Aftermarket V Original this Time its Science

[moanthebairns]

Instead of speculating on aftermarket v original equipment it’s time to do some testing. I used to be a lab tech don’t you know.

I for one am sick fed up of seeing videos pop up of some guy with a hammer smashing open a Chinese helmet and a UK spec one, wtf does that prove, fk all, yet many people go look at the difference, WOW.

I however am choosing to test visors, one they are cheaper than helmets and two because I have and manufactures on and an aftermarket one sitting doing nothing after my hizzy lid died, (god rest its soul)

So the test.

Straight forward one here it’s going to be an impact test

Apparatus will be as follows

2m surveyors rod
One 0.5 kg tipped weight
Visors
One glamorous assistant
A calculator
A video (not my dad’s old jvc you’ll be happy to hear)
and a step ladder

I will drop the 500 gram weight from a height of 2m onto the centre of the chosen visor. Record the results, simple.

Now for some math, which I’m ste at so I might need some help here

F = M x A

The objects mass is 0.5 kg and is dropped from 2m so the velocity before impact is 6.260990 m/s the kinetic energy just before impact is 9.8 Joules

I am assuming after dropping the weight a few times that the distance travelled upwards after the impact is almost fk all 0.005m meaning an average impact force of 1960 N

As it’s a tipped end I decided to work out the pressure as it will be greater due to a smaller impact point.

P = F/A the force is 1960 over (pie mmmm pie over 4 x 0.002 squared = 624 x 10 to the He-man power of 6 n/m2 basically 624 bar is going to be the pressure in this area for a split second.

Anyone disagree? It was a lot easier with a computer, I’ve forgotten a lot due to drink, drugs and hitting my head.

[LiamB]

You have too much time.

[thatdude]

Here's one I would like you to do

a water beading test off the visor surface. Does water flow better off one than the other?

[moanthebairns]

thatdude said:

Here's one I would like you to do

a water beading test off the visor surface. Does water flow better off one than the other?

I cant be arsed, just ride quicker in the rain.

[Prof Prolapse]

Firstly it's spelt "maths". This is also where you're about to make the same fk up as the youtubers in that you're not going tp satisfy the stat requirements to claim anything.

I don't do this anymore but given your grievance before you do anything look at your sample size. Depending on the size of change you wish to detect this will vary enormously. E.g. To detect a 10% change you may need ten thousand visors! A 90% may require only a few hundred. You don't guess.

You then need work out a confidence interval. This is probably p= 0.05 or 95%. So if you need to ensure this is met. After 10,000 visors you may find it only reaches 93%. So you still have to conclude no positive result, this is why many people don't understand global warming, and why CERN is the size of a small Polynesian Island.

With proof of concept being established there's no value otherwise.

[gwm]

Prof Prolapse said:

So you still have to conclude no positive result, this is why many people don't understand global warming, and why CERN is the size of a small Polynesian Island.

Que?? No idea what this means...


How about repeat testing on same visors and average out? Mythbuster style testing as opposed to proper science.

[moanthebairns]

Prof Prolapse said:

Firstly it's spelt "maths". This is also where you're about to make the same fk up as the youtubers in that you're not going tp satisfy the stat requirements to claim anything.

I don't do this anymore but given your grievance before you do anything look at your sample size. Depending on the size of change you wish to detect this will vary enormously. E.g. To detect a 10% change you may need ten thousand visors! A 90% may require only a few hundred. You don't guess.

You then need work out a confidence interval. This is probably p= 0.05 or 95%. So if you need to ensure this is met. After 10,000 visors you may find it only reaches 93%. So you still have to conclude no positive result, this is why many people don't understand global warming, and why CERN is the size of a small Polynesian Island.

With proof of concept being established there's no value otherwise.

