BikeTechReview.com

  • Increase font size
  • Default font size
  • Decrease font size
Home Forum
Welcome, Guest
Username Password: Remember me

Comparison of chain lubes
(1 viewing) (1) Guest
  • Page:
  • 1

TOPIC: Comparison of chain lubes

Comparison of chain lubes 5 years, 11 months ago #21680

  • tigermilk
  • OFFLINE
  • Expert Boarder
  • Posts: 107
  • Karma: 0
I recently completed a set of tests to see if there was a difference in lubricants. The lubricants tested included lubricants directly marketed to the cycling market (e.g., Pedros or Pro Link) as well as general household products (WD-40, 3-in-1). A leap of faith is taken in terms of this testing in that it is assumed that the materials sliding against each other don’t matter to the degree as the lubricant. A 440C ball was slid against an Inconel sample on a pin-on-disk tribometer. Dry sliding tests were performed to establish a baseline (dry friction above 0.60). A wear track was developed, and this wear track was used for all tests rather than starting a new track. This was done to keep the effect of roughness in the testing. After a lubricated test, the Inconel sample was cleaned and stripped of the lubricant. A dry run-in test was performed to verify the cleaned sample was free of lubricant and that the dry sliding friction was at least the initial dry value (i.e., over 0.60). Contact force was set at 5 N. Tests were run wet, immediately after application of the lubricant (later tests will including wiping off excess and allowing lube to "dry").

Below are the results when the lubes were performing their best. Some started to break down significantly. Which performed the best? 3-in-1, a heavier household oil (would attract dirt), was darn good, but the one that took the prize was, of all things, WD-40. Pro-Link rated 3rd in my tests, with the others jumping up in friction near the conclusion of the tests. Should be noted that the expected range below is simply 3-sigma min/max values. I doubt the first 2 on the list would go as low as written. Just the large deviation (in other words, poor stability) brings the values down that low.

Lube Average Deviation Expected Range
Pedros 0.1381 0.0163 .089-.187
Purple Extreme 0.1119 0.0140 .070-.154
3-in-1 0.1028 0.0017 .098-.108
WD-40 0.0961 0.0007 .094-.098
Pro-Link 0.1106 0.0015 .106-.115
KryTech Wax 0.1264 0.0105 .095-.158

Re: Comparison of chain lubes 5 years, 11 months ago #21682

  • toigtfiw
  • OFFLINE
  • Junior Boarder
  • Posts: 35
  • Karma: 0
Those are interesting results!

You mention a 5N contact force - what sort of pressure does that correspond to? What are typical bicycle chain contact pressures, anyway? [I know, I should do the measurements and calculations myself, but my calipers and bike chain are ten miles away.]

Can I talk you into testing my favorite chain lube? It's just Mobil 1 5W-30 motor oil. I mix it with mineral spirits for better penetration, but that part evaporates pretty quickly.

Thanks for sharing your experiment,

Eric

Re: Comparison of chain lubes 5 years, 11 months ago #21683

  • JV
  • OFFLINE
  • Platinum Boarder
  • Posts: 1101
  • Karma: 0
Thanks for sharing!

Pedros makes a lot of lubes, which one did you test?

Re: Comparison of chain lubes 5 years, 11 months ago #21684

  • tigermilk
  • OFFLINE
  • Expert Boarder
  • Posts: 107
  • Karma: 0
Hertz stress (sphere on plate) is around 160 ksi for the experiments (minimal, if any, wear on the slider so the contact stress was pretty consistent). I was going to try to work out the stresses involved with the chain, but 1) just haven't gotten around to measurements and 2) hard to tell since not all chains are made the same (i.e., thickness of the plates at the pin ends). I'm making the leap of faith that the lube performance will be comparable as load varies.

Pedros was "Extra Dry All-Purpose" aerosol. The irony is that it was the only one that in its description referenced an ASTM standard (3233).

I plan on testing some motor oil. I used it for a few TTs. Yes it attracts dirt, but it certainly stays slippery.

