BimmerFest BMW Forum banner

Got the 1.2 bar expansion tank cap from German Auto Solutions

60K views 113 replies 36 participants last post by  KWN-E39 
#1 ·
From Gary (German Auto Solutions). It's a thing of beauty. The top is engraved steel and the relief valve is "Reutter" - the OEM supplier for the OEM expansion cap. The beauty is that "if" the temp is getting too high, it will release the pressure much sooner than the OEM cap, thus having a chance for the plastic parts to last longer. I would venture to affirm that if the cooling parts will rupture due to pressure, at a lower pressure the damage could be smaller, i.e. the crack could be just that: a smaller crack, which could allow you save the engine. At higher pressure, a crack will develop very fast in a big rupture, where one could loose cooolant very quickly.

Thanks Gary. Beautiful product. :thumbup:



 
See less See more
2
#10 ·
#11 ·
I also ordered one of the 1.2 bar caps from German Auto and can't wait to get it in my car. But, remembering some of my college chemistry, I don't think it will result in increased overheating. It would probably increase the likelihood of overboiling, i.e. coolant out of the radiator cap due to the lower pressure. But, the coolant would still have to reach a certain temperature before that could happen. In this case, the 1.2 bar cap is still above 1.0 bar or sea level pressure and would work (like the 2.0 bar cap) to prevent overboiling.

Also, the point of antifreeze/coolant is to both lower the freezing temperature and increase the boiling temperature. It just seems like the new 1.2 bar cap will work. Still, there must be a reason BMW is using a 2.0 bar cap.

I will update my experience once I have replaced my current cap with the one from German Auto.
 
#12 · (Edited)
I think BMW did a mistake with the 2 bar expansion cap, and fixed it on the newer BMW models to 1 bar expansion caps to prabably avoid the catastrophic cooling system failures (ruptured radiator necks, split expansion tanks to name a couple). Pretty much for the same engine (N52 vs M54 - not much different).

Edit: The above statement is wrong. The source I quoted was off.
 
#15 ·
If you choose the lower pressure cap, make sure the expansion tank is topped to the LOWER LIMITS.
Quote from the "Fanatics", real-life test:

Originally Posted by kmulder
Hello everyone,

We've done extensive testing on this subject, and wanted to share some of the results. For testing, we used a 1997 E39 at idle in a garage. The main radiator fan was removed, and the auxiliary fan blocked off in order to force the system into higher temperatures and pressures. We first tried with only the main fan removed, but the auxiliary fan was successful in preventing the temperature from rising above 103 degrees C. It would turn on at 103 C and lower the system temperature to 95 C at which point the fan shut off again. We monitored coolant temperature both through the on-board diagnostics and through external software. Each test began with everything at ambient temperature. The expansion tank was tapped and a low-pressure gauge installed to monitor the pressure in the expansion tank.

Results:

We saw a maximum temperature of 126 degrees C (~260 F) before ending the test. We were unable to increase the temperature above this due to the efficiency of the system even without airflow. Keep in mind that a 50/50 mix of coolant will boil at approximately 135 C (275 F), so we were very close. Pressure at this temperature (126 C) was 20.5 PSI. The temp gauge needle was in the red zone well before this and the coolant temp warning light was on. Pressure never increased above 15 PSI before a temperature of 120 degrees C. We ran the test several times and achieved virtually identical results each time. We also tested the system with purposefully low coolant to see how that would affect the results. At a given coolant temperature, the pressure is lower in a system with low coolant.

During normal operating conditions, the system never operates anywhere near 2 bar. Assuming normal operating temps of 105 C (a bit on the high end), the system won't see pressures above 10 PSI.
 
#18 · (Edited)
people here dont understand the cap doesnt cause pressure, the heat does. heat is produced and maintained by thermostat opening and closing to route water to the radiator. rad cap is simply a safety device to allow pressure to escape if necessary. venting in a car ejects precious coolant and introduces air into the system; this is NOT the solution to lower pressure. lowering temperature is the ONLY way to reduce pressure of a ethylene glycol/water mixture. this total failure of people's understanding may be especially bad for the 540is(as they run hotter) and hit that 1.2 bars sooner then the i6 guys.
 
