Split Lock Washer_Fail :(

Status
Not open for further replies.
Rupunzell

Rupunzell

Bernice Loui
The 74' developed a rattle/click and weird brake pedal feel recently.

Finally got a chance to have have a look-see and figure out what is wrong.
Turns out, the stock Fiat brake caliper bracket bolts with retained split lock washer failed.

The split lock washer shattered causing the pre-load on that cap screw to go away allowing the brake caliper bracket to rock and more than it should.

Here is what the remains of the shattered split lock washer looks like and the two Fiat cap screws involved. These are without question OEM from 1974 complete with rust from being in service all those years.
IMG_2017.JPG


The rust only added to the inherent problems with split lock washers.
To have any degree of spring, they are heat treated to add hardness making them springy. Problem with this, harder means brittle, brittle means likely to shatter when stress on the part is cycled. Enough load cycles and corrosion induced cracks equals failed part.

Over the years, I have been advising to remove ALL split lock washers in any assembly that has significant mechanical stress on them and replace split lock washers with Loctite, proper flat washer, conical tooth lock washers, Nord lock washer, prevailing torque lock nuts and do what ever is possible to prevent and control corrosion as these details often result in part failures leading to even more serious failures.

Split lock washer have been tested and well proven not to prevent loosening of threaded, they also cause threaded fasteners to loosen easier and faster. On this bolt science web page there is a description of these findings dating back to 1969 and two videos of split lock washers being tested on a Junkers machine.
http://www.boltscience.com/pages/helicalspringwashers.htm

And this video we clipped from else where:
https://www.youtube.com/watch?v=ArBxOclBuSQ

As stuff gets worked on, items like split lock washers, old corroded hardware and all related get's replaced with new equal or better. This is all low cost insurance that bad stuff does not happen when least expected and goes a long ways to keep Murphy in check.

Oh, split lock washers are forbidden in the shop, if any of them appear they ended up in the circular file. Same applies to old hardware.


Bernice
 
I found a few broken split washers while replacing my clutch. Fortunately, that was the first thing that needed to be replaced when I got the car. Loose bolts attaching the flywheel to the crankshaft could have been disastrous!
 
The flywheel to crankshaft bolts shouldn't have washers. There should be a thin plate there instead.

AZDrifter said:
I found a few broken split washers while replacing my clutch. Fortunately, that was the first thing that needed to be replaced when I got the car. Loose bolts attaching the flywheel to the crankshaft could have been disastrous!
Click to expand...
 
WHOA! (Pun intended...)

I have never seen that kind of failure... but I do admit I live in a vacuum.

Its a good thing I only use my brakes sparingly and as a last resort...

Braking (Pun intended again...) Split Washers??? GEEZ! Obviously you are too hard on yours to the point you are abusing them!
 
Belleville washers

In a number of places, Fiat uses a very thick belleville washer, also known as conical disc springs. They are keep a constant pressure on the bolt, reducing the impact of vibration on loosening of the nut.

I have read that they only effective in situations with low dynamic loads, but I wonder if the thick ones on our Fiat engines overcome that issue. I have only been able to find thin ones listed in with fastener suppliers. You can stack them for more load capacity, only I have not been keen to go this route.

What do you all think?

The Nord-Lock system looks great in the video. Anyone have experience with it?

Paul Davock
 
I think you are overlooking a key point...

By the way have you all see this one:
[ame="http://www.youtube.com/watch?v=v_CgPsGY5Mw"]French Model Commercial (CC) Full-length - YouTube[/ame]

There are limitations on the fasteners for sure but the key is the breakaway torque. Once the preload is lost then the split washer loses it benefit and is arguably worse than a flat washer base on the loss of surface area contact as the whole thing unwinds. But in reality it is much much more complicated than you lead on, considering thread class and fit. So with all that cool stuff said you need to focus on the point of where it starts to lose its clamp load. I believer you will see the split washer has a higher breakaway before it starts to lose its load - this is the key point. Failure is a failure and once it goes the rest of the story is some nice philosophical chat over a beer(s).

When you ad the thread locker you get a twofer where you are taking up the tolerances in the thread fit - filling the gaps that allow movement and sliding of the surfaces and leading to loosening. It also has a "bonding" (more to this story too) effect with adhesion characteristics for the surfaces and resistance to flexure under load- this is why there are different products available for permanent and semipermanent applications. It works and works well.

