5w20 Havoline 4850mi, Subaru 2.5L w.90k mi

Status
Not open for further replies.
Joined
Apr 30, 2003
Messages
7,076
Location
Ontario, Canada
Apr thru Aug, hottest summer in memory here, I decided to try 5w20 in my car. I tried lugging the engine as much as the auto transmission would let me, but still I could not bring out much lead.

The engine felt more responsive and powerful with this oil, and there was only a little more noise. It sounded like the old QS winter blend after it had sheared down to 9.2 cSt, and while climbing mountains in the heat during the road trip. Fuel economy dropped a tad after I changed the oil back to Pennzoil 5w30. The reason I put in clean oil was because I had one time where the timing belt tensioner failed to pump up quickly. This makes a rubber slapping against plastic sound for a few seconds as the loose belt apparently slaps against the cover.

Iron in this UOA is 3 ppm, if you subtract the 2 ppm from the VOA. BUT, the drain plug magnet collected three times as much powder as last summer's road trip of almost twice the mileage, where the UOA showed a higher iron reading.

Why does the UOA not reflect the amount of iron powder caught by the magnet. Is the magnet removing ALL the iron from mechanical wear? The ppm/mile in the UOA for this summer interval was the same as last year's, but the magnet had three times as much stuff on it. Could the UOA only be detecting iron corrosion, with my magnet removing all the wear particles?





Code:

2001 Subaru Legacy 2.5L non turbo



7926km Wal Mart recycled 10w30, 11123km Mobil 1 0w20/0w30 mix,

4500 km Redline 5w20, 12,000 km Quaker State Winter Blend 5w30,

9000 km QSWB5w30, 7800km Havoline 5w20



Jul-Oct, Dec-Jun, Nov-Feb, Oct-Dec, Dec-Apr2006, Apr-Aug 2006, (VOA)



Fe 5.3 12 6.6 6 7 5 (2)

Cr 0.8 1.4 1.1 0 0 0 (0)

Ni 0 0 0 0 0 0

Ti 0 0 0.1 -- -- --

Ag 0.3 0 0.5 0 0 0

Al 2.8 6.4 7.1 4 8 3 (1)

Pb 0 1.6 4.7 1 1 1

Cu 3.0 1.9 2.8 2 1 3 (1)

Sn 0 0 0.1 0 0 0

Si 9.9 11 14 7 10 8 (4)

Na 0 5.7 15 9 9 4 (3)

K 0 0 1.5 4 0 1 (1)

B 3.0 133 16 -- -- --

Ba 0.4 0.2 1.1 -- -- --

Mo 1.2 67 560 71 63 343 (482)

Mg 11 29 10 -- -- --

Ca 1969 2709 2805 -- -- --

P 850 771 1203 -- -- --

Zn 970 1009 1329 842 804 688 (823)

S 3052 1938 5022 -- -- --

Mn 116 77 3.0 -- -- --

V 0 0 0.3 -- -- --

Fuel 0 -- -- 0 0 0

Gly 0 -- POS 0 0 0

Oxi 42 82 114 112 79 --

Nit 31 63 14 38 46 --

Sul 52 73 78 62 79 --

ZDDP -- -- 39 -- -- --

TBN 4.40 5.22 5.48 2 2 --

Kv40 -- 52.8 -- -- -- --

Kv100 10.4 10.2 9.2 9.2 9.5 8.0



VI -- 185 -- -- -- --

H2O 0 0 0 0 0 --




Ok, so the CODE thing isn't working so well, but it's good enough to see the recent results.

oil filter was Supertech 3593, oil was Havoline 5w20 SM, air filter has about 25,000 mi of use. Magnetic drain plug is a couple of rare earth bar magnets sitting in a couple of pockets I drilled into the drain plug. One magnet has N pointing up, while the other has the S pole up, so there's a really strong field set up in the 1/8" gap between the two.

The soft metal wear looks really low this time, despite the low viscosity, but there was much more iron slime on the magnet than I expected after this distance. The last time I got this surprise from the magnet was when I used Auto Prix 10w30 SM for 5000 km one summer. I wonder if the oil chemistry is at fault.

If I can find a xw20 that doesn't allow quite so much iron wear during short trips, I'll be very happy. I liked the extra power and fuel economy.

Oil consumption with 5w20 was minimal, just like usual. It might use one quart in 25,000 miles. No visible metal bits were found in the pleats of the oil filter.
 
Oil analysis only measures particles smaller than about five microns, along with dissolved ions. Larger wear metal particles don't show up....

I'd believe the magnet and not the analysis - I think you're seeing severe rubbing wear with the SAE 5w-20.

TS
 
Quote:


Oil analysis only measures particles smaller than about five microns, along with dissolved ions. Larger wear metal particles don't show up....

I'd believe the magnet and not the analysis - I think you're seeing severe rubbing wear with the SAE 5w-20.

TS




If that were the case I think he'd be seeing much higher iron levels in the UOA.
 
Quote:


I tried lugging the engine as much as the auto transmission would let me, but still I could not bring out much lead.




Lugging may aslo lead to higher iron from valvetrain. Here it is not. Looks like Havoline moly helps a lot to fight against valvetrain wear.
 
Gman,

In these cases, I'd normally recommend doing a particle count to see if the thinner lube is generating significantly more wear. Spectrographic analysis is sort of a blunt instrument when trying to pin down specific types of wear mechanisms.

Ted
 
Interesting. Ted if you believe what you said then true component wear on a larger scale is not being detected via spectrographic analysis. Sounds familiar....RL?
 
I'm remembering that Amsoil chart with results from a 4-ball wear tester, that showed Havoline and two others' 10w30 with 3x as much wear as the rest.