I understand where your coming from, ideally id have 10 - 20 samples and could perform a repeated test, giving a failure ratio if any. But a few problems with that, A) im not made out of money, B) I have no knowledge of a specified pass threshold allowed in the tests to make if fit for sale C) I don't know the load at which the visors are tested to or even the test method.

what I am doing is carrying out a consistent test for both visors, that's all. Much better than hitting with a hammer.

[balls-out]

moanthebairns said:

stuff
One glamorous assistant
more stuff

can we have pictures of the assistant please? thats why I gave up engineering to become a manager - I know how to cut to the important stuff.

Edited by balls-out on Thursday 21st August 13:49

[moanthebairns]

gwm said:

Prof Prolapse said:

So you still have to conclude no positive result, this is why many people don't understand global warming, and why CERN is the size of a small Polynesian Island.

Que?? No idea what this means...


How about repeat testing on same visors and average out? Mythbuster style testing as opposed to proper science.

the test is likely to be failure or pass, it is for many plastics on an impact its not how badly it deforms, an impact test is normally to see if it brakes or not. the only averaging done on impacts is if the number of samples tested pass the threshold fail rate, which typically is none, if a percentage was to fail over the allowed threshold, further testing of a predefined number would be done, if any where to fail in that set of tests the unit would receive a fail over all.

things like creep, elongation, deformation, tensile etc then you would get an average from samples

[moanthebairns]

Prof Prolapse said:

Firstly it's spelt "maths". This is also where you're about to make the same fk up as the youtubers in that you're not going tp satisfy the stat requirements to claim anything.

I don't do this anymore but given your grievance before you do anything look at your sample size. Depending on the size of change you wish to detect this will vary enormously. E.g. To detect a 10% change you may need ten thousand visors! A 90% may require only a few hundred. You don't guess.

You then need work out a confidence interval. This is probably p= 0.05 or 95%. So if you need to ensure this is met. After 10,000 visors you may find it only reaches 93%. So you still have to conclude no positive result, this is why many people don't understand global warming, and why CERN is the size of a small Polynesian Island.

With proof of concept being established there's no value otherwise.

oh and by the way the scales at which your talking about are way off from the dangerous goods testing I've done in the past.

lets just say that for a simple jerrican that will be sold in hundreds of thousands the test sample size could be as small as 16 jerricans

9 for impact, 6 sets receiving two drops and 3 sets receiving one drop on the likeliest impact zone to fail on
3 for stack testing
3 for air leakage and hydraulic testing
1 for specification check

that is all that is required, its simply a pass or fail when going for accreditation in this field, the initial R&D would have a much higher range but im testing on items that have went through this

I assume you are talking about more of a product recall test, given the number you have quoted

Edited by moanthebairns on Thursday 21st August 13:57

[Biker's Nemesis]

Prof Prolapse said:

Firstly it's spelt "maths". This is also where you're about to make the same fk up as the youtubers in that you're not going tp satisfy the stat requirements to claim anything.

I don't do this anymore but given your grievance before you do anything look at your sample size. Depending on the size of change you wish to detect this will vary enormously. E.g. To detect a 10% change you may need ten thousand visors! A 90% may require only a few hundred. You don't guess.

You then need work out a confidence interval. This is probably p= 0.05 or 95%. So if you need to ensure this is met. After 10,000 visors you may find it only reaches 93%. So you still have to conclude no positive result, this is why many people don't understand global warming, and why CERN is the size of a small Polynesian Island.

With proof of concept being established there's no value otherwise.

You tell him Prof.

[Renn Sport]

moanthebairns said:

thatdude said:

Here's one I would like you to do

a water beading test off the visor surface. Does water flow better off one than the other?

I cant be arsed, just ride quicker in the rain.

Lol... you're going to all this trouble and cant be arsed to do this simple test.

I love it.