There are better fastening systems but it is all application based. For safety applications I have my stuff wired - CV bolts and brake bolts for example. I also use split washers, prevailing torque, star washers, bevel washers, and thread locker it all depends.

Looking at the picture of your hardware I have to say your bolts are jacked! Those guys need replacement. No surprise that the split washer broke just look at a S-N curve and you will see you are on borrowed time considering the corrosion rolled into the equation. Now is a good time to look at all your key fasteners and make sure all is good.

I did check out the links but this was my favorite on the topic :clap:
https://www.youtube.com/watch?v=z9gGBSVwKSE
 
The point of this post was to share the fact:

*Old knackared hardware will fail.

*Split "lock" washers do not work for mechanical assemblies that are subjected to cyclic stress.

*Rust and corrosion add significantly to risk of parts failure.

Using a flat beam torque wrench with floating memory indicators and some new thread fasteners, we will run a little experiment to see if the breakaway torque with and without split washers indeed have higher break away torque. I have yet to see split washers used on any aerospace mechanical assembly or current real race car.

Threaded fasteners are basically a tension spring. The amount of compression is set by the amount of torque applied to the helical threads. Once the compression is set, or pre-loaded, all is happy until that preload on the joint is exceeded. Keep in mind metal is elastic and better to think of metal being more like rubber than solid item. Exceeding the preload set in the threaded fastener will cause the joint to move, once that joint has moved, it can and will work loose over time. This is the basis of how the Junkers test machine functions.

As a rule:

*Threaded fasteners are intended to provide clamp force.

*Pins or similar positive locating devices are intended to locate the parts to be assembled.

They should not share their intended function.

Clip from a NASA seminar on fastening.
https://www.youtube.com/watch?v=ArBxOclBuSQ

The NASA 1228 fastening hand book:
http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19900009424.pdf


Yes, I have designed helical split washers into production products that have been on the market for decades with zero failures over thousands upon thousands of devices installed. They are used as a low force spring to keep power semiconductor devices compressed on to their heatsink thermal transfer pads. The low force spring action compensates for thermal expansion and contraction as these devices heat up and cool off. Split helical washers have their place and function. They do not "lock" or prevent loosening of threaded fasteners on stressed mechanical assemblies. This is proven and well tested fact, yet the common belief is they "lock" stuff together which is false.


Bernice
 
I may disagree that a helical washer is not a lock washer if you use one that was manufactured as they were originally.

If you compare a new split washer and compare it to an old one, the old one had one and sometimes two tangs on the high edge of the split.

This tang digs into the nut and bites the nut when torqued. It is therefore also obvious that once this arrangement has been unfastened, you should not reuse the washer nor the nut as there would be a grove in the nut made by the tang when loosening it.

I suspect the issue is that to save money, manufacturers made split washers without a tang on one side or both, which changed the design from something that works to something that fails.

M
 
Well, no.

Two parallel surfaces against another essentially parallel surface, how does the digging in occur?

BTW, did a quick test to see if the break-away torque on a threaded joint is higher. Results are NO, the split lock washer produced lower break-away torque compared to a plain flat washer.

The amount of clamp load produced in a threaded fastener joint far exceeds the spring force produced by a helical split "lock" washer. This instantly flattens the thing into nothing more than a flat washer. For the helical split washer to go any good at all, it must produce a LOT more spring force than they currently do, otherwise they are simply another point of failure for a threaded joint.

I'll produce another video and images showing this.

What I would like to understand is, why are folks so attached and dug into the belief these "Helical Split Lock Washers" actually work when the body of facts both theoretical and experimental has proven time and time again they do not perform as advertised?


Bernice


BellaMacchina said:
I may disagree that a helical washer is not a lock washer if you use one that was manufactured as they were originally.

If you compare a new split washer and compare it to an old one, the old one had one and sometimes two tangs on the high edge of the split.

This tang digs into the nut and bites the nut when torqued. It is therefore also obvious that once this arrangement has been unfastened, you should not reuse the washer nor the nut as there would be a grove in the nut made by the tang when loosening it.

I suspect the issue is that to save money, manufacturers made split washers without a tang on one side or both, which changed the design from something that works to something that fails.

M
Click to expand...
 
Some, the correct ones have tangs and others, the ones that will fail, don't have tangs.

On the ones with tangs, they will dig into the nut.

For your test to be conclusive you need to test both types of helical washer.

