Current 5W-20 Roundup

[millerl]

Having just purchased a new 05 Accord with the 2.4L VTEC, I decided to visit the local stores in Oklahoma and gather up all the 5W-20 I could find and send it off to Blackstone for analysis. I overnighted them all to Blackstone via FedEx and they ran all of the analyses the same day they arrived. I also pulled a sample of my Honda factory fill (350 miles on it) and sent it along with them just to see how that compared...

Here's what they found. (I'll try my best to create a proper table.)

code:

---

M-1 Honda Castrol Pennzoil Honda

0W-20 5W-20 5W-20 5W-20 Fact.

GF3/SL GF3/SL GF4/SM GF3/SL Fill



Mo 51 64 34 120 281

Po 1 1 1 0 3

Bo 139 42 17 36 44

Si 2 1 6 2 50

So 5 210 150 0 5

Ca 2445 794 1133 991 2093

Ma 7 1 1 4 4

Ph 737 315 417 356 770

Zi 878 343 429 377 868

---

So, the high silicon in the factory fill is easy to understand...engine seals. Remember, the car had 350 miles on it when I pulled the factory fill sample.

However, what's going on with the sodium levels in both the Honda HG and Castrol oils? The Honda HG bottle clearly states that it is made by Exxon, and claims to be GF-3/SL oil.

I had all of these analyses done at once by Blackstone to try to minimize any relative inconsistencies in laboratory procedures.

I will try to look around the neighborhood and see if I can find any other 5W-20 oils to send off to complete the current roundup.

Any votes on which oil deserves to replace the factory fill when the time comes?

P.S. I did find the new M-1 EP in 10W-30 locally (Wal-mart), so I've sent that off for analysis too, but that's another thread.

 

[sbc350gearhead]

I would ask them to re-test the honda, pennz, and castrol..........they don't look like accurate reports to me.

 

[ruking77]

As you have noted the oem fill Honda 5w20 is made by ExxonMobil. ExxonMobil also has its own brand Superflo 5w20. If you stay conventional, the ExxonMobil Superflo 5w20 would be the ticket especially at a dollar a qt. If you decide synthetic, Mobil One 0w20 or the very new 5w20 extended would be my two choices.

 

[buster]

quote:

---

I did find the new M-1 EP in 10W-30 locally (Wal-mart), so I've sent that off for analysis too, but that's another thread

---

Can't wait to see that. How long ago did you send it off? Your talking about the Mobil 1 Extended Performance 10w-30 right?

I like the M1 of the 20wts bc it has more Boron and Calcium.

 

[buster]

quote:

---

How long ago did you send it off?

---

I'm waiting to see if anyone has sent off a sample of the new Mobil 1 EP. If not, I'm going to.

 

[millerl]

Fellas, the sample went to Blackstone already...I should have results by Tuesday evening to post.

I sent the new M1 10W-30 Extended Performance.

I even asked for a TBN just because I knew we'd all want to know that too.

I changed the oil in my wife's Jeep over the weekend and put in the new M1 10W-30 EP. I'll start to pull UOAs on it as we go. I plan to put this new M1 to the test. If it can't hack the wear and tear, at least it will be in my wife's vehicle and not in mine!

(Doh, I'm gonna catch heck for that!!!)

I'm also going to call Blackstone on Monday morning and ask them to re-run some of those 5W-20 results. They just don't seem right.

 

[buster]

quote:

---

I sent the new M1 10W-30 Extended Performance

---

Awesome, thanks!

 

[G-MAN]

quote:

---

Originally posted by sbc350gearhead:
I would ask them to re-test the honda, pennz, and castrol..........they don't look like accurate reports to me.

---

Agreed. Calcium levels are WAY too low.

 

[millerl]

Fellas,

Blackstone ran all 5 of the analyses listed above again and I have new results to post. The additive levels are HIGHER than shown above, and I will post an updated chart later this evening. Please pardon the delay, but I am awaiting two more analyses (Valvoline GF-4 and Motorcraft GF-4) to add to the chart, and since I can't edit the chart above making a new chart just one more time will save me some typing.

I also expect the results of the new M1 10W-30 EP to arrive via email later today, and I will post those results in a separate topic.

 

[buster]

quote:

---

I also expect the results of the new M1 10W-30 EP to arrive via email later today, and I will post those results in a separate topic

---

You da man. Seriously, thanks for doing this!

 

[millerl]

Blackstone did not email the results back to me today. I will call them tomorrow morning and ask what happened. FedEx definitely delivered the oils to them, so the delay is at Blackstone. Sorry....

This also means we are still waiting on the M1 10W-30 EP results...

 

[ToyotaNSaturn]

Something's up with Blackstone, both of our very recent VOA's are showing VERY low add packs.

 

[millerl]

Having lost my self control waiting on the other test results for the Valvoline and Motorcraft 5W-20 SM/GF-4 oils (not to mention the new M1 10W-30 EP results

) I provide the amended test results that I received from Blackstone Labs. Things look better in these tests, but something really wierd is going on...

code:

---

M-1 Honda Castrol Pennz. Honda

0W20 5W20 5W20 5W20 Fact.