#20 ·
people here dont understand the cap doesnt cause pressure, the heat does. heat is produced and maintained by thermostat opening and closing to route water to the radiator. rad cap is simply a safety device to allow pressure to escape if necessary. venting in a car ejects precious coolant and introduces air into the system; this is NOT the solution to lower pressure. lowering temperature is the ONLY way to reduce pressure of a ethylene glycol/water mixture. this total failure of people's understanding may be especially bad for the 540is(as they run hotter) and hit that 1.2 bars sooner then the i6 guys.
Never said it reduces pressure in the cooling system.
 
#19 ·
Has anyone actually had a pressure gauge hooked up to the cooling system in the following conditions:

Idling when cold

Idling at operating temperature

at 2500 rpm with fan blowing air through rad to simulate highway cruising

I'd be curious to see all the pressure readings under these conditions.
And another thing that hasn't been clarified, is the 2 bars of pressure gauge or atmospheric? I assume it's gauge since that's how we measure everything else (tires, fuel pressure, etc.).
 
#41 ·
Here ya go. This is on my 2000 540 (M62tu engine) with the 2 bar cap. I do have a 1.4 bar cap but didn't use it as the cap doesn't effect the pressure.

I did this last fall, so the outside air temps were not in the 90's, no AC either.
I'll do it again this summer and see if it changes much.
It was posted and graphed on the other message board.
I heated the engine up in my driveway with the electric fan off to allow the engine to heat up quicker.
When the KTMP reached 103C I took the car out for a spirited drive, the temp never rose above 105C.

KTMP Reading/PSI Reading
40C/1.5 PSI
45C/2 PSI
50C/2.25 PSI
55C/2.5 PSI
60C/3.0 PSI
65C/3.25 PSI
70C/3.5 PSI
75C/4.0 PSI
80C/5 PSI
85C/6 PSI
90C/6.5 PSI
95C/7.5 PSI
100C/9.0 PSI
105C/16.5 PSI

When I returned from my drive I released some pressure via the cap. It was then down around 5 PSI. I went out for another drive, when I returned it was still at 5 PSI.
I'm going to modify the t-stat heater circuit to drop the temp which will drop the pressure.

 
#22 ·
The beauty is that "if" the temp is getting too high, it will release the pressure much sooner than the OEM cap, thus having a chance for the plastic parts to last longer
...but your statement here may not be correct. in a rice cooker, releasing pressure is fine. in a automotive cooling system, releasing pressure sooner also releases coolant(and introduces air), which actually will force the engine temp to rise even FASTER. in an overheating scenario, shutting engine down(and breaking the cycle of heat production) is the only way to stop it.
 
#23 ·
Having read most of the e46fanatics thread and other posts here and elsewhere, I have one question which I haven't seen answered directly - why did the BMW engineers decide on a 2 bar cap in the first place? As noted in previous posts, testing under extreme conditions couldn't get the coolant temp above 126 deg. A 1 bar cap will bring the coolant boiling point up to 130 deg and a 2 bar cap will raise it to 154 deg. The cap has two primary functions - to pressurize the system and to act as the "weakest link" if over-pressurization occurs. So what scenario were the engineers seeing when they decided on a 2 bar cap?

I took a quick look at two ends of the automotive spectrum and found:
Toyota Corolla average 1.2 bar
Ferrari (a couple of models) 0.9 to 1.1 bar

From what I can gather, 1.1 to 1.2 seems to be an industry norm. This seems to be independent of the use of plastic or aluminum radiators. Perhaps someone knows of exceptions? It's doubtful that BMW engines are sufficiently unique to warrant such a departure from industry norms.

BMW has been known for missteps - lifetime tranny fluid (although this might have been driven by marketing), vanos seals, the truly horrible design of the CCV system etc. Could the 2 bar cap be another?
 
#33 ·
The cap may look nice, but to reduce the pressure in the system you need to change the coolant not the cap. I have Evans NPG in it with a zero pressure cap on, (I removed the O-ring on my cap).
 
#37 · (Edited)
I don't get on here as often as I would like, so I just stumbled across this thread.

