Na and K Values in Coolant for UOA Interpretation

[edhackett]

I've analyzed a handful of coolants for sodium and potassium. This data should prove to be helpful in diagnosing coolant leaks.

These values are for undiluted coolant.

code:

---

Brand Type/Color Na K



Havoline DexCool/orange 50 11,100

Zerex G-05/pale yellow 4780 10

Peak Conventional/green 1940 5640

SuperTech Conventional/green 4830 40

Subaru Factory/dark green 4480 5200

---

I'll start a new thread in the Coolant Analysis forum to coordinate having people send in additional samples so that we can build a fairly complete database of coolants.


Ed

 

[Ray H]

How recently did you come by your sample of the SuperTech "conventional/green"? The reason I ask is because SuperTech conventional silicated antifreeze hasn't been available for over a year. It's been an extended-life dexcool "clone" since around June, 2004, though it's still dyed green.

 

[Al]

Great information. The Dexcool is puzzling. Leaks I have seen with DexCool have sodium and Potassium numbers close to the same. But anyway..the good news is that sodium and/or potassium will show up. The bad news is that using an oil with high antiwear sodium compounds and an anti-freeze with sodium = a problem at low leakinjg levels. :

Thanks for posting.

 

[dickwells]

One thing to keep in mind. When my intake manifold gasket was on its way out, my analysis company switched a time or two between reporting the contamination as sodium or postassium. This was with the factory dexcool in my o2 Silverado. So Caveat emptor.

 

[edhackett]

Ray H: The SuperTech conventional is several years old. It's out of my tractor stash. I figured I would go ahead and run it as there are probably people that have it in use.

Al and ****: Yea that is sort of strange. It is possible that the factory coolant formulation is different than the Havoline version. It's also possible that the sealant put in at the factory contains significant sodium.

Can anybody get a sample of the factory DexCool? It would also be interesting to run a sample of factory coolant from a vehicle.

Ed

 

[Al]

quote:

---

Originally posted by edhackett:
Ray H: The SuperTech conventional is several years old. It's out of my tractor stash. I figured I would go ahead and run it as there are probably people that have it in use.

Al and ****: Yea that is sort of strange. It is possible that the factory coolant formulation is different than the Havoline version. It's also possible that the sealant put in at the factory contains significant sodium.

Can anybody get a sample of the factory DexCool? It would also be interesting to run a sample of factory coolant from a vehicle.

Ed

---

You know I would. But I have decided the far better approach is to use either Toyota or Nissan coolant which are better than DexColl ever will be. BTW I had switched to DexCool in my nissan and 2 years later I was replacing the heater core.

 

[ekpolk]

Greetings. I'm here looking for clues about what's up in my just-posted, new UOA. The sample is from my 14 month old G35 which has 37k miles now. The Potassium, which should be at almost zero has blipped up to 7. Not a big jump, but certainly not profile. I was looking to see if anyone knows about the stock Nissan coolant, which is still in, looks very healthy, and has shown no decline in level. Obviously, I'm hoping I don't have a tiny coolant-to-crankcase leak.

 

[edhackett]

Here's the latest batch of cooolants:
Na and K in PPM.

code:

---

Brand Type/Color Na K



John Deere Yellow-Green 9,710 40

TY16034



Motorcraft SCA/Blue 17,500 76,700

FW-16



Texaco AF Coolant Conc. Conv./green 2,100 5,810

ChevronTexaco



Texaco Extended Life HD w/SCA/Orange 550 5040

Equilon (Diluted 50/50)



Prestone Conv./yellow-green 5,280 16

---

Ed

 

[edhackett]

Here's another set of data, values in PPM:

code:

---

Brand Type/Color Na K



Honda Type II ethylene glycol/green 760 7280



NEO propylene glycol/purple 4740 2



NEO KEEP COOL /pink 14700 16



NEO AF Additive /clear 45400 24

---

Ed

 

[edhackett]

The latest set of coolant values, in PPM:

code:

---

Brand Type/Color Na K



Honda OEM 50/50 ethylene glycol/blue 610 7480



BMW OEM ethylene glycol/blue 9800 19



Toyota OEM ethylene glycol/pink-red 4160 4540

---

Ed

 

[Al]

Thanks Ed.

 

[shanneba]

All together now......

code:

---

Brand Type/Color NA K



BMW OEM ethylene glycol/blue 9,800 19

Equilon (Diluted 50/50)Prestone Conv./yel-grn 5,280 16

Havoline DexCool/orange 50 11,100

Honda Type II ethylene glycol/green 760 7,280

Honda OEM 50/50 ethylene glycol/blue 610 7,480

John Deere Yellow-Green 9,710 40

Motorcraft TY16034 SCA/Blue 17,500 76,700

NEO propylene glycol/purple 4,740 2

NEO KEEP COOL /pink 14,700 16

NEO AF Additive /clear 45,400 24

Peak Conventional/green 1,940 5,640

Subaru Factory/dark green 4,480 5,200

SuperTech Conventional/green 4,830 40

Texaco FW-16 AF Coolant Conc. Conv./green 2,100 5,810

Texaco Extended Life HD w/SCA/Orange 550 5,040

Toyota OEM ethylene glycol/pink-red 4,160 4,540

Zerex G-05/pale yellow 4,780 10

---

[ February 19, 2006, 09:41 AM: Message edited by: shanneba ]

 

[edhackett]

The latest additions to the coolant list:

code:

---

Brand Type/Color NA K



VW G12 Longlife EG/pink 10,900 45

Ford Motorcraft Premium Gold G-05/pale yellow 11,480 8

Ford Motorcraft Low Silicate conventional/green 8950 1330

Fleetgaurd Low Silicate 50/50 w/DCA4/blue 1200 3250

Prestone HD Diesel w/DCA2/purple 3210 14

---

I have reason to believe that the sample of Zerex G-05 that I tested was diluted 50/50. Double the values given previously.

Ed

 

[Al]

Looks like origional Nissan coolant is the same as used by Subaru. According to this post the K=74 and Na = 60

http://theoildrop.server101.com/cgi/ultimatebb.cgi?ubb=get_topic&f=3&t=003367

 

[Razl]

Good stuff. This may support my position that the high K values in my wife's 2004 Honda CR-V are not due to a coolant leak because Na values were extremely low. I think the spike in K for my UOAs was from thread sealant I used on replacing the oil sending unit.

 

[Ray H]

Something's seriously amiss with the extraodinarily high potassium readings listed for DEX-COOL. The stuff has no potassium intentionally added (though a low potassium result could be evidence of a minor contamination source at the blending plant). However, if that 11,100 ppm level is correct, it has to be serious contamination introduced into the test sample or equally serious lab technique procedural lapses.