GF3/SL GF3/SL GF4/SM GF3/SL Fill

Iron 1 1 1 0 8

Moly 56 102 42 182 298

Pota 1 1 2 1 4

Boro 161 63 20 53 49

Sili 3 3 7 4 57

Sodi 6 243 148 1 6

Calc 2864 1467 1453 1686 1939

Magn 13 6 6 6 7

Phos 831 590 546 607 835

Zinc 920 675 619 711 886

---

This data continues to show the high sodium counts in the Honda HG and Castrol oils. Even though the Honda bottle wasn't labeled as SM, any chance this is a part of the new SM formulations? The Castrol is definitely a SM oil.

I'm hoping the Valvoline and Motorcraft SM results will help shed some light.

OK, I know Blackstone is a site supporter and all, but I'm not quite sure just how much I trust their results right now. I sent all of these oils at one time to the lab to try to eliminate inconsistencies in methodology from one day to the next. Unfortunately, even though the tests were run on the same day by the same person, some of the results look strange. Lab error, or new unique formulations? I say this also in light of ToyotaNSaturn's Pennzoil HM analysis shown in a separate post. Not trying to start any fights, just asking an honest question.

 

[sxg6]

i remember people saying that mobil bottled gf-4 m1 in gf-3 bottles a few months back.. so maybe that explains the low additives in the mobil1 and pennzoil.. they couldve done the same thing as mobil and were really looking at their gf-4 oils??

 

[millerl]

I have received back from Blackstone the additional Motorcraft (M-C) and Valvoline analyses. I have updated the table from above to include these two additional oils.

code:

---

M-1 Honda Castrol Pennz. M-C Valv. Honda

0W20 5W20 5W20 5W20 5W20 5W20 Fact.

GF3/SL GF3/SL GF4/SM GF3/SL GF4/SM GF4/SM Fill



Iron 1 1 1 0 0 1 8

Moly 56 102 42 182 30 0 298

Pota 1 1 2 1 0 0 4

Boro 161 63 20 53 1 11 49

Sili 3 3 7 4 3 2 57

Sodi 6 243 148 1 0 1 6

Calc 2864 1467 1453 1686 1477 1782 1939

Magn 13 6 6 6 5 6 7

Phos 831 590 546 607 487 561 835

Zinc 920 675 619 711 695 738 886

---

For all you moly lovers out there, you'll be plenty disappointed with the new SM/GF4 Valvoline...big zippo for moly.

So, now the real question comes...how to decide on which one to put into the new 05 Accord 2.4L VTEC?

 

[ToyotaNSaturn]

Use the M1 0w-20. Just about every UOA from this oil a is good one.

 

[Bryanccfshr]

Yes, do look at the UOA's. You will find that the MC performs with similar #'s to the M1. Despite the lightweight results on the VOA.

I am starting to believe that their is much more to motor oil formulations than the add packs we can see on VOA's.

Thanks for the interesting information

 

[Jelly]

quote:

---

Originally posted by millerl:
Any votes on which oil deserves to replace the factory fill when the time comes?

---

Olefin Oligomer 0w-20...

 

[sxg6]

how do we know companies just havent lowered additive leveks for gf-4? i wouldnt blame blackstone

 

[millerl]

Turns out, Blackstone answered the question for us three days later...lab error. To Blackstone's full credit, they (without question and without additional cost I might add...

) re-ran their own analyses on the exact same samples they analyzed the first time around.

Let's look at the Honda HG sample for example:

code:

---

First Second Percent

Test Test Increase



Moly. 64 102 59%

Boron 42 63 50%

Calcium 794 1467 84%

Phosph. 315 590 87%

Zinc 343 675 96%

Sodium 210 243 15%

---

I'm not trying to slam Blackstone at all. In fact, just in the process of performing the tests to generate this single thread I spent over $125 with Blackstone. Rather, my concern over lab error is that with this percentage shift in the results, are the lab results actually scientifically valid enough to draw ANY meaningful conclusions? From the first tests, the oils look very weak. From the second test, they look about what we would have expected, except the very high sodium content. So, is the sodium really there in these two oils? Would that be a desired quality or not? Or is there really no sodium present in the sample but a lab error has introduced sodium into the results?

When we don't actually know the repeatabilty or probability of error rates then it becomes difficult to know what conclusions to draw based upon the numbers.

Said another way, if this was a UOA instead of a VOA, and you had just received the first report showing low additive levels and a very high sodium content of 210, wouldn't you be concerned about your engine status? Or would you just ignore it assuming the engine was fine and that the lab was in error? If you would assume the latter, then why would you even run a UOA/VOA?

We run VOA/UOA to get hard "factual, objective" knowledge. But is that what we actually get? Notice in the results of this one sample above that the error rate is not consistent from one element to the next. I just don't get it.

Perhaps Molakule or Terry could shed some insight into the probable error rates in a typical analysis performed without any lab process error (i.e., the remaining error rate is strictly due to the calibration tolerances of the actual equipment). When a lot of UOA reports have elemental rates well down into the single digits (in some cases 1s and 0s), how statistically valid are those values?

Until now, when Blackstone has sent me a report showing 2ppm of a wear metal, I assume that is exactly what the sample contained. But now I wonder, is it?