Just a Theory Between Rad size and HG failure

[Petersubaru]

...Is it possible that the subi cooling system is under sized to maintain headgasket integrity..the reason I am asking is because I had a visit from a friend who showed me his 2007 Jetta and somehow the topic of discussion moved over to HG failure concerning subarus..it was pointed out to me that the 2.5L jetta motor had a much larger cooling system...10qts vs 6.7qts for the subi and the jetta rad was slightly larger when comparing the LxW...but what surprised me the most was that vw rad is almost twice as thick...

[zyewdall]

Hmmm. But I've never seen the temperature guages on the newer subarus move as long as the cooling system doesn't have anything wrong with it -- same place on the gauge whether going downhill, or flogging it uphill. If the system was undersized, wouldn't you expect the temp to increase when really working the engine?

 

Z

[nipper]

Nope. HG's are extreemly complicated parts and no one issue kills a sooby HG. u look at the history of the automotive industry during this time, emissions went from OBDI to OBDII with tighter specs. There was a very short time to adapt and desighn to the new standard. Everyone had head gasket issues during this time. The materials at the time were at the edge of what they can do (which is why the failure rate industry wide was from 100% to 15% and everything inbetween depending upon how cheap the HG's were made to begin with). Also some things need real world testing. Even lab testing takes time for analysis and to come up with a correction for the failure. When a critical part fails a test, the test starts over. Sometimes the tests themselves, while consistant, do not reflect real world conditions, and sometimes they just can't (like reproducing 150,000 miles of city driving).

 

Failure starts in the combustion chamber, where spot temps are a few 1000 degres and combustion pressures can be in the 15,000-18,000 psi or more range. This is all precisly controlled by the computer, except when you get knock. Knock is one of the biggets detriments to HG (assuming the engine doesnt already have a design weakness)

 

Subarus tend to run fairly consistantly at 175- 205 depending upon the outside air temp and load. ine to market (computers

 

We expect alot of our cars, and sometimes, we just have to be living labratories. It can take up to 10 years to bring a new engine to market, though computers (for development) have gotten faster and have reduced that number to 6, unless something in the market changes, then it is back to square 1 again.

 

 

nipper

[nipper]

PS. You can also add in open deck vs closed deck design (and even hybrid deck....confused yet ?).

 

To me i know how it fails in subarus, i just dont know what kicks off the entire process. Since it is an open deck, the HG is not fully supported. Imagine trying to pry against a hollow core door vs a solid door. the solid door wont give, or will take a very long time and alot of work. The hollow door will yield fairly quickly.

 

Since they were commited to hollow core doors, they went with a 3 peice door (HG) where the outer skins were steel. Now it is as strong as the solid core door (still with me?).

 

Now since there is not a 100% failure rate (i'm looking at you Dodge Neon) the question is what sets off the failure. That I dont know, and i dont know if anyone will ever truly have an answer, but you can design around all the known causes and beef them up, and erradicate the unknown cause. Maybe it is just one bad tank of gas that causes ping, maybe its a bad batch of HG's, maybe it was just the monday crew in japan hungover.

 

nipper

[Petersubaru]

Hmmm. But I've never seen the temperature guages on the newer subarus move as long as the cooling system doesn't have anything wrong with it -- same place on the gauge whether going downhill, or flogging it uphill. If the system was undersized, wouldn't you expect the temp to increase when really working the engine?

 

Z

.. the newer gauges don't show moderate temp changes ..for example if your coolant was 20deg hotter from normal, you still would not notice it

[nipper]

.. the newer gauges don't show moderate temp changes ..for example if your coolant was 20deg hotter from normal, you still would not notice it

 

 

Youll see it on an OBDII readout, but yes the gauges are numb in the normal range, otherwise they will get alot of people saying thier temp gauges are moving all over the place. And since the majority of people are stupid, it ties up shops (80's gauges were very reactive, and this is the lesson that OE's learned... that customers can be idiots) As you start getting out of the normal range they are more accurate.

 

nipper

[zyewdall]

Youll see it on an OBDII readout, but yes the gauges are numb in the normal range, otherwise they will get alot of people saying thier temp gauges are moving all over the place. And since the majority of people are stupid, it ties up shops (80's gauges were very reactive, and this is the lesson that OE's learned... that customers can be idiots) As you start getting out of the normal range they are more accurate.

 

nipper

 

Wow... and here I was thinking that the new cars just hold temperature very steady compared to my '70's pickup with an actual mechanical thermometer that goes up and down depending on what I'm doing with it.

[Petersubaru]

Nope. HG's are extreemly complicated parts and no one issue kills a sooby HG. u look at the history of the automotive industry during this time, emissions went from OBDI to OBDII with tighter specs. There was a very short time to adapt and desighn to the new standard. Everyone had head gasket issues during this time. The materials at the time were at the edge of what they can do (which is why the failure rate industry wide was from 100% to 15% and everything inbetween depending upon how cheap the HG's were made to begin with). Also some things need real world testing. Even lab testing takes time for analysis and to come up with a correction for the failure. When a critical part fails a test, the test starts over. Sometimes the tests themselves, while consistant, do not reflect real world conditions, and sometimes they just can't (like reproducing 150,000 miles of city driving).

