The Verdict Is In: Sasquatch Exists, Oil Filters Spin Off, And I'm

[Guest Gary Cheek]

The ar instead of the er is for Pirate, ahhrrrrr! Or maybe it is for the drawl.

The K is for Kommie. Or maybe the ignorantz.

Canadienne is my misspelling of Canadiene, as in the greatest hockey team that ever was, The Habs!

 

 

Oh yah. I gotcha thar matey. I already heerd enuf of the Canadienne round here. We get ta speak with them regular. Take the trip south across the border often to pick up a load of Sleeman's at the Canada beer store. Also pick up a 40 ouncer of VO at the duty free. Not ta mention gitting a look at the dancers over at the Windsor Ballet. (Code for topless dancers, the wife hasn't found oot they do bottomless too Eh?)

 

Goozer, I stand corrected . Sorry for the mis credited quote. THanks for the friendly correction.

[helicopterjim R.I.P.]

Why, Heli-Jim, if I were light in the loafers , I do b'lieve I'd be on that man like white on rice! Wouldn't you?

 

 

Only if he lets me ride his Ducati as well!!

[Ryland3210]

Ryland, how many turns again to seat the UFI filter?

 

In the case of the older design (the date code on the one that came with the bike was 16.01.03), the cross section of the gasket is larger than the groove, so the gasket will deform and extrude out of the groove as it is tightened. In the case of the newer design (the date code on the one I purchased more recently was 29.08.05) the filter can will bottom out at only 0.80 turns (288 degrees). At that point, the gasket is compressed to 82%.

 

To reach 70% sqeeze, would require 1.32 turns.

 

So in the case of the newer design, I would suggest tightening until the filter can bottoms out or you reach 1.32 turns, whichever comes first. It should be easy to feel the sudden increase in torque as the can bottoms out. Because I feel 82% squeeze is not as much as I would like to see, my own practice would be to tighten until I feel the sudden increase in torque required, and tighten a tad more. I say "would be", because I prefer to use other brands that are designed to provide more squeeze of the gasket before bottoming out, grooves wide wide enough to prevent extrusion, and higher sealing pressures.

 

In the case of the older design, it may not be so obvious as the extruded part of the gasket is crushed.

[Guest ratchethack]

. . . my own practice would be to tighten until I feel the sudden increase in torque required, and tighten a tad more. I say "would be", because I prefer to use other brands that are designed to provide more squeeze of the gasket before bottoming out, grooves wide wide enough to prevent extrusion, and higher sealing pressures. . .

John, just wanted to say again that your conscientious research on this has been much appreciated.

 

You've added considerably to my understanding, and I'm pretty sure I'm not the only one.

[Ryland3210]

John, just wanted to say again that your conscientious research on this has been much appreciated.

 

You've added considerably to my understanding, and I'm pretty sure I'm not the only one.

 

Thanks.

 

I'm a big believer in the laws of physics. It's a good tool to resort to when there is a controversy on a issue vital to the health of our beloved toys that is in need of objective numbers.

[dlaing]

...the filter can will bottom out at only 0.80 turns (288 degrees). At that point, the gasket is compressed to 82%.

 

To reach 70% sqeeze, would require 1.32 turns.

I also appreciate your research, but so far I don't know what to make of the numbers.

Shouldn't more turning increase how much the gasket is compressed or squeezed?

It might also be good to know if the failed UFIs were torqued till the tin hit the chin.

I suspect most of us need a wrench to get the metal to metal contact.

Any idea what the torque is at the various degrees of turning?

It has been a while since I read your other research on the matter, but I recall I came away thinking there was not a big difference between filters.

I guess I'll have to go find the old posts and review.

Thanks for providing us with real physical data

[Ryland3210]

I also appreciate your research, but so far I don't know what to make of the numbers.

Shouldn't more turning increase how much the gasket is compressed or squeezed?

It might also be good to know if the failed UFIs were torqued till the tin hit the chin.

I suspect most of us need a wrench to get the metal to metal contact.

Any idea what the torque is at the various degrees of turning?

It has been a while since I read your other research on the matter, but I recall I came away thinking there was not a big difference between filters.

I guess I'll have to go find the old posts and review.

Thanks for providing us with real physical data

 

Yes, more turning will increase the compression of the gasket. You should be able to understand it better by following the math.:

 

In the case of the UFI '05 version, the groove in the can was 0.225 inches deep.

The gasket was 0.275 thick.

The thread was 3/4-16, or a pitch of 0.0625 inches per turn.

From the point where the gasket makes contact with the block, it can be squeezed another 0.275 - 0.225 = 0.050 inches before the can contacts the block. That requires 0.050/0.0625 = 0.8 turns

At that point the gasket is squeezed to 0.225/0.275 = 81.8%. That is the maximum the filter can design will allow. Good industry practice calls for squeezing gaskets to 70%, which would require more turns than the UFI filter allows. That's why, if I were to use the UFI's, I would turn them until the filter can bottoms out at about 0.8 turns, and apply a tad more torque to hopefully prevent them loosening.

 

By comparison, in the case of a Purolator filter I measured the following:

Gasket was 0.250 thick, groove was 0.175 deep, same thread pitch.

The filter can bottoms out at 0.075 inches, or 1.2 turns.

At that point, the gasket is squeezed to 70%, consistent with industry practice.

Other factors are in the design of the groove, which also favor the Purolator design.

Details on this are on the earlier threads.

