Low Temperature Thermostat

Isn't that a 997 problem/fix?

my understanding it's any post '01 996 bar GT2/3/RS.... could be wrong.

Hi, booking the 996 (2003) in for a coolant change in a couple of weeks and wondered if anyone has experience of the above? Reccomended by Hartech to help maintain cylinder lubrication. The car won't be going on track but maybe low revs / high torque is the most testing time anyway. Any opinions welcome! Tim

Yes I'm sure it's for all M96 and M97 motors. Seriously considering doing that mod to my 997S. OPC quoted £200 for fluid and labour, I supply new thermostat. I don't have Porsche warranty so no issues with non original parts.

Having re-read Bazs posts, I think it is directed at the 997/987 block 'a piston siezure on bank 4-6 seizing one side of the piston - due to the cylinder running too hot' He does mention that high mileage 996s suffer from cracked liners - but of all the posts I've read over the years the only terminal problem I keep reading about on 996s seems to be the IMS. With the 997 it seems to be the overheating problem leading to piston seizure on bank 4-5-6. The thermostat is cheap, so maybe worth getting anyway but it may affect your warranty.

Warranty runs out in four months so that's not an issue. I was thinking that if the problem is mainly down to high temperature thus lack of lubrication strength this could indicate a need for a slightly thicker oil. Maybe will also change up to a 5W-50, i'm a bit uneasy about the Magnatec 10W-40 as it's unsupported and the car takes a while to warm up. I'm hoping 2003 cars are after the IMS problem? Tim

Baz explains it that the engine is getting too hot and as a result the oil is becoming too thin for purpose. Once you fit the new stat the oil should be fine:- 'The biggest problem I can see occurs when the engine is driven spiritedly. At that moment the cylinders will imediately get hotter and conduct that heat to the cylinder block and the coolant in it - but in that block the amount of coolant is small (compared to the amount passing through the cylinder head) and the cylinder head is no where near as quick to conduct the heat outwards and to the coolant (which is flowing through it in large volumes). So the small amount of hotter cylinder block coolant mixes back with the larger amount of cylinder head coolant (which is still cool) and passes at little increased temperature to the radiators - which therefore have a marginal affect on the temperature drop (being a function of the temperature differece which has been diluted by the mixing ratio). It then passes back to the engine thermostat which hardly notices the difference and in any case is designed to react slowly - and back round the engine again. While all this is going on the temperature between the cylinder wall and the piston has increased significantly and with it the oil seperating the two gets thinner and does not provide the support it should (while the coating on the pistons may get too hot). Eventually it all settles down to a constant temperature again until the driving style changes. If you compare this with more tradditional engine design, all the coolant frst goes into the block and cools it quickly first - then it passes through the head (so even if the cylinder block temperature has increased - the head raises that temp little more than before for a period of time) and the result then hits the thermostat (which sees the higher temperature and opens more) and then it almost imediately goes into the radiator. If you then fit a higher temperature thermostat (to help run the engine with reduced emissions) and cheat the dashboard reading (to avoid upsetting owners) - you get the scenario in full. Without re-machining inside the engine - unless the thermostat is changed to reduce the temperature it opens at - there is nothing any other changes will do to influence the cooling rate of change inside the clinders - because that thermostat controls the flow rate and the balance of coolant flow inside the engine is unchanged'

The IMS was still an issue until 2005 until they introduced the new design in MY'06

Thanks for that and the previous explanation, good oil and water temp control will be a bit more important than lowest emissions for me, fingers crossed that i don't get the IMS problem.

So is this a big job then i thought the thermostat is quite wasy to get to?

Replacing the thermostat is easy. Its at the bottom on the engine near the water pump (on the nearside). You'll see a inch and a half diameter hose exiting from the housing. You will need to drain the coolant first then disconnect the hoses and remove the housing cover. Porsche specify a special tool to wind the thermostat to remove it but on other forums, people have done it with long nose pliers. After that its just a matter of replacing the thermostat with the new one, use a new gasket for the housing cover, re connect the hoses and then fill the engine with coolant slowly and then bleed it using the Porsche stated technique.

The thermostat itself only costs around £60+VAT.

i have heard that its tricky to bleed the coolant where can i find the best method thanks peter

Could you advise from where? The LN Engineering item is $160, which includes the complete housing with stat installed.

That's the price http://www.hartech.org/ are selling them for.

Workshop Manual is freely available online somewhere. Hartech persuaded me to book in with a local specialist. Correct coolant is also vital, i think either OPC or Comma G30. HTH Tim

Thanks, Looked at website but can't see any reference to it. So I have emailed them for info.

You're right it's not on their site. I got the info from a thread where Baz mentioned his price.

I phoned and got someone who seemed to know what they were talking about. Pity they're so far away; always like to support a good business

I agree. He's one of the most helpful and knowledgeable Porsche people out there - I've never heard anything but the highest of praises for him and his business. Him and Chris Franklin should set up together as the ultimate one stop Porsche service centre.

Preferably more upnorth too its a 3 hour drive to coventry for me!