Cheers,
M
 
How deep or surface area could those "tangs" possibly dig into the nut/bolt/joint material or ... to do any real good when there are parallel surfaces involved?

That tang, barb or what ever single point will be flattened nicely once the force generated by torque applied to the threaded fasteners helix. Then we have the claim that the tang, sharp bit or .. will dig in once the nut/bolt loosens. If the threaded joint becomes un-done that far, pre-load in the threaded joint mostly gone, game over-failed joint. This has been an on-going discussion between me and the guy who sells hardware at the local industrial supplier. He restores cars and noted this effect. The reply, "the joint is already loose. how much residual clamp load is on that joint, little if any significant amount." Game over.

If the split washer is hard enough to dig, this mean it is hard, hard means brittle, brittle means the risk of shattering is high. This why split washers are often found cracked into bits in cyclic stressed threaded joints.

As for digging into grade 12.9 or similar hardware, with hardness of RC38-44, it will need to be greater to do the least bit of scratching the surface with a LOT (pounds by the thousands) more force behind it to even begin to dig or cut.

If the intent is to force teeth into the joint surface, an internal tooth, external tooth or Whizz-Lock nut could be better at that. Except, those don't prevent loosening much better than similar preventive measures. Beyond that, those bites into the joint material are stress risers that can grow into serious cracks causing the part to fail.

One example of why digs, pits, number stamps, sharp edges and similar are serious trouble makers:

"The number 4 piston exhibited a hole burned through the crown structure of the piston. The hole was located along a crack that extended across the top of the piston, in-line with the piston pin.

>>>Metallurgical examination determined that fatigue cracking of the piston initiated from a stamp mark "4" on the top surface of the piston. The stamp mark, which was made by the manufacturer during engine buildup, acted as a stress raiser and was responsible for initiation of the crack.<<<

Once the crack extended to the interior surface of the piston, hot combustion gasses created the burn through hole and escaped into the crankcase, leading to engine failure."

http://www.ntsb.gov/aviationquery/brief.aspx?ev_id=20001212X22291


There is stuff worth learning about from these incidents. Including how to prevent them, how to never allow the to happen, how to inspect for them and most important of all how to design and not repeat these same errors that can result in tears.

Threat this car stuff as it your life and the lives of others depended on what is done. Never, every take any of this stuff for granted.



Bernice


BellaMacchina said:
Some, the correct ones have tangs and others, the ones that will fail, don't have tangs.

On the ones with tangs, they will dig into the nut.

For your test to be conclusive you need to test both types of helical washer.

Cheers,
M
Click to expand...
 
This is all fascinating to me as a non-mechanic. I bought my Fiat to learn about mechanics, and how this little car works. My background is in opera which I studied academically. So I understand most of what you are saying but maybe not at the same level.

I do see the benefit to this but what is the outcome? Do I disassemble my car now and take out the split washers? I have been as far into the car as a top end rebuild and the only washers that have failed, aside from the ones that were corroded beyond recognition, are the ones holding down the cylinder head, which cracked. I ordered new ones, which I suspect were off another old car.

I was really only pointing out that there are different split washers available and was more questioning why in the original test done by bolt science, why were not both versions tested.

Cheers,
M
 
BellaMacchina said:
T
I do see the benefit to this but what is the outcome? Do I disassemble my car now and take out the split washers?
Click to expand...

No need to do that for a street-driven car that is showing no signs of a problem. Next time you're in there (and there will be a "next time") inspect and replace as necessary.

(And of course you always and immediately investigate unexpected changes in brakes and steering - it's worth noting that Bernice started this thread by saying "developed a rattle/click and weird brake pedal feel recently").
 
Precisely..

Address these issues/problems as they appear. This is why the 74' has been a rolling restoration project. Each time something comes up that is re-done, it gets re-done to a very high standard so the problem will not occur again for a very, very long time and any problems baked in originally can be properly addressed.

One of the reasons why I post this stuff is partly education on the Science and current academic understanding of how this stuff actually works based on work that has been proven theoretically, experimentally and in actual practice for technology items produce today and in the past.


Here is one example of crack growth. It begins with at the sharp apex of the Vee.

https://www.youtube.com/watch?v=Bgsi3AK1ju4&list=UUBZqUHA2vnZmsIwwavVGN3A



Bernice



EricH said:
No need to do that for a street-driven car that is showing no signs of a problem. Next time you're in there (and there will be a "next time") inspect and replace as necessary.