This oil ended up 8.0 cSt, only 13% thinner than the QS (9.2 cSt) I used last summer. And the soft metal wear was very low, so I'd say it was not a viscosity problem. Maybe the HTHS was much lower. I don't know.

I've had a drain plug magnet for five years, and consistently see more buildup during winter use. This summer had about the same buildup as last winter, which is unusual. The only other time I've seen so much buildup is with the cheap Auto Prix 10w30 over 5000 km of summer driving.

Trips for this past UOA were medium length, or about 10 miles, the same as last winter's UOA.

The engine is making a slight ticking sound, probably from valve clearance. Perhaps that is partly to blame for increased iron on the magnet.

I'll probably use up the rest of my Havoline by mixing it with the SJ rated Diesel 10w30 I can get at Canadian Tire, wihich has 1200 ppm or so of phosphorous.

I have a microscope that can see the rough shape of particles of around 3 microns and bigger. It might be interesting to look at what the magnet catches.
 
Quote:


Interesting. Ted if you believe what you said then true component wear on a larger scale is not being detected via spectrographic analysis. Sounds familiar....RL?



Ted better quite before the hole he's digging gets deeper.
laugh.gif


Back to the original poster. Like UOAs, I highly recommend you don't judge your results based on one OCI's drain plug collections. Something could actually be wearing abnormally, and the oil will unjustly get the blame. Could aalso be that the oil filter media has ruptured, or the bypass valve is defective.
 
I thought I had checked the filter, but after looking at my records, I notice I had cut open the previous one, and left on the current one.

I'm wondering a little about the drop in moly and zinc. Is this from normal lab variance, or have the zinc and moly become bonded to the surface metal, worn off into little bits, and then caught in the oil filter? 274 ppm of stuff has gone missing. In 4 qts, that would be 1036 mg. But there wasn't that much on the magnet, and if not all of it was caught by the magnet, then the UOA should have shown way more iron.

-maybe the drain plug magnet was covered in a paste of iron bonded with MoS2, and that added volume and darkness.

-maybe there was a buildup of iron powder in the oil pan that didn't drain out the last time I changed the oil, and the thinner Havoline was better at stirring it up off the bottom. But last time, I changed the oil when very hot, just minutes after shutting down, so everything should have been suspended.

or... maybe during hot summer, the oil temps are nearing the Curie point of the magnet, and it isn't collecting the iron as well. But then I'd see higher Fe in the summer UOA's.



The main things I've learned this OCI
--5w20 will not ruin the bearings, but does improve mpg
--drain plug magnets might significantly affect Fe in a UOA.
 
It looks like you ran Redline in this engine a while back? Did it show lots of ferrous materials on the magnetic drain plug?

Buster,

What I'm saying is that the observations about the drain plug and the UOA data seem to contradict one another. In a case like this, I'd like to collect more data before deciding that continuing to run an SAE 5w-20 is a good idea.
So the idea of doing a particle count or even ferrography (a more sophiscated technique) would be on my list of things to do.

TS
 
too slick I agree and will add that the drian plug can also hold more sludge and LOOK worse than last drian.

Also unless the plug is weighted there is IMHO alwways a error in memory in that I think there was less fuzz last time.
Or thinner oil different chemistry from last chnage is cleaning up some small amount of sludge or debris which sticke to plug and LOOKS like more stuff.
bruce
 
Thanks for the clarification fellas. I'd like to see the results minus the drain plug.
 
I actually do have a photo of the drain plug from last summer. There was surprisingly little iron for 7500 miles. Primus has an idea, when he mentioned lugging. Low speed means more % of time in boundary lubrication, which means more wear. I tried Redline twice, and only remember noting that it produced a normal amount of iron. I was expecting some kind of miracle. This was a mix of 5w30 and 5w20, then straight 5w20. The 5w20 was over winter, so the iron on the magnet was expected.

Correct that I can't prove that there wasn't any iron powder that was cleaned up with this oil. I'd have to do a Patman Flush to be sure of what's going on. All I can say for sure is that the drain plug magnet does seem to affect Fe readings in the UOA. I might try it without the magnet next year.

My best guess is that most of the magnetic powder comes from cylinder bore wear. Valvetrain wear should not get worse in the winter, since the cold oil is very thick, and should protect very well. Bore wear should be much worse, since the first three minutes of operation the oil gets washed off the walls, and it's basically dry. There's also more surface area there to give off iron. Maybe the iron was from the surface roughness being polished down to a slightly lower level, and it was a one-time thing.
 
Hehe, I think it's funny that the term "Patman flush" still exists here.
smile.gif


I don't even do the Patman flush myself anymore, not even when I switched my wife's Civic from GC to regular Syntec.
 
Apparently there is a way to measure big particles.

Particles Counts:
>5__________44864____81592____90038_____18349
>10_________3703______4273____22902______3508
>15_________318_______608______1050_____1255
>20_________109_______350_______136______523
>25_________42________252_______86______239
>50_________5_________118_______18_______27
>75_________1__________90________6_______6
>100________0__________64________2_______1

aehaas

maybach AEHaas
 
Quote:


MoS2, formed locally when MoDTC is exposed to extreme temperatures at wear zone.


Very good. But your magnet theories didn't lead one to expect that level of understanding.
 
"I've had a drain plug magnet for five years, and consistently see more buildup during winter use. "

This is a good observation. This to me indicates large particles and may represent cavitation damage from the cold weather start ups.

I also think that large particles are not picked up by UOA, only dissolved metals. One would think that the dissolved metals would be higher with increased amounts of particulates but who knows. I have not seen enough evidence to make any conclusions.

aehaas
 
Status
Not open for further replies.
Back
Top