Re: Comparison of chain lubes 5 years, 11 months ago #21685

  • JV
  • OFFLINE
  • Platinum Boarder
  • Posts: 1101
  • Karma: 0

Re: Comparison of chain lubes 5 years, 11 months ago #21699

  • tigermilk
  • OFFLINE
  • Expert Boarder
  • Posts: 107
  • Karma: 0
JV wrote:

Yes, but that's why you do testing rather than go by opinions of one person. Data trumps opinion!

On the chain bearing stresses, hard to really nail that down since power is all over the place. But the bearing stress should be:

Fchain=Fpedal*Crank_Length/Chainring_radius

where we assume Fpedal=average pedal force from power, or

Fpedal=P/omega/Crank_Length (omega the angular velocity of the crank)

So chain force should simply be

Fchain=P/omega/Chainring_radius

Bearing stress on the chain is Fchain/Dpin/h_plates

My Wipperman chain has a pin of 3.6 mm and plates of 1 mm (so h_plates=2*1 mm=2 mm). Assuming a cadence of 90 rpm (omega=3*pi rad/sec), P=300 W, and chainring radius of 107 mm leads to bearing stresses on the order of 41.3 MPa (6 ksi). I'll have to scale back the contact force on my experiments, but I can only get down to 1 N, which still leads to an almost order of magnitude higher contact stress than the bearing stresses associated with the chain. Clearly 6 ksi is well below the yield (let alone ultimate) of the material, so strength is never an issue. Just those pesky wear particles forming and degrading the material.

Re: Comparison of chain lubes 5 years, 11 months ago #21703

  • JV
  • OFFLINE
  • Platinum Boarder
  • Posts: 1101
  • Karma: 0
tigermilk wrote:
Data trumps opinion!


I agree

Just making sure you've seen the various opinions, as that sometimes gives ideas as to what (else) to test. Thanks again for testing and posting.

Re: Comparison of chain lubes 5 years, 11 months ago #21717

  • toigtfiw
  • OFFLINE
  • Junior Boarder
  • Posts: 35
  • Karma: 0
tigermilk wrote:

My Wipperman chain has a pin of 3.6 mm and plates of 1 mm (so h_plates=2*1 mm=2 mm). Assuming a cadence of 90 rpm (omega=3*pi rad/sec), P=300 W, and chainring radius of 107 mm leads to bearing stresses on the order of 41.3 MPa (6 ksi). I'll have to scale back the contact force on my experiments, but I can only get down to 1 N, which still leads to an almost order of magnitude higher contact stress than the bearing stresses associated with the chain. Clearly 6 ksi is well below the yield (let alone ultimate) of the material, so strength is never an issue. Just those pesky wear particles forming and degrading the material.


Cool. Thanks for the estimate! Since peak pedal forces are quite possibly an order of magnitude higher than the average, 1 N contact force might well be appropriate. I sure wish we had a resident tribologist here on BTR - I'm sure we'd have lots of questions for them!

My experiences with motor oil have been pretty good. Regarding cleanliness, I find that the chain stays clean a lot longer if I wipe the chain clean with a soft rag after the mineral spirits have had a chance to evaporate a bit. I just wonder if I shouldn't be looking for a straight 30W rather than the 5W-30 I use. Does straight 30W even exist in a synthetic oil? Where's that tribologist?!

-Eric

Re: Comparison of chain lubes 5 years, 10 months ago #22156

  • tigermilk
  • OFFLINE
  • Expert Boarder
  • Posts: 107
  • Karma: 0
As a followup, after doing some additional work related testing, I have done a couple more "off-the-clock" tests do determine longevity of various lubes. To mimic real-life more (lubing chains and then wiping them down), lubes are being applied to the surface and then wiped down, leaving a thin film of lubricant. Sample cleaned with solvents to degrease the surface between tests. I started with WD-40 first since it performed best in terms of friction. Unfortunately the sliding interface was lubricant starved and some time and consequently friction and wear skyrocketed. A significant wear track was left in the specimen. Went to 3-in-1 next which performed very well. I stopped the test after 24+ hours. The specimen slid for over 5.7 miles (120,000 laps at a ~24.4 mm diameter) and friction was rock solid in the .10-.11 range. In addition, there was virtually no wear on the specimen. As a boundary lubricant, 3-in-1 is incredible. It makes sense since it is a fairly heavy oil, and any wear particles generated would be carried away better. The surface of the specimen had only a faint circular track, barely scarring the surface. Currently Purple Extreme is under the microscope.