There is a lot of good information here as well as a lot of misinformation.

When developing our coolant cap we risked possibly damaging two engines, an M52 e39 and an m54 e46. We tapped a pressure sensor into the air space at the top of the expansion tank and data logged system pressures from cold start up to full overheat (temp gauge pegged). To cause an overheating situation we removed the clutch fan and slid a full size sheet of cardboard between the radiator and AC evaporator to block the flow from the electric fan. We ran tests at proper fill levels, low fill levels and high fill levels.

We found the following:

  1. The system runs below 1.0 bar at all normal operating temps.
  2. A lower fill level results in lower pressure and a higher fill level results in higher pressure.
  3. Overfilling the expansion tank beyond the proper cold fill level can result in system pressures over 2.0 bar, at which point the OEM cap will start to vent excess pressure.
  4. Since the cooling system pressure, with a proper amount of coolant in the system (not overfilled), does not exceed 1.0 bar at any normal operating temperature, or even moderate overheating, the use of a 1.2 bar cap will not effect cooling system efficiency in any way.
  5. 1.2 bar equates to a system temperature of approximately 126 deg C (260 deg F), which also equates to a fully pegged temp gauge.

For those that argue that the engine was engineered by BMW to use a 2.0 bar coolant cap and you shouldn't mess with it, consider this. BMW has used a 2.0 bar coolant cap on many models from 1982 to late 2000 something. These models have included iron block iron head engines, iron block aluminum head engines, and aluminum block aluminum head engines, in 4 cylinder, 6 cylinder, 8 cylinder and 10 cylinder configurations. Either they start from the coolant cap and design their engines from there, or the coolant cap is not an integral part of the engine design.

BMW has also moved away from the 2.0 bar system with most if not all current models using a 1.2 bar cap. I think the original thought process was to use a cap that was at the limit of the cooling system rated pressure to give the greatest amount of extreme overheat boil over protection. What they failed to take into account is that a 5 or 10 year old system has lost some of it's original strength, and that reaching a 2.0 bar pressure in these vehicles can cause things to go boom.

By the time my engine has overheated to the point that the bong and dash message have warned me, and my pegged temp gauge has warned me, I would rather pull over, shut off the engine, and let the 1.2 bar cap vent any excess pressure, rather than let the pressure rise until something expensive goes pop and I also need to call a tow truck. I also like the peace of mind that comes from knowing that if while in a hurry I overfill my expansion tank a little, that I don't run the risk of a catastrophic cooling system failure.

I would like to thank everyone again for all the support you have shown, and all the kind words you have expressed to German Auto Solutions. It's been a fun and busy first year with many good things in the works for the coming year.

Gary
 
#39 · (Edited)
#38 ·
Gary, we agree on everything, but your statement, as written in #5 is simply not true UNLESS you also include your assumptions! Really not trying to nitpick, I swear! You can't just make a statement like that..It is probably an oversight, but is really misleading as written. You must state what material you are heating to 126 Degrees C to equal 1.2 bar and how much free air space existed. If you want to say that the E39, when filled to the correct COLD level and containing a perfect 50-50 mix of anti-freeze will be at 1.2 bars of pressure if it reaches 126 degrees C then you have a possibly valid statement. I trust your math, but you can't make a statement without specifying the liquid involved and the air gap. Like you pointed-out earlier in a way-air can compress, a liquid can't. Water can only compress if it changes state; like from a liquid to a solid.

The single most compelling reason in my mind for the lower pressure cap is for those that over-fill their reservoir on a regular basis. Then you stress the entire system to 2 bar to let the excess coolant out so the system can return to under 2 bar. If you fill the system full, the pressure in the system will exceed 2 bar and release ONLY enough coolant to get the pressure back to under 2 bar...say 1.9 bar for arguments sake. The next time you start and run your car, the coolant level will still be too high and the pressure will likely not reach 2 bar a second time, but may reach 1.9 bar quickly and stay there, in theory, forever, or until you somehow lose enough coolant to create the proper air space. For this reason, the lower pressure cap makes a lot of sense, but not to control your temps or what your cooling system wants to do. It is only an advantage for those who A) like the look and B) over-fill their systems regularly.