[Prof Prolapse]

moanthebairns said:

oh and by the way the scales at which your talking about are way off from the dangerous goods testing I've done in the past.

lets just say that for a simple jerrican that will be sold in hundreds of thousands the test sample size could be as small as 16 jerricans

9 for impact, 6 sets receiving two drops and 3 sets receiving one drop on the likeliest impact zone to fail on
3 for stack testing
3 for air leakage and hydraulic testing
1 for specification check

that is all that is required, its simply a pass or fail when going for accreditation in this field, the initial R&D would have a much higher range but im testing on items that have went through this

I assume you are talking about more of a product recall test, given the number you have quoted

Edited by moanthebairns on Thursday 21st August 13:57

No I picked numbers out of the air to illustrate the point that to do anything scientifically you need to use stats which dictates if it's feasible before you crack out the step ladder and start dressing the Mrs up in sequins.

What you describe is not a scientific experiment. For tests like this the theory is decided, you simply need to test they meet a defined standard. You instead initially appeared to be trying to answer a question, this is much more difficult and expensive, which is one of the reasons definitive answers with safety gear is so tricky.

What you suggest is fine of course. But it's more like quality control.

[Prof Prolapse]

Biker's Nemesis said:

You tell him Prof.

[Prof Prolapse]

moanthebairns said:

what I am doing is carrying out a consistent test for both visors, that's all. Much better than hitting with a hammer.

Can I bring my air rifle around to simulate stones? Probably.

[moanthebairns]

Prof Prolapse said:

moanthebairns said:

oh and by the way the scales at which your talking about are way off from the dangerous goods testing I've done in the past.

lets just say that for a simple jerrican that will be sold in hundreds of thousands the test sample size could be as small as 16 jerricans

9 for impact, 6 sets receiving two drops and 3 sets receiving one drop on the likeliest impact zone to fail on
3 for stack testing
3 for air leakage and hydraulic testing
1 for specification check

that is all that is required, its simply a pass or fail when going for accreditation in this field, the initial R&D would have a much higher range but im testing on items that have went through this

I assume you are talking about more of a product recall test, given the number you have quoted

Edited by moanthebairns on Thursday 21st August 13:57

No I picked numbers out of the air to illustrate the point that to do anything scientifically you need to use stats which dictates if it's feasible before you crack out the step ladder and start dressing the Mrs up in sequins.

What you describe is not a scientific experiment. For tests like this the theory is decided, you simply need to test they meet a defined standard. You instead initially appeared to be trying to answer a question, this is much more difficult and expensive, which is one of the reasons definitive answers with safety gear is so tricky.

What you suggest is fine of course. But it's more like quality control.

yes I know all this............. so unless someone is prepared to donate some visors it will need to do. Even then they would need to be from the same manufacture, the same batch ideally, the same age, exposed to the conditions for the same amount of time, i could go on.

why am I doing it, one main goal really.

if it fails its a complete waste of time, but going on the calculations I have done this would far exceed and impact test for a polymer of this type that I have done before due to the force exerted and the area of impact.

I done a similar test on Volvo s40 spoilers, these were cracking due to stones getting flung up and hitting them. The force I should be delivering would be 3 times what they were tested to by increasing height, load and area of impact. Yes I had a dozen of them to fk about with so that was scientific and this really isnt. Also im basing it on what I remember to be the height, load and area of impact almost a decade ago.

BUT with this harsher impact if it passes one could reasonably conclude that the average punter doesn't need to worry about a stone hitting their visor if they buy an aftermarket one. probably.

Edited by moanthebairns on Thursday 21st August 18:31

[strudel]

Wouldn't it be easier to do some hardness or three point bend tests so you can get a reasonable range of results?

[Prof Prolapse]

Visor death match sounds much more fun. Sign me up for that.

[WaferThinHam]

graphene said:

We are getting a bit carried away with words 'science' and 'lab'.

A visor deathmatch is sufficient: Take a visor in each hand; smash together repeatedly until one fails. Declare the winner.

Now THAT'S science.

[moanthebairns]

visor conkers!!!!!!