(And of course you always and immediately investigate unexpected changes in brakes and steering - it's worth noting that Bernice started this thread by saying "developed a rattle/click and weird brake pedal feel recently").
Click to expand...
 
I think there still is a misunderstanding of the fundamentals

This was a bit unsettling:
"BTW, did a quick test to see if the break-away torque on a threaded joint is higher. Results are NO, the split lock washer produced lower break-away torque compared to a plain flat washer."

So I really struggle with this because doing a test on a single sample in itself really raises many questions. Modern engineering bases results on statistics and a conclusion derived from the data, at least that is how we do it in the automotive industry. So doing this kind of study is normally done with a torque and rotation sensing tool like an Atlas Copco unit that measures the run up and break away from which a super cool curve is generated showing all kinds of key data. Modern practices are done with 25-30 samples to verify and assure the data is normally distributed (everything falls into a symmetrical bell curve) ; this sorts out the one off flukes from reality. The tests are controlled too where the fastened joints are also allowed to normalize because there is a time dependent factor on the clamping load and breakaway. All kinds of groovy science on run up RPM and final angle of rotation too but maybe that is a bit over the top for this general conversation.

So that is a bunch of cool stuff but a single sample measured with a general tool, I am jumping to a conclusion on this but I suspect the evaluation was done with a standard beam wrench - I may be wrong though, there is not enough data to make a conclusion. This really goes to the level of granularity in that you just can't get the level of visibility to really know what is happening a the point of breakaway.

Looking at the linked supporting data I loved, loved, loved the space professors video not just because it is so old school but that it is so lofty in that the information is very thin on support. I do like the other videos and rate them high in coolness but more importantly if you look at the curves I think you will see the breakaway torque is higher with the split washer than without- I'm just sayin.
 
Much about EGO, NO ?

This was not the first time you pulled this. It happened with Steve Hoelscher over suspension design, the engineering and his methods.
http://xwebforums.com/forum/index.php?threads/3119/


Then we had an incident over the Porsche 901 shifter for the LeMons racer.
http://xwebforums.com/forum/index.php?threads/17780/

*Which has, by the way been abused in more ways than ever intended, tormented to extremes and survived just fine after several LeMons races, endured thousands upon thousands of shifts and has continued to work fine without any galling of the cam. Worth adding is every Porsche-phile that has seen it wants one to correct the designed in shifting problem of the 901 used in the 914.

There is no need to believe anything I or anyone else writes or publishes about the effectiveness of split lock washers, Do the work yourself, test the idea on your own. Then write it all up complete with theory, calculations, experiment methodology and results then publish your study for all peers to test and verify your publication and claims.

I'm at the end of any further discussion about this.

And yes, I'm quite familiar with Atlas Copco's latest and greatest torque tools. They are not the only ones who offer this type of torque tool.

Bernice


Dallarax19 said:
This was a bit unsettling:
"BTW, did a quick test to see if the break-away torque on a threaded joint is higher. Results are NO, the split lock washer produced lower break-away torque compared to a plain flat washer."

So I really struggle with this because doing a test on a single sample in itself really raises many questions. Modern engineering bases results on statistics and a conclusion derived from the data, at least that is how we do it in the automotive industry. So doing this kind of study is normally done with a torque and rotation sensing tool like an Atlas Copco unit that measures the run up and break away from which a super cool curve is generated showing all kinds of key data. Modern practices are done with 25-30 samples to verify and assure the data is normally distributed (everything falls into a symmetrical bell curve) ; this sorts out the one off flukes from reality. The tests are controlled too where the fastened joints are also allowed to normalize because there is a time dependent factor on the clamping load and breakaway. All kinds of groovy science on run up RPM and final angle of rotation too but maybe that is a bit over the top for this general conversation.

So that is a bunch of cool stuff but a single sample measured with a general tool, I am jumping to a conclusion on this but I suspect the evaluation was done with a standard beam wrench - I may be wrong though, there is not enough data to make a conclusion. This really goes to the level of granularity in that you just can't get the level of visibility to really know what is happening a the point of breakaway.

Looking at the linked supporting data I loved, loved, loved the space professors video not just because it is so old school but that it is so lofty in that the information is very thin on support. I do like the other videos and rate them high in coolness but more importantly if you look at the curves I think you will see the breakaway torque is higher with the split washer than without- I'm just sayin.
Click to expand...
 
Last edited by a moderator:
Status
Not open for further replies.
Back
Top