G'day folks
I'll keep my question to point so that it attracts some straight forward answers without being inundated with a lot of theories. Please do not find the manner of of the question abrupt or rude. Not being an engineer or having access to wild computer design and test programs, I just wish to learn by observation and setup to conduct some practical tests to help me do that.

I am going grab multiples of a random blank model. I am happy to destroy a blank or two to find this for it. Once I have found the stress point then I'll experiment with guide count, height and spacing to see if they can help prevent blank failure under the same load as the naked blank. I have already chopped off a reel foot and stem to anchor the line to once I run it through the guides to compare against the naked blank.

I'll be happy to share blank OD's for any section of the blank upon request and any other such info for those who are interested to follow my tests. I will add 3 of the same blanks to my next order

Before I can setup a test I need to find out what is a reasonably accurate and practical way to find the weakest point on a naked blank clamped to the 3 o'clock position?

I have another question. After watching a manufacturer sand blanks after the celophane is removed to get rid of the spiral epoxy ridges I noticed how "hands on" this evolution is, so when I see that the OD measurements are all the way into the 1000th, How can these measurements be accurate when the worker is using a subjective observation that all the ridges and the epoxy gloss is removed and where in fact some areas got a little more sanding than others?

Everytime but once that I've seen a rod correctly stressed to failure the failure always occured in the butt section not far from the front grip far below any guide. The one time it didn't I saw on a video by St. Croix where they tested with a machine replicating a very hard hook-set (I think?) and the blank shattered in 3 or 4 pieces and even then one of the breaks was in the lower section of the rod.

Good luck with your testing.

Raymond Adams
Eventually, all things merge, and a river runs through it..

Rohit,
Could I ask you to collect some data for me ?
Before breaking could you take diameter measurements of the subject blank every 2 inches from the butt to the tip.
If you could please post the data and allow me to review to look for potential stress risers.
After blank destruction please measure the point from the tip to the butt to see if that matches my review.
If too much trouble, so be it.

Thanks
Gene

Like Raymond I've observed the shattering of blanks and would strongly suggest the wearing of eye protection!

In addition I would suggest wearing gloves if you intend to do this by hand as I've also known it to send a person to the emergency room because of the wounds to his hand.

You're going to have one major problem - no two blanks will be identical. Breaking one and then adding guides on another and comparing break strength will not provide accurate results. There is too much hand labor involved in blank construction to allow for verbatim construction. They're all likely to vary by as much as 5% anyway. But the location where the break occurs will be roughly the same, assuming the blank has not been damaged or impacted anywhere.

To get any sort of decent empirical results, you'll have to break 10 naked blanks and then take an average to arrive at the bare blank's deadlift limit. Then set up a particular guide setup, 10 times, and break 10 more. Then take an average. So on and so forth for each additional guide setup. Just breaking one of each isn't really going to tell you anything at all - any one of those blanks could be unusually stronger, or weaker, than the rest. Only by testing numerous blanks per group can you weed out the highs and lows and arrive at any sort of reliable results. And even then, 10 blanks per group isn't many if you're really after good results.

I can already tell you that nothing you do with regard to guide additions will increase the dead lift limit, power or strength of the rod blank. You can reduce it, but you will never increase it.

The problem with locating supposed stress risers in this manner is that it will also require that you take measurements of the ID of the blank at each of those locations where you're taking the OD measurements as well. Diameter is not your only concern - wall thickness plays a role too and it is not consistent all along the blank's length.

If you want to bust a few blanks - clamp the rod horizontal and attach a line to the tip and attach a bucket. Pour in water, slowly, until you reach the blank's inherent dead lift limit. It will be more than you think and you need to wear safety glasses. After you add guides, run the line through the guides and attach the bucket to the line and add water.

Rich Forhan has done quite a bit of this type stuff over the years.

I 'm no statistician, but wouldn't the number of samples required be dependent upon the variation in the results? If 3 random blanks break at exactly the same load, then 3 or 4 might be all that would be necessary to test. On the other hand, if the first 3 and subsequent blanks break at dramatically different loads then 50 blanks might not be enough. Does sample size have something to do with the standard deviation of the results?