A system will only be at 1.2 bar @ 126 degrees with a very certain liquid in the closed system. I seriously doubt that it is possible to know exactly that you are at 50-50. Maybe close, but you never have a completely bone dry system to start with. Let's agree to not mix pressures and temperatures- just to know the relationship and the properties of the liquid in the system to make "assumptions". I really appreciate your testing and your thoroughness. Your statement was obviously making some assumptions that some might not get while others would. I especially like your objective evidence of the pressures observed with the proper fill level and the right mix of coolant. The car never exceeded even 1 bar. That was enlightening to me and thank you for that. Beautiful piece BTW! definitely FTW. Link to order please guys.
 
#44 ·
540iman,
Gary, we agree on everything, but your statement, as written in #5 is simply not true UNLESS you also include your assumptions! Really not trying to nitpick, I swear! You can't just make a statement like that..It is probably an oversight, but is really misleading as written. You must state what material you are heating to 126 Degrees C to equal 1.2 bar and how much free air space existed. If you want to say that the E39, when filled to the correct COLD level and containing a perfect 50-50 mix of anti-freeze will be at 1.2 bars of pressure if it reaches 126 degrees C then you have a possibly valid statement. I trust your math, but you can't make a statement without specifying the liquid involved and the air gap. Like you pointed-out earlier in a way-air can compress, a liquid can't. Water can only compress if it changes state; like from a liquid to a solid.
Yes, I was making the assumption of a 50/50 mix of water and OEM BMW antifreeze, filled to the correct cold level, and as measured in an m52 or m54 engine. It was a bullet point list and I didn't see the need to get much into the engineering math behind the measurements. But, you are absolutely correct.

The single most compelling reason in my mind for the lower pressure cap is for those that over-fill their reservoir on a regular basis. Then you stress the entire system to 2 bar to let the excess coolant out so the system can return to under 2 bar. If you fill the system full, the pressure in the system will exceed 2 bar and release ONLY enough coolant to get the pressure back to under 2 bar...say 1.9 bar for arguments sake. The next time you start and run your car, the coolant level will still be too high and the pressure will likely not reach 2 bar a second time, but may reach 1.9 bar quickly and stay there, in theory, forever, or until you somehow lose enough coolant to create the proper air space.
Thank you for pointing this out as I forgot to mention it. If you over fill the expansion tank, the system can run at excessive pressures for many drive cycles, not just one.

It is only an advantage for those who A) like the look and B) over-fill their systems regularly.
I disagree with this statement. Due to any number of reasons your vehicle can overheat. Just a few examples would be broken fan belt, bad clutch fan, bad electric fan, clogged radiator fins due to leaves or dirt, etc. In any of these situations you have ample warning via the dash display and temp gauge that something has gone wrong. I would rather the cap limit the max pressure to 1.2 bar than to have something in my coolant system explode, leave me stranded and cost me much more money.

98-104 degrees C iirc
Correct, the m54s run a little hotter at times due to the ECM controlled thermostat which allows hotter temps at part throttle cruise conditions.

Dragan,
Thanks for that! Now if someone would do the same for the I6.
That is what all the data from my post was taken from.
.
 
#46 ·
I take it as the former too as that weblink you provided also states:

The above form works if you are measuring differential pressure, such as the difference in psi between two points. It also gives the correct answer for absolute pressure, assuming you are measuring psia, which is the pressure relative to absolute zero vacuum
I presume the typical dial guage is calibrated to zero vacuum/pressure at a zero reading.
 
#54 ·
The values that JimLev posted are gauge pressures. That is pressure above atmospheric pressure. To get absolute pressure you would need to add the local barometric pressure at the time he made the readings.

I think the gauge pressures should be the same regardless of the local barometric pressure (or altitude). Since the system is closed, I assume the temperature readings would also be the same.

Bluebee's conversions seem to be correct. The different conversion factors are related by standard atmospheric pressure, 14.696 psi. If you add the value of the first conversion factor to the second, you'll get the standard pressure value (14.504 + 0.192 = 14.696).
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top