore and stroking a 1.4 litre engine to this extreme. Part of the problem may also centre around the issue that the block is primarily oil cooled whereas the head is primarily water cooled - so if there is instability in the water temperature control, it means that the head and block will be at different temperatures. Thus the problem does seem to be centred on the location and operation of the thermostat. Okay, that's my opinion - but there is evidence from what Power Train have been up to in terms of modifying the cooling system that would appear to support this hypothesis. |
| Rob Bell |
| >>Other engines in other cars are cooled in a similar way There is one aspect that is different to the average engine/car, of course, and that is the front radiator/long underfloor pipes/rear engine layout. OK, this doesn't excuse the K series in other cars (although the Elise is similar to the MGF), but it does seem to increase the propensity for leaks and may exacerbate thermal shock. |
| Dave Livingstone |
| >Part of the problem may also centre around the issue >that the block is primarily oil cooled whereas the head >is primarily water cooled Having compleatly taken apart the engine, I no longer beleive this is true, the water jacket extends for some distance into the block, and the oilways are quite small, certainly there are areas only cooled by the oil, but they are much lower down and unlightly to effect the head gasket at all. |
| Will Munns |
| >increase the propensity for leaks I beleive that the diffrences in occurance you see between other cars and the F/Elise are due to two things, firstly there is the tortuous path, this must mean that the water flows slower (much more drag, but the same water pump). Also the K is known for HGF's, and it could be the extra potential water leaks from the long runs, and possible air lock issues could push the numbers up to unaceptable. Also the stress the engine is put under is much greater for the F/Elise/LR due to the way the cars are driven. >may exacerbate thermal shock I don't see why it should - execpting the fact there is little airflow over the engine - where in a normal car the air thru the rad also flows around the block |
| Will Munns |
| >> Having compleatly taken apart the engine, I no longer beleive this is true, the water jacket extends for some distance into the block, and the oilways are quite small, certainly there are areas only cooled by the oil, but they are much lower down and unlightly to effect the head gasket at all. << Well, we both know of a certain Elise owner who seems to think he looks like Tom Cruise who might vigourously disagree with you there Will ;o) LOL IMO this argument only serves to increase the need for more empirical data collection. As I recall, Carl mentioned that he could detect external block hot spots (in the region of the alternator heat shield) using an IR temperature probe... Whether this tells us anything about block cooling or not remains to be seen of course. |
| Rob Bell |
| I wouldn't worry so much about this one Mr. Press. Although Rob is right that prolonged exposure to a minor issue can be as bad as a major issue, on balance, I still feel that this is a relatively small cause of potential HGF, unless, perhaps, you have an uprated engine and are given to wellying it around a race track. Using the new gaskets, dowels and having sound pipework are more essential. Two comments from reading online to add to Will's: Scuffam's notes mentioned that the Elise gets no cooling to the underside of that K series so any differential between the block and head temperatures is more pronounced in that installation than in the MGF. The second point is that the results of the online log of HGF (see Rob's site for 2002's analysis) show a marked fall off after 1999, suggesting that later cars (even before the latest gasket) are less prone to HGF. However good the statistics are, the thermostat is the same. Re gauges, I'd prefer analogue anyway, so I looked up Smiths (now called Caerbont Automotive) which seems appropriate since this is an MG! Their helpful technical people said that an electric water temp gauge would be less responsive and accurate than a mechanical one, but nonetheless better than the current one. Mechanical gauges require a capillary tube with ether in it to be immersed in the coolant, but they have sold gauges for K series engines so it is possible to install. Costs: £23 for an electrical gauge and £12 for matching sensor; £50 for mechanical and £80 for dual oil and water temp if you don't want to replace the clock. Sounds to me that someone with the technical nouse could install the electrical one and blue-tack it to the dash for £35 all in. Final comment - B&G's MGF cup car (been running for 7 years and has won the title twice) has the standard thermostat and no HGF problems. Her ends another long post from me. Well, I wouldn't use "multifactorial", but I am a lawyer! Cheers Chaz |
| C Golvala |
| >> Using the new gaskets, dowels and having sound pipework are more essential. << I'd definitely agree with that Chaz :o) But HGF occurs on cars with perfectly servicable cooling systems (although the coolant resevoir cap could be the 'Joker' in the pack - a number of faulty caps are 'known' to be in circulation) >> Scuffam's notes mentioned that the Elise gets no cooling to the underside of that K series so any differential between the block and head temperatures is more pronounced in that installation than in the MGF. << That's true too - and even the LR Freelander has an excess of HGF, and presumably that car has even better 'sump ventillation' than the F (not sure as to the extent of the sump guard which *could* dramatically reduce this cooling effect). But remember that the problem that is believed to be occuring is a relative difference between oil and water temperature. So although there may be a significant difference between an oil temperature of 110C [MGF say] and 120C [for the Elise], in terms of differential to water temperature, if the water temperature still falls to 60C on thermostat opening, that's STILL a 50C difference between the two. >> The second point is that the results of the online log of HGF ... show a marked fall off after 1999, suggesting that later cars (even before the latest gasket) are less prone to HGF. However good the statistics are, the thermostat is the same. << Tricky this, the stats aren't all that representative as we well know - and what if the new gasket and dowels are only delaying the time to when the gasket fails? Oops, here I am playing Devil's Advocate ;o) Thanks for the smith's gauges tip. :o) Mind you, I am hoping that Tim's telemetry system will work out even cheaper! :o) |
| Rob Bell |
| Interestingly, B&G pressure test expansion tank caps separately before fitting because of the high incidence of duff ones. I understand what you say about the differential still existing and agree with you about the statistics being only a rough guide. Playing devil's advocate myself, the changes that Powertrain are making do suggest that there is an issue to be addressed - did I read that the PRT 'stat is only being put on the 160 engine or is it for all? Going back to the original list's number 2, has anyone a view on using an aluminium radiator (either the TF or an ex-cup car one) instead of the standard one? Clearly corrosion should be less of an issue, but could the more efficient cooling add to the risk of thermostat recycling as discussed above? Thanks Chaz |
| C Golvala |
This thread was discussed between 05/01/2004 and 08/01/2004
MG MGF Technical index
This thread is from the archives. Join the live MG MGF Technical BBS now