[Closed] Monitoring Components

SPLAT

Graham where are you man, I thought I'd at least get your support in this!!!!

Wow,

How badly did I miss read Graham's thread, no one has even bothered flaming me.

This reminds me of Fight Club...

I'm still perservering. Gonna keep everyone who is interested updated.

That'll be me then! I'll update tonight when I get home from work.

Isn't this the wrong way round, you're writing a program that'll tell you when to replace parts? Surely you want a way of checking how many miles/revolutions a part has done by the time it's knackered, rather than routinely replacing parts after x miles, which tells you nothing about component longevity.

You might be better setting up a blog where you can document progress and offer downloads of the finished code etc?

I really can't see the point, you can tell when I part needs replacing by looking at it, how it feels or works. Not sure how a database is going to tell me, I mean a few hundred miles in perfect dry conditions will be a big difference to 100 miles in grinding paste mud of winter.

Or two miles in the peak. Or if badly fitted, not bedded in properly etc. Valiant idea, but I think there are too many variables

Like Drac I can't see the point of an application to tell me when something needs replacing but I might, in theory, be interested in something that let me log how many miles (or how long) something has been used before it needs replacing.

What language are you planning on learning? Rather than starting from scratch have you considered finding a suitable open source project and either volunteering to join the development team or extend the existing application with you required features.

For example you can develop your own plug-in for [url= http://www.zonefivesoftware.com/sporttracks/developers/?sid=db249d8bb1de7758cc9a83ceb6b4996e ]SportTracks [/url]although I'm not sure if you would be able to decompose a bike into a hierarchy of components.

You have kind of missed the point of the question, but answered it anyway.....

I was asking what people would want to see on the top level and giving a few random ideas. The problem with asking for too much detail to early is you can so easily lose sight of what the intended outcome, ie what information the user is give, is supposed to be. I like to start with what the user wants to see and then work out how to give them that in the most accurate way possible!

I would set up a blog, but this is just as useful to me, and I can also gauge how many people are likely to use it by the replies I can get on a well known well attended forum rather than trying to pursuade people to view my own web page!!!!

As far as programming languages go, I'll be using Python as it is 100% free and open source, how compatible it is with smart phone stuff I don't know.

Do people want a smart phone app? I personally don't, but I can easily look into it.

Jeez, who would be so anal as to need a ****phone app to tell them their chain's knackered?

Just get out and ride man!!

Me 😀

Anyway, needs and wants are two very different things.

How could it possibly be accurate?

You would have to factor in rider weight, rider style, rider power, climate, type of terrain, type of mud, etc etc etc. Too many variables.

Oh and then theres the problem of littigation? If you programme says your component will last another 30 hours use and then snaps and kills some one after only 10 could they not sue you?

If not then the app is pointless and you may has well go back to taking a bit of a risk as we all do normally.

Maybe all you can do is publish manufacturer data which says how many hours abuse they think it should with stand?

How could it possibly be accurate?

It couldn't, that's why it's a bobbins idea 🙄

But as above, how does any app know if a chain's knackered, all it can do is tell you it's done x miles - at the very best you can average chain wear, but that's going to be very inaccurate. A chain may last 10 times as long in some conditions compared to others.

Anyway, needs and wants are two very different things.

In this instance it would appear nobody either 'needs' or 'wants' it

You seem to be assuming what the end result will be.

Graham, I kind of need to learn from scratch,and this way allows me to learn doing something I find quite interesting even if noone apparently wants it. But you'd be interested in something which would show you how much your bits have been used?

You would like something that says this component has been used for this long?

"You would like something that says this component has been used for this long?"

Why ? You change a chain when it starts to slip, breaks etc, not when it has done 1000 miles etc.

Ive got bars and stems that are over 15 years old that I still use. I dont know how many miles they have done, but I know they are older so higher chance of snapping, but telling me how many miles they had done wouldnt quantify anything as 10 years of me using them might be equivalent to some one else using them for 2 years.

I think sh1tforbrains deserves an award for his outstanding achievement in missing the point.

Just get out and ride man!!

DrDomRob, I'm not sure exactly what I want, I've just got vague ideas at the moment.
I'm currently logging my rides on Endomondo, BikeJournal and Fetcheveryone.
If you look at [url= http://www.bikejournal.com/profiles.asp?rname=VeganGraham ]my profile on BikeJournal[/url], you can see my distance for each bike.

I think what I want is some sort of calendar where I can make a note of when I fit a component and when I replace it, then easily work out the distance between those two dates.

Actually, it may be a bit more complicated than that.
To give a simplified hypothetical example;
Let's say I start on 1.1.11.
On 20.1.11, I swap to my spare wheels (wheel set 2).
On 22.2.11, I swap back to wheel set 1.
I swap my wheels several more times over the next few months.
On 3.2.11, I fit a new chain.
On 8.8.11 I fit another new chain
On 9.12.11 the bearings fail on rear wheel 2.

I now want to know how many km I did on that bike on wheel set 2 between 1.1.11 and 9.12.11.
I also want to know how many km I did on that bike between 3.2.11 and 8.8.11 to check the chain life.

Most of all, I want it to be easy to use.
I think I might, just about, be able to do something with a spreadsheet that would cover it.
Something like BikeJournal with a drop down menu where I can set the defaults would be a lot easier though.

If you register on BikeJournal and add two or more bikes, you will see what I mean about the drop down box when you add a ride.
You can set which bike it defaults to, then scroll down to a different one if you want.
What I would like is further customisable sub categories for components.

If you register on Fetcheveryone, you get the option to assign a pair of shoes to each run. It then logs your distance against those shoes.
They did a survey recently where they used this information to compile all sorts of data on shoe life, I think it was based on over 30 000 individual runs.
Just as with the issues raised about bike component life, you're going to get light people running on grass and heavy people running on gravel, but with a big enough sample, it should smooth out the extremes and give a reasonably reliable guide to shoe life.

But that then makes it useless for the light person on the grass, heavy guy on gravel, or anyone at Swinley at the moment!

I still have no idea why you'd want to know you'd done x miles since you put bearings in your other wheels or how you think a spreadsheet can tell you chain needs replacing.

No it doesn't.
Let's say all the light runners wear Nike, all the heavy runners wear Adidas.
This would slew the results and make Nike look like the longer lasting shoes.
With a sample of 30 000 runs though, that's pretty unlikely. There's going to be a good mixture of all types of runner, surface and shoe.
They did make some allowances in the data for trail running shoes as it was to be expected that they wouldn't last as long as road shoes.

I think you, and a few other people, are missing the point as well, Drac.
I'm not looking for something to tell me when to replace a component.
This isn't like helicopter maintenance, where if a component is tested to failure at x hours, it is routinely replaced at x/2 hours in service.

The thing that inspired me to start this was getting two pairs of Egg Beaters back from repair.
Look at any thread on here about Egg Beaters and anyone will tell you they break and fall apart.
How long do they last before they break though ? Nobody knows.
My idea, once I've fitted the Ward titanium spindles