 

Failure starts in the combustion chamber, where spot temps are a few 1000 degres and combustion pressures can be in the 15,000-18,000 psi or more range. This is all precisly controlled by the computer, except when you get knock. Knock is one of the biggets detriments to HG (assuming the engine doesnt already have a design weakness)

 

Subarus tend to run fairly consistantly at 175- 205 depending upon the outside air temp and load. ine to market (computers

 

We expect alot of our cars, and sometimes, we just have to be living labratories. It can take up to 10 years to bring a new engine to market, though computers (for development) have gotten faster and have reduced that number to 6, unless something in the market changes, then it is back to square 1 again.

 

 

nipper

I hear what you are saying and it all makes sense..but, is there any possibility, that with a larger cooling system, you could have more consistent coolant temps and not as much temp recycling within the "normal" temp range of 175-205..thus bringing about less expansion and contraction of the head..in an age where cost cutting is across the board, you would think that other auto manuf. would have smaller cooling systems and yet they don't...the rad sizes just got me thinking, that's all..subi 2.5 turbo 7.6qts..subi 3L 8.4qts as oppossed to our families 3L lincoln 11.2qts...freinds jetta 2.5 10qts

..even my old '86 diesel has a larger all aluminum rad..perhaps the larger rads are for cars with higher top end speeds

Edited January 16, 2010 by Petersubaru

[Petersubaru]

Wow... and here I was thinking that the new cars just hold temperature very steady compared to my '70's pickup with an actual mechanical thermometer that goes up and down depending on what I'm doing with it.

..both my '86 vw and '86 subi will do as your truck does

[Sonicfrog]

Some older Sub's did have that correlation, just ask anyone who has an EA82 Turbo car like the RX or GL-10. The problems with the newer Sub's and HG's, between 97 and 2001 IIRC, were mainly from a batch of sub-par quality head gaskets from a manufacturer. In this case, the problem was not limited to just Subaru, other Japanese auto manufacturers had similar problems about the same time.

[edrach]

.. the newer gauges don't show moderate temp changes ..for example if your coolant was 20deg hotter from normal, you still would not notice it

 

You have two temp sensors in your car, one feeds the gauge and the other the ECU to control mixture and such. After engine warm-up, the instrument temp gauge doesn't budge at all, no matter the conditions unless there's an actual issue with the cooling system.

 

I installed a ScanguageII on my Impreza since I wanted to have real-time indication of engine temp, volts, rpm and to monitor CEL codes. I found the temp does vary quite a bit. I can actually see when the thermostat opens to allow more cooling. During regular driving the engine temp varies from a low of 182 degrees to a high of 196. When I'm running rallycrosses in the summer I'll sometimes hit 205 degrees. During all that spread the instrument gauge never moves.

 

I bought the scangauge primarily to have an on-board unit to check and remove CEL codes. But I've found the gauge to be extremely useful in monitoring the cooling and alternator output during everyday driving. It hasn't happened yet, but I suspect I'll get better warning when there's a cooling system failure or alternator failure.

 

Just so no one has to ask about the scangauge accessory, I bought mine through Rallitek. I see from the link that the price has gone down; I paid over $200 for mine a number of years ago.

http://search.store.yahoo.net/cgi-bin/nsearch?catalog=yhst-49187380673845&query=scangaugeII&searchsubmit=Go&vwcatalog=yhst-49187380673845&.autodone=http://www.rallitek.com/

Edited January 16, 2010 by edrach

[Fairtax4me]

Wow... and here I was thinking that the new cars just hold temperature very steady compared to my '70's pickup with an actual mechanical thermometer that goes up and down depending on what I'm doing with it.

 

Hook up an old school temp gauge on the Subaru it will do the same thing.

Fuel gauges are the same way. They have circuitry built in to keep the needle from swinging when you go around corners too fast or are facing up or down hill at a stop light.

 

From what I've read, Subaru HG failure is usually failure leads to overheating. Where if the cooling system was inadequate, you would see the problem as Overheating leading to HG failure. And the failure in that case would typically be because of a warped cylinder head. (some early 90's GM cars had that trouble)

[nipper]

Hook up an old school temp gauge on the Subaru it will do the same thing.

Fuel gauges are the same way. They have circuitry built in to keep the needle from swinging when you go around corners too fast or are facing up or down hill at a stop light.

 

From what I've read, Subaru HG failure is usually failure leads to overheating. Where if the cooling system was inadequate, you would see the problem as Overheating leading to HG failure. And the failure in that case would typically be because of a warped cylinder head. (some early 90's GM cars had that trouble)

 

 

Well a failed sooby HG wont lead to overheating immediatly, but a cooling system failure (broken hose no coolant) will lead to failed HG on almost any aluminum cylinder head engine.

 

nipper