Like TK stated earlier and KH eluded to. There is a human factor to consider, Starting with who cut out the pattern, who attached the pattern to the mandrel, who laminated the the fabric to the mandrel, Who attached the cellophane to the blank before cooking Now after the blank has been cooked for over 2 hours and the cellophane is removed the blank is now sanded then coated and dried. As you can see there are many individuals involved. During any one of these steps if a process is flawed the blank will fail. So what you are looking at is almost impossible to find. This is not saying you might be able to find a fail factor. But when you do it will not be at the same exact point. The only way I can see this being done is to eliminate the whole human factor . I am not an engineer. So to answer this question you well have to go a engineer or Todd Vivian at Mudd Hole. He designed blanks at Lamiglas for almost 20 plus years

Good Wraps Bob

The number of samples in the group has to do with the validity of the results. Even if the first 3 broke at exactly the same load (they won't) you'll have a hard time convincing anyone that has dealt with such testing that the next 3 will do the same.

The bottom line is that Rohit is not going to increase or improve the deadlift limit with any sort of guide placement. If he ran the line through the center of the blank, which in theory will provide 100% perfect stress distribution, the deadlift limit or strength of the blank is not going to be increased.

But if he wants to do this for his own satisfaction that's certainly up to him. You just don't want to draw conclusions from tests that aren't necessarily scientific in nature. And what he proposes, isn't.

...................

Whoa
very good question Rohit

A couple of observations
This is accelerating into the exacxt realm that Bill Hanneman ably pointed out...................the 1000 negative tests.
1.
You dont have to do destructive testing of a blank under load to identfy stresspoints.
2.
Whether you do destructive testing or progressive data analysis to identify stress points you need to know before hand what you are testing for, failure or the reason for failure.
3.
So you need to know:-
- what you are testing for
- the nature of the beast you are testing
- design the test protocol so it teases out what you are setting out to test for.
otherwise you end up doing 1000 tests that tell you nothing about what you want to know.
4.
That damn fishing rod is a complex piece of equipment......................you are entering the realm of multi-variate testing.

The questions that need to be asked ...........FIRST................are:-

- what are you testing for................failure or the cause of failure.
- is the test protocol representative of the typical use of the item
if not .........design the test protocol so it is.
- failure of a rod = excessive stress.
- what is stress
- how do you determine the stress in the item as a result of the failure ( or just prior to failure)
- what are the parameters causing the stress in the item.
- how will the test protocols measure those parameters during the test.

Then you can start to test

Tom makes a very cogent point.........................the variables from blank to blank is such that destructive testing is not very instructive as a lot of blanks are going to be busted in the name of poor science.

John makes a good point in a similar vein
If you don't know the inherent variability in the item to be tested ..............or the parameters that affect failure ...............you must do a large number of tests to establish that the possible range of variability has been adequately sampled.
The direct answer to John's question is that it is a matter of luck ( in real life ).
The typical, outcome in real life is that the mean and the standard deviation shift with increasing tests until the natural variability of the item is adequately covered & the law of diminishing returns kicks in & little change in either is experienced.........................the point at which this occurs in a normally distributed variability is about 30 random samples.
less than 10 and you are playing with fire.......................more than 30 & you are probably wasting time.

The normal experimental protocol in experiment design is :-
- review the physical parameters involved & establish their relationship
ie identify the parameters you need to collect data for to undertake a meaningful test.
- test to validate that relationship ( with an appropriate protocol )
- test validate the overall effect of all the parameters at play in the item being tested ( with an appropriate protocol )
- THEN sensitivity testing of the likely range of variability in the parameters involved in the item being examined............in their effect on the item.

The first questions that need to be answered before even thinking about breaking rods are:-
- what are you testing for..............failure or stress at failure
- what is stress...............why is stress what causes failure.
- what parameters do I need to collect data for ..............to determine stress...........
- what are their relationships in generating stress.

Then you can design the testr rig & the data you need to collect in the tests you do

Otherwise..................charge off & break that 1000 rods & you will identify little of any relevance..................other than every rod can be broken.............if you can put in enough energy.

Lets start by working thru the 4 questions directly above.............so we can communally identify what we should be doing and how.

We can get thru the basics of this if everyone is willing & stay focussed on the specific issues.

My suggestion is to do nothing other than use an adequate number of guides for the rod you are building. This information is readily available. Beyond that, you may have fun breaking some blanks, but you will not improve anything. Using enough guides will provide you with adequate stress distribution and preserve the strength of the blank to the greatest degree possible. Educate your customer on the normal and most common causes of broken rods - abuse (impact) and poor fish fighting technique. These two problems can be at least partially corrected through education of the customer by the custom rod builder.

............

What instrument would one use to measure the ShockWave of Energy going through the rod per square inch?

Every new rod builder comes into the hobby with their own personal background. And being new to the hobby they find what they think are problem areas which they have the unique ability to solve or improve. Because they are new to the hobby they have no idea that many other people of equal or greater intellect and experience have come before them and been through the same process.