Just too bad dirt would be attracted to the 3-in-1 heavier oil, though dirt certainly doesn't have the stiffness necessary to induce significant wear. Field testing could demonstrate if chain wear is occuring more rapidly or if the dirt is simply unsightly.

Re: Comparison of chain lubes 5 years, 10 months ago #22161

  • andy497
  • OFFLINE
  • Gold Boarder
  • Posts: 289
  • Karma: 0
One more thought, if you're still in the testing mood. Have you tried chainsaw oil?

That one's my favorite. It's already designed with the penetrator as far as I know, and of course one would assume it's good for chains

As far as the dirt, I put a copious amount on (not exactly expensive), then wipe off well. That seems to work ok.

Re: Comparison of chain lubes 5 years, 10 months ago #22173

  • triguy42
  • OFFLINE
  • Platinum Boarder
  • Posts: 357
  • Karma: 0
andy497 wrote:
As far as the dirt, I put a copious amount on (not exactly expensive), then wipe off well. That seems to work ok.


Why would you want to put a copious amount of dirt on your chain and then wipe it off???? Great results tigermilk, it pretty much makes sense that WD40 didn't survive a "life test" very well...considering the fairly high level of volatile solvents. I wonder if it would be good enough for a raceday use for friction reduction? I've been using Pro-Link for years and it definitely does a good job. Wattage-wise it doesn't seem like there's a huge gap in performance between Pro-Link, Purple Extreme and the WD-40/3-in-1 though.

Re: Comparison of chain lubes 5 years, 10 months ago #22179

  • tigermilk
  • OFFLINE
  • Expert Boarder
  • Posts: 107
  • Karma: 0
I did the Purple Extreme test and it bombed quickly. Significant wear path was carved out. Thank goodness it was run overnight rather than working hours. The screeching would have been unbearable to those around. I expect pretty much all the marketed "bike lubes" will fare poorly for long-term wear. Have to give up the machine for awhile for "real work" but when available again I'll test some other heavier lubes (e.g., motor or chainsaw oil).

New chain 5 years, 10 months ago #22263

  • toutenhoofd
  • OFFLINE
  • Expert Boarder
  • Posts: 109
  • Karma: 0
Okay, this all interests me quite a bit, but I'm wondering what you think is the best "race day" chain solution. Should you start with a brand new chain, clean the factory grease off of it with solvent, and then use something like 3-in-1, WD40 or ProLink?

Thoughts?
<a class="postlink" href="cyclingtechnology.com">http://cyclingtechnology.com -- Lightweight & High Performance Cycling Parts and Accessories

Re: New chain 5 years, 10 months ago #22279

  • tigermilk
  • OFFLINE
  • Expert Boarder
  • Posts: 107
  • Karma: 0
toutenhoofd wrote:
Okay, this all interests me quite a bit, but I'm wondering what you think is the best "race day" chain solution. Should you start with a brand new chain, clean the factory grease off of it with solvent, and then use something like 3-in-1, WD40 or ProLink?

Thoughts?

I haven't tested much grease. In fact, the only grease I have truly tested is Braycote 602EF at work. It's hardly a grease you'd want to use on the bike. At $2k+ for a 1 pound supply, it's a little on the expensive side. I also tested it with a different set of materials (440C on 15-5PH). However, it was VERY effective at maintaining friction, even when artificial debris was mixed in. I wouldn't expect too much difference in friction coefficient if I were to test using the same materials as I did for the liquid lubes (as they all serve the same boundary lubrication purpose - to keep the metallic surfaces out of contact). Sliding friction values were around 0.15 (deviations for tests were consistent in the .003x range) using the Braycote. But the Braycote did perform exceptionally well over the long-term, lasting well into the hundreds of thousands of revolutions (one test taken to 1.2 million cycles).