With this type of filter design, I find it requires quite a bit of torque to get to one turn, which is my personal minimum. Even with my good squash playing arm, I find this difficult without a wrench unless the filter location is easy to get good leverage on. A filter wrench is a must for me on the Guzzi. With the filters I use, I'll tighten more than one turn to about 1-1/8 turns from gasket contact, unless it bottoms out, which I can feel by a sudden increase in torque required.

 

Hope that answers your questions.

[dlaing]

Hope that answers your questions.

Yep!

Thanks for the clarification.

[Guest ratchethack]

Hope that answers your questions.

It was further clarification f'er me, John. Thanks again.

 

There seem to've been many misunderstandings in this thread. While we're clarifying things, I may've been guilty of feeding some of this. Lest further hysteria, conniption fits, and projectile diarrhea ensue , I throw meself upon the mercy of the Forum and beg y'er kind indulgence of my hastened plea to explain.

 

It may be a semantic problem. When I think of "hand tightening" a filter, I think in terms of tightening "by hand" with a standard wrench and filter socket, as opposed to tightening with a torque wrench. It occurs to me that when others think of "hand tightening", they think in terms of tightening the filter directly with their hands. Of course, getting it adequately tight this way is challenging enough on the Guzzi even with the sump off, but with the sump ON, it's 100% impossible through the manhole cover! I've always used a filter socket on my filters and again, never had one loosen up, even the 2 UFI's I'd used. Even with the sump off, I use a filter socket. But that's just me.

 

Can I imagine someone trying to "hand tighten" a Guzzi filter through the manhole cover WITHOUT a filter socket, using only their fingers? Yes indeed. As I've often observed, stranger things, more absurd than this IMHO, happen every day.....

 

Your work, combined with the observations and direct evidence of others here, is an indication to me that UFI's carry an unwanted margin of risk. I don't like the idea of having to put more turns on the filter than it was obviously designed to take to get a seal. Seems to me that if you tighten it past the point of hard contact (or bottom-out), it has to deform the filter flange, which to my way of thinking would distort it, thereby throwing more risk into this jolly pot o' stew.

 

Of course others will disagree. But for me, this exercise takes UFI's firmly out of future consideration in favor of alternatives known to deliver solid test results and 0% failure rates.

 

BAA, TJM & YMMV

[dlaing]

I went and measured my Walmart filter:

Gasket 12/64"

Groove 9/64"

Squeeze 75%

YMMV (Your Measurement May Vary )

[Guest ratchethack]

Dave, how many turns does it take to get 75% squeeze?

[dlaing]

Dave, how many turns does it take to get 75% squeeze?

I think the most accurate way to determine that will require mounting the filter.

Perhaps a project for later.

But if my measurement is accurate, then the squeezable distance is about 3/64 or 0.46875" (no way I am that accurate)

So, plugging in pitch of 0.0625 inches per turn we get 0.046875/0.0625 = 0.75 turns.

I'll bet with a bit of muscle you could get more than 3/4 turn.

[Guest ratchethack]

Dammit, Dave, now I've gotta take this into the Hatchracket Lab to confirm. More later.

 

Cdr. Hatchracket, Mad Scientist

 

[. . .disappears into Laboratory with various instruments and calculator. . .]

 

[. . .sounds of instruments clinking and calculator keyboard tapping. . .]

 

[. . .somewhat later. . .]

 

OK, here's wot came back from the Lab:

 

(All measurements accurate to +/- .001" via vernier dial caliper/depth gauge, unless otherwise noted)

 

ST3614 Oil Filter

gasket thickness = .208" (+/- .002")

gasket width = .155"

groove width = .155"

TPI = 16

gasket compression per turn = .0625"

fom contact to achieve industry standard 70% of uncompressed dimension = 1 turn

 

NOTES:

 

The thickness of the filter "flange", the part that needs to be strong enough to carry the pressure of the compressed gasket, can be seen, but cannot be measured without cutting the filter apart. The metal part that CAN be seen more directly without cutting the filter (because it covers the flange underneath) is probably AN ORDER OF MAGNITUDE THINNER, and has holes in it, through which the MUCH THICKER flange may be observed underneath. Looking at the edges of the holes inthe flange, it appears to be ~.0625" steel.

 

If I thought (as Tom evidently did in his post #38), that the very thin gauge COVER over the much more substantial flange were in fact, the flange itself, carrying the full pressure of the compressed gasket, I would indeed be concerned. But if this were the case, it would quite obviously flex and deform into uselessness with much of any pressure at all, causing the filter to fail instantly.

 

Since I don't have the oil pressure relief assembly out, I can't make any assumptions about what it would take to bottom the filter in terms of compressed dimension or number of turns. From memory, however, I'm very certain that I can get more than a full turn on these filters from initial contact of the gasket without bottoming.

 

Well then. There you have it.

 

Dave, I don't know wot you were referring to by your post by "75% squeeze"?

[dlaing]

Dave, I don't know wot you were referring to by your post by "75% squeeze"?

Clousseau exits, stage right, into the garage...

[Guest ratchethack]

Dave, note my edit and revisions above. Be careful of getting things bass ackwards here as I did at first.

 

I'd initially been working on the mistaken idea that industry standard is 70% compression of the uncompressed thickness. WRONG.

 

After re-reading John's posts, I now understand that it's achievement of 70% of uncompressed thickness that is industry standard.

 

Semantics again. It's a difference between 30% and 70% of uncompressed thickness.

 

Good thing we've got engineers about the place.

 

1 turn and Bob's y'er Uncle.