In Southern California U.S.A, Our earthquake are measures by a Richter Scale. Maybe something like this device could be of use in your research for finding the break point and data collecting. [en.wikipedia.org]

Rob
Some others , have been building & using for 50 + years and have learnt a few things along the way that do matter, and can make a difference.
Where the intellect & experience you refer to has resulted in documentation there is no reason to revisit history
Documentation in the rodbuilding community of the parameters involved and proof of concept don't seem to exist at the present time...........hence questions & discussion amongst those with the capabilities to do so..............such is the history of the human species in the evolution of human knowledge.

Ben
One of the issues in technical development and discussion is trying to jump too far ahead and guessing where the future might be leading you.
Its guessing that often gets you into incorrect assumptions of the value of the work, and particularly so where the guesser does not understand the parameters or their relationships in that work.

Basically shock loading is merely the rate of change of the bending moment relative to the inertial moment of the section being considered.
That leads back to the question............what parameter are you going to monitor & for what ..............then you design the test rig accordingly.
No big deal where the test rig has been designed appropriately...............if shock testing is the objective.

Earlier post, someone mentioned that the rod breaks into multiple pieces and sounded loud like a gun shot. What causes this ?

Dennis,

Thanks for opening such a great subject !... Love to learn and Looking forth to read all post from you and other.

Ben

Ben
any carbon fibre rod break, whether shock loaded or just over stressed slowly will go off like a gunshot...........its a catastrophic instant failure.
A excessive shock load can, and usually does, but not always, result in more than one break.

Rohit Lal Wrote:
> Before I can setup a test I need to find out what is a reasonably accurate and practical way to find the weakest point on a naked blank clamped to the 3 o'clock position?


I take it that increasing the load suspended from the tip in measured increments will take THE blank to ITS failure point.

My findings from your replies
A particular model has to be sampled multiple times to find an average for that model as there is very poor tolerance. This tells me that computer programs can only show best case results and not real world results. It also appears that a system that uses blank OD's to determiine any steps in taper to find a weak point is not accurate as there might be a reienforcement component in made from either a different material or added wall thickness. The dstruction method is acurate per blank but one cant use it as its already bye bye blank. The OD method is not looking at enough information so is inaccurate

My further conclusions
This leaves a end user like me with the manufacturers ratings together with the current tools like the static deflection test and CCS as the best guideline to work with. I do think that the deflection test should be done at what would be considered a heavy realworld load for that blank

Tests
As destruction tests and OD methods are deemed useless, I will pick a blank and use an extreme deflection test to to check if it is within the manufacturers design limits. Then I will setup a guide layout that will mimic the blanks natuaral bend. Then I will test for bowstring effects as well as any change to the bend with different guide heights. I will make the OD increments along the blank available for EM and I hope Denis will also chime in also with a guide layout from his formula.

The Blank
American Tackle AXLB60ML. I use this a fair bit and am famalier with the real world performance and its ability. Lets use this awesome stick as a test monkey. Its a great value blank enabling testing withought breaking the bank

The line rating for this blanks is 12-30lb, may I please have your opinions as to what pound test weight would be considered an extreme deflection load for this. I have tested it at 17lb drag setting on a vertically diving yellowtail kingfish. The blank handles this load well so I think 22lb would make an ideal test weight with the rod clamped at 3 o'clock

The naked blank
length is 6'
ODs every 2" starting at the tip - .122, .125, .131, .137, .146, .160, .174, .188, .204, .220, .238, .260, .280, .302, .322, .339, .360, .386, .408, .432, .454, .477, .498, .520, .537, .554, .569, .585, .601, .620, .639, .657, .677, .693, .708, .722, .745..........butt OD .750

Rohit
Thats a fine piece of information you have enthusiastically generated there, but what is a blank's natural curve..............it isn't found in a conventional horizontal static test rig.
The naturalm curve of a blank is only really described by the curvature it can achieve without failing................that depends on the angle of attack of the load and the stress that load generates on the blank....................unfortunately that is a dynamic thing.
you need to go back and look at the explanation of the failure driver in a tubular blank under deflection and reconsider whether your test design will replicate other than less than 5% of the deflection curve experienced by the blank in actual fishing use , and what the drivers of stress will be in those other situations before you rush into testing.
It would be far better if we devoted some time to discuss the parameters we need to collect data on before you start testing. otherwise you stand a terribly high chance of doing one of those 1000 tests that produces no real information ............................but what you propose would be a fine test of the blank in less than 5%of its operating conditions in the real world in use.

I take it you are in the western suburbs of Sydney
Let us know where your shop is & we can discuss face to face , if we have mutual time available, the next time I am in Sydney