From my experience, the grease that comes on new chains doesn't have a long life to it. It's good for a few rides as a general lubricant, but that's it. For performance, in the past I've used motor oil, though based on my testing I may just stick with 3-in-1. Has all the properties I'd want in a race-day lube - life and low friction. While I typically use Purple Extreme for everyday chain lubrication, I have noticed that that particular lube is hit or miss. I can lube each link, wipe it down a few hours later, and consequently hear some squeaks, a clear indicator of dry sliding.

Re: Comparison of chain lubes 5 years, 7 months ago #23132

  • tigermilk
  • OFFLINE
  • Expert Boarder
  • Posts: 107
  • Karma: 0
For a practical update, I applied some 3-in-1 oil to my chain a bit over 1000 miles ago. The only cleaning has been wiping the chain down a couple of times in that span. Been running in a dry (albeit humid) and relatively clean environment (relatively clean roads), and the chain hasn't squeeked at all. Seems lately the lubricant may be starting to starve a little bit, but I'll certainly take the low friction properties, long life, and ultra cheap cost of 3-in-1 now over boutique bike lubes.

Re: Comparison of chain lubes 5 years, 2 months ago #24245

  • Neal
  • OFFLINE
  • Platinum Boarder
  • Posts: 420
  • Karma: 0
Hello Tigermilk and All,

Did you ever get around to finishing the friction tests of the chain lubes?

You have published the only data I can find on chain lubes. Thanks.

Seems like there should be a lot of data from oil companies and mechanical engineering applications.

I am wondering if some super synthetic nano microsphere product has come along since your original tests that offers practically no friction.

Or DuPont Teflon motorcycle chain oil.

Or chains with ceramic pins that require no lube and are very light.

You said the Purple Exteme bombed quickly on long term wear test.

"I expect pretty much all the marketed "bike lubes" will fare poorly for long-term wear."

Did you ever test some of the others for long term wear?

In your opinion is long term wear relevant for normal bike use where we oil/wipe our chain before each ride?

I am currently using Pro-link and it seems to work OK but would switch if there was a better choice.

Which lube do you prefer after your tests?


Cheers,

Neal

Re: Comparison of chain lubes 5 years, 2 months ago #24268

  • tigermilk
  • OFFLINE
  • Expert Boarder
  • Posts: 107
  • Karma: 0
I haven't done any additional lube tests since the summer. I'd post what I finished with, but I can't upload a file to BTR larger than 500kb. I didn't test any more for long term wear since in my opinion I had a winner - el cheapo 3-in-1 oil. I've been using that on the "daily driver" bike to the tune of 800-1000 mile "oil changes". Clean the chain, let it dry, lube liberally with 3-in-1, set overnight, wipe down, ride. Generally I'll wipe down after each ride the first couple of days and then less often.

I certainly don't have the discipline to lube my chain before each ride, and nor should I. I don't lube anything else before each use.

The only thing I've tested so far as good for life as 3-in-1 is Braycote 602EF grease (developed for applications in vacuum). But at a couple of thousand dollars per pound, I doubt it's coming to a bike shop near you anytime soon.

Re: Comparison of chain lubes 5 years, 2 months ago #24348

  • Neal
  • OFFLINE
  • Platinum Boarder
  • Posts: 420
  • Karma: 0
Hello Tigermilk, Toigtfiw, and All,

I looked at some other lubrication tests:

[url:152jwxin]neptune.spacebears.com/cars/stories/oil-life.html[/url]

Testing synthetic motor oils seems to show the benefit of Mobil 1 and AMS Oil.