Porsche now recommend using the vacuum fill method but this was the recommended method when the 996 was launched: 1) Lift spring on bleeder at the top of coolant tank 2)Run engine at idle speed and add coolant until no more coolant flows into the cooling system when the engine is revved at moderate speeds. Coolant level is now at lower edge of filler neck. Coolant temp must not exceed 80 degrees. 3)Close coolant tank and warm engine to operating temperature at 2500 rpm until thermostat opens approx 10 minutes later. The radiator supply and return lines in the front wheel arch housing must be hot. 4)Forcibly engage both secondary fans by using a relay jumper at relay 20 and 22. 5)Allow engine to run an additional 5 minutes at 2500 rpm. Every 30 seconds briefly rev engine to 5000 rpm as the surges is important to purge the air. The coolant light may light during the bleeding process. Reset by restarting engine. Interrupt procedure if coolant level falls below min mark. Open coolant tank carefully to depressurise it then fill with coolant and intermittently rev engine again. 6) Deactivate the auxillary fans by replacing the relay jumpers with the relays. 7)Wait until coolant has cooled to 80 degrees (approx 25minutes) 8)Open the coolant tank carefully again and fill with coolant to the lower edge of the filler neck which corresponds to the max level when cool. 9)Reposition bow on bleeder valve and collant tank cap.

thanks for your help again ive printed and filed it thanks peter

996 = IMS Problem.

the upgrade stat opens 20 degree's lower/before the standard oem. Id like to do this too when I have a bit more spare cash.

Wow 20c? With a 997 it may be a good idea, but on a 996 I think you're trying to gild the lily while at the same time potentially losing significant power as the engine is running so much cooler than optimal. Did Baz mention anything about this?

I am not at work as I answer this (so don't have figures for power increases - but it is significant) ast the best temperature to run an engine for power is not the temperature that these cars run at. They run hotter than that to reduce emissions but produce more power when run cooler with the lower temperature thermostat and more at the temperature racing engines run at - so that criticism is inaccurate - sorry! I keep explaining - that the temperature of the cylinder heads and cylinders are running different to the last decades of engines designed the original way. Basically they are splitting the coolant path to run the heads cooler and the cylinders hotter than before. Hotter cylinders can reduce cylinder sealing and increase cylinder to piston wear (hence problems with scoring) and although the engines can be altered when apart to change that coolant flow balance - the only way to lower the cylinder temperature without is to fit a lower temperature thermostat. The temperatures you can read either on the dashboard or the air con unit are the result of mixing back together again - the coolant coming out of the heads with the coolant coming out of the cylinders and does not reveal the true cylinder temperatures. We have tested this with temperature sensors fitted into different parts of different engines before and after re-machining and with and without the lower temperature thermostat - so we do know exactly what we are talking about! Unfortunately on the bank 2 side of the engine (cylinder 4, 5 and 6) the cylinder coolant is hotter on the thrust side (the side the piston is pushing on the cylinder bore to produce power) but not on the bank 1 side - so the most vulnerable side is bank 2 (where 95% of failures also occur for that very reason). Part of the problem is bubbling in which localised boiling occurs at the cylinder outer surface and the result is magnified as gas bubbles don't convect temperature as well as a well vented coolant and cause a rapid localised rise in temperatures - which build more bubbles and so on - all in a few seconds of running. Therefore - although we cannot always predict which cylinder has failed - it is usually on that side. Once one has scored thereis usually an increase in oil consumption but the extra oil leaking up the cylinder combined with a lower compression pressure resulting tend to move the problem on to another cylinder that may then follow on failing later - and so on and so on until we have seen cylinders with all 6 bores damaged. Lots of other factors lead to unpredictability of the exact outcome - worn water pump impellors (plastic and as the bearing wears - often touch the rear of the housing in use), leaking radiators or expansion tank caps or seals - etc can all lead to lower coolant pressure which increases the risk of bubbling as liquids boil (or bubble up) at lower temperatures as the pressure reduces. Any bubbling not only upsets the cooling of the cylinder that does it first - but alters the flow of the other cylinders on that bank and also the other bank as well - so the permutations of how it all changes the flow rates and therefore individual temperatures is too random to be specific about. Blocked radiators (and air con condensers) also alter the temperatures and balances - it's a complex system although most regard it is simplicity itself. Although it would always be best to alter the flow inside the engine - at least lowering the normal running temperature is a step inthe right preventative direction - is inexpensive and a simple way to actually run with more power available - not less. Baz

Baz, I have emailed Hartech for price inc. delivery for the stat but received no response.

I see, thank you Baz. It would be interesting to see what power increases are available when you get back to work. Surely if you think there is no downside to this (apart from higher emissions/fuel consumption) this could be touted as an extremely cheap power upgrade and better for the engine overall - and therefore sell by the bucket load?

Baz - have you tried measuring temperatures with the clip on the coolant tank valve in the raised position? I believe this it a thermostat bypass valve, used when bleeding the system. When I had a coolant change on my C2 I ran it with the valve open for a couple of days to help bleed the system, and noted that it took a little longer to warm up, and that it ran a few degrees cooler. Cup cars run with the clip up all the time, by the way.

Thanks for that full reply. Is it really 20 degrees Celsius cooler? i thought more like 5 degrees. Got the 928 in for undertray straightening at the moment but hope to do the 996 thermostat asap Tim

No it is more like 20 degrees Fahrenheight but of course varies with the car condition, ambients etc - it is just that opening the thermostat earlier always helps. Sorry if we are a bit slow to respond at the moment, our engine rebuild inquiries have become so frequent we are hardly ever off the phone dealing with them and the consequential increases in workload – orders – manufacturing – progressing – testing etc. This is not just from the UK now – presently 9 ovverseas orders – the furthest being an engine coming from Korea.. Despite expanding staff and facilities – and investment - we are never going to be able to satisfy demand and so – we are forced to prioritise inquiries until we get up to speed coping with the increased order rates. Please be patient - we are trying our best. Baz

No worries, if you didn't have any customers your opinion would probably be worthless anyway! Tim

Hi, this thread is now 10 years old. Has the advice regarding the best way to protect a 996 engine from bore scoring changed with 10 years greater experience? I’m thinking of having a LTT fitted to my 4S. thanks

After reading various articles i changed mine, its a reasonable diy job if you have a jack etc.