Hi guys,

I rarely post, but have observed the following in today's coolant drain from my R230. 2005, M113 engine. The usual car. I thought this might help some folks down the road.

CONDITIONS

1. Following exactly MB service document AR20.00-P-114RVK. That document is attached as a pdf
2. This factory document claims a coolant capacity of 11.5 liters for the M113 as installed in an R230.475
3. A careful removal of coolant, when done exactly as prescribed in this document yields about 7.5 liters.
4. Using compressed air to aid in the further run-out of additional coolant, this yields an absolute total of 8.8 liters of coolant (NOTE: compressed air can be added at the upper radiator hose to coolant reservior bleed line point. Do not exceed 12PSI. Open radiator drain before pressurizing. Pulse the air. Then wait for the liberated coolant to flow back out. It takes time... I repeated the compressed air cycle until no further coolant would materialize out of either block drains or radiator drain.)

Test condition notes: Yes, this result includes opening both block drains and using hoses to recover. Right drain output was no more than 75ml, left drain was about 100 ml. Yes, drains were completely removed to ensure that there was no corrosion blockage.The car was tilted nose-down at 5%. Virtually 100% of the drained coolant was recovered. Spillage was very minor (almost nonexistent). A spillage estimate of 100ml was added to the figure for consistency, since technically there was a tiny amount of coolant on the floor and on the frame rail which could not be measured.


OBSERVATION: After such a run-out (conducted using factory instructions, and then aided by compressed air blow out) on my '05 R230 with an M113.963 engine there is still ~ 2.5 to 2.8 liters of old coolant left in the total system, I suspect mostly in the heater core.
Exact coolant total capture was calculated using a professional quality HVAC refrigeration scale to measure the liquid by weight. My refrigerant scale is accurate to +/- 1gram. The specific gravity of MB coolant 50/50 mix is 1.06, which works out to 2.337 pounds per liter. or 1.06 kilograms/liter. My container had 19.67 pounds (tare). Not too hard to figure from there.

WHY DOES THIS MATTER?

It has been suggested on this forum that a double refill using clean water to twice dilute the remaining coolant is an acceptable way to achieve practical replacement with low cross contamination. But is this true?

I have done the math and it indeed appears that two full flush cycles of 9 liters clean water each time will result in a left over mixture of better than 95% water and less than 5% old coolant. (overfill coolant reservoir during flush cycles to achieve 9 liters clean water in each flush cycle. Don't overfill at the final coolant addition of course! :-) What follows is based on a sure knowledge that the maximum observed drain out is no more than 8.8 liters (without manipulating the heater core valve, etc., something not provided for in the MB drain/refill document)

Here's the breakdown of the calculation,

• Initial State: I start with 11.5 liters of 50/50 coolant water mix, aka (solution B) which contained 5.75 liters of used coolant and 5.75 liters of distilled water (50% each). Standard R230 fill for the M113.963
  
• Target State: I want the final solution remaining in the radiator (solution A) prior to the final fill to be at least 95% water, meaning 5% or less exhausted coolant mix. For 11.5 liters, this would be 0.575 liters of exhausted coolant mix.
  
• Dilution Process: In each cycle, 9 liters of clean water are installed, mixed and then removed. This means the amount of exhuasted coolant/water decreases with each cycle.
  
• Calculation:
  
  The remaining fraction of the original mixture after each removal in each cycle is (11.5−9)/11.5=2.5/11.5 which is approx 0.2174.
  To find the number of cycles (n), we use the formula: B_n=B_0times (textfractionremaining) n Where B_n is the target amount of B, and B_0 is the initial amount of B. 0.575=5.75times(0.2174)n
  
  Solving for n, we get approximately 1.51 cycles. Since we don't do "fractional cycles" this means you would need 2 complete physical dilution cycles to achieve a solution of (better than) 95% pure distilled water left in those 2.8 liters of undrainable coolant.
  
  OBSERVATIONS/ACTIONS
  
• A.) The idea of a doubled-up clean water flush, as proposed elsewhere appears VALID. One needs only two complete clean water flushes of the cooling system, after a maximum practical drain-out. When one does this, and as long as one gets close to 9 liters with each drain cycle, this is more than adequate to achieve virtually fresh coolant in your system.
• B.) It is virtually pointless to take the time to access the block drains. Since less than 200ml of coolant mixture came out of both drains combined, the dilution method is faster and much easier. Adding 200ml of old coolant mixture to the formula makes virtually no difference in the final re-dilution ratio prior to fresh coolant refill. WHY BOTHER?
• C.) If someone wants to be extremely **** about skipping all that block-drain fun, then a triple flush is for-sure enough without draining the block. Triple flush will result in better than 99% clean water in the what remains in the cooling system, after the drain off from flush #3 and prior to adding the final coolant solution.
• D.) When I say "Clean water" I mean either known soft tap water (per MB document) or distilled water. Check with your local water authority for your annual water quality report. The exact mineral content of your water is included in such a report and most are avilable online. I am fortunate. My local water is both clean and soft.

Hope this helps some folks.

CBG

 

With old iron blocks, you would get some nasty rust/sludge and flushing was necessary. With aluminum engines, coolant stays pretty clean, but additives will be depleted. I agree with MB engineers that a drain and refill is sufficient.

 

Yeah, agree in principle. I wish is was that simple in the real world. In the real world there is one major caveat that will bite many DIY and older-automobile owners. That is if one is CHANGING coolant classes. (I am)

Coolant class limitations exist for a reason, and one key reason is their interactivity (or lack thereof) with anti-corrosion chemistries in other classes of coolants. LIke everything else, coolants made nowadays are different than what came in your typical early 2000's R230.

In the case of a coolant class change, you want minimal coolant type interaction, which is why MB requires a complete FLUSH if moving away from Q1030002 (the yellow stuff) that came in a 05 R230 like mine (for example) and into either Q1030005 (the MB branded pink stuff) or Q1030004 the blue stuff (generically available as G048).

Since they put the yellow stuff in back in the 00's, and then later the blue stuff, one can surmise that a dealer serviced vehicle originally equipped with the yellow stuff and now more than 15 years old SHOULD have the blue stuff in it, (OR the pink stuff.) Either is o.k. But neither can mix with the other.

If one wishes to move from pink to blue for DIY reasons, a full flush is required. If one wishes to move from Blue to Pink because one believes that MB branded coolants are magical, again a full flush is required. If one has yellow, it is discontinued, and therefore a full flush is again required.

As you can see from this very brief review, there are many reasons for a DIY to accomplish what amounts to a complete flush.

Hence my post.

Cheers!
cbg

 

Finishing up this thread I started - I want to say that in the real world, after doing this process by the numbers I first posted, I was not fully satisfied with the results. Therefore, I decided to do the full three flushes with clean water, leaving the remnant of the 3rd flush - approximately 3.5 quarts - as the water mix with the new straight coolant.

I did this because the second flush came out with more remaining old coolant color visible and taste discernable in the real wolrd than the formula I worked suggested should be present.

It was not terrible, but it was detectable.

In contrast, after flush #3 and a run to operating temperature, the ~8 liters of clean water that drained out from the third draining was materially clear and had no discernable taste of coolant left in it.

From there, fill and finish was routine, following the MB sheet recommendations for 100% coolant fill quantities. After putting straight coolant in to the MB quantity, one need simply fill to normal level with additional clean / distilled water in the usual fashion.

cbg