--------------------------------------------------------------------------------------------------

Synthetic oils were originally developed more than 50 years ago and became widely used in jet engines. Less than -120ºF ambient temperatures, 60000 shaft rpm, and 500º+F exhaust temperatures proved too much for conventional oils. Synthetics were created specifically to withstand these harsh conditions and to date every jet engine in the world uses synthetic lubricants. Amsoil introduced the first synthetic oil for automotive use in 1972 and have continued to be at the leading edge of development ever since. Mobil-1, undoubtedly the most recognized name in synthetics, was introduced in 1976. Many companies have jumped on the bandwagon and have since released synthetic lubricants for automotive use and all are becoming increasingly popular for their superior lubricating properties, superior ability to flow at cold temperatures, and their ability to withstand high temperatures for extended periods of time. Several new cars including the Porsche 996 and the Chevrolet Corvette LT-1 are delivered with synthetic oil in the crankcase and require synthetic oil use throughout the life of the car.

=======================================================================

Mobil 1 just started offering a new oil 0W-20W that has a lower viscosity.

I am going to try it for my chain and perhaps also my Speedplay pedals.

The Speedplays will require frequent lubing with the lighter oil.

Mobile 1 by the quart is much cheaper than most 'bicycle oils'.

If the viscosity tuns out to be a bit thick I will cut it with odorless mineral spirits (sulphur removed) for better penetration as noted by Eric.

For future planning this looks promising:

[url:152jwxin]www.nanotechwire.com/news.asp?nid=5455[/url]

ApNano Materials, Inc., a provider of nanotechnology-based products, today announced that recent tests conducted at its customers labs have shown that when added to a regular oil, NanoLub, the world's first nanotechnology-based solid lubricant, continues normal lubrication of machines and engines in severe oil loss conditions. According to the tests, engines and machinery can work for about 7 times longer after severe oil loss when the NanoLub formulation was used compared with the base oil.

NanoLub is made up of particles of tungsten disulfide (WS2) that have a structure of nested spheres, called inorganic fullerenes, which lubricate mainly by rolling like miniature ball bearings. When used as an additive to liquid oil or grease, NanoLub significantly enhances the lubricating properties of the oil or grease with respect to wear and friction by an order of magnitude versus the same lubricant without this additive. In addition, NanoLub "wraps" the moving parts with a lubricating thin film -called a tribofilm - and continues to lubricate normally for a long time, as an "uninterruptable lubrication source" during severe oil leakages.

The superlubricity of the NanoLub particles has been discovered by the group of Professor J. M. Martin of Ecole Centrale de Lyon, France. Their experiments demonstrated that when NanoLub was used as an additive to oil, it caused a very low friction coefficient and low wear. From the characterization performed after friction, several phenomena have been proposed to explain these friction reducing properties: NanoLub's fullerenes delamination, formation of tribofilm made of WS2 sheets on the surfaces, superlubricity of the sheets and a rolling/sliding effect of the particles. According to the conclusions of the work of the French group a combination of all these effects can explain the excellent tribological (friction and wear reduction) properties of NanoLub.

Cheers,

Neal

Re: Comparison of chain lubes 5 years, 1 month ago #24391

  • Neal
  • OFFLINE
  • Platinum Boarder
  • Posts: 420
  • Karma: 0
Hello All,

I did an experiment (very crude) with my Speedplays to see how many watts are lost in pedals.





The string wound around the tape roll on the pedal unwinds with 20 grams of weight for a rather constant unwinding.

I stuck various items (small bolts, hex keys, etc) on the tape to get the weight to rotate the pedal at a constant velocity

I cut the string off and weighed the 'stuff'.

The radius to the string from the center of the spindle is 36mm.

This pedal has regular Speedplay grease with about 1500 miles on it.

Just measuring by moving the Speedplays with my fingers I can see that lubing with a lighter oil reduces the force required considerably.

I will do another experiment later on my new pedals with the lighter oil.

I have not made any calculations yet to determine watts. 20 grams at 36 mm for 90 rpm x2.

What do you think?

Cheers,

Neal

Re: Comparison of chain lubes 5 years, 1 month ago #24393

  • kraig
  • OFFLINE
  • Administrator
  • Posts: 3283
  • Karma: 4
Neal\n[quote]

What do you think?

/quote]

Cool!

Back in the day, we used a conceptually similar approach to measure/quantify bearing friction and moment of inertia for bicycle wheels. We measured the acceleration of the falling mass as a function of different masses, IIRC.

My initial thoughts are that over the distance you show in your pictures, things probably aren't "steady state" and you'll have some rotating pedal mass effects going on. Also, not sure about the whole "loaded bearing losses" vs "unloaded bearing losses" deal either.

Good stuff, though, Neal!
-kraig

Re: Comparison of chain lubes 5 years, 1 month ago #24394

  • Neal
  • OFFLINE
  • Platinum Boarder
  • Posts: 420
  • Karma: 0
Hello Kraig and All,

You are right.

As I say, it is a crude experiment - just hoping to get some rough idea of the magnitude of the watts used in the pedal.

This was a first cut at the question of how many watts are used for pedal friction.

I may put my bike on the roof so the weight has a longer time to fall and I could time it.

Also there is a slight amount of additional weight to 'break out the pedal initial friction and start it turning' that is not accounted for.

I could use my finger to start the pedal turning to reduce this error.

And the string weight increases as the weight turning the pedal moves toward the ground.

I am going to measure my pedals lubed with half odorless mineral spirits - half Mobil 1 0W-20W and see what values I get.

I used odorless mineral spirts from an art store since it has the sulpher and impurities removed and should not cause any corrosion.


More later.

Cheers,

Neal

Re: Comparison of chain lubes 5 years, 1 month ago #24396

  • tigermilk
  • OFFLINE
  • Expert Boarder
  • Posts: 107
  • Karma: 0
Work of friction over a revolution is

W=pi*D*F_mu

so power is

P=W*Cadence*2/60=pi*D*F_mu*Cadence/30

at 90 rpm, 20 grams (0.196 N), and 36 mm spindle, that's a whopping .067 W. Hard to get much savings out of that.

Re: Comparison of chain lubes 5 years, 1 month ago #24397

  • Neal
  • OFFLINE
  • Platinum Boarder
  • Posts: 420
  • Karma: 0
Hello Tigermilk and All,

Thanks for doing the math.

I agree.

Not much of a watt there to trim even I could get it to zero friction. (like with a magnetic bearing)

Any thoughts as to how the friction would change with load of pedaling?

I have read that ceramic bottom bracket bearings save about 1 watt.

So lubing your pedals might get you about 50% of that at a lower cost.

My pedals spin nicely now like my old Schwinn.

I will have to remember to lube them regularly.

I got a nice high pressure oil can ($2.99) at Tool Crib that does the job.

It has enough pressure to force out the old grease by the O ring.

Cheers,

Neal

Re: Comparison of chain lubes 5 years, 1 month ago #24429

  • Neal
  • OFFLINE
  • Platinum Boarder
  • Posts: 420
  • Karma: 0
Hello Tigermilk and All,

Just to finish up my experiment .... I tested my brand new (never used) Speedplay pedals lubed with Mobil 1 and got 7 grams vs 20 grams for the standard Speedplay grease on the old ones with about 1500 miles. As Kraig pointed out my test was unloaded and bearings with a load perform differently so I am not testing real world values here.

I would guess that the pedals might improve a gram or so as the bearings get broken in. (for my unloaded experiment.)

When I try the Mobil 1 on my chain I will cut it about 1/2 and 1/2 with odorless paint thinner to make it more flow into the chain more easily.

If you get a chance please let me know how Mobil 1 synthetic 0W-20W fares in your tests compared to 3 in 1 oil.

PM me with your address and I will send a sample of Mobil 1 if you like.

I plan to use which ever one tests better.

Cheers,

Neal
  • Page:
  • 1
Time to create page: 1.35 seconds

Poll

Which type of tire is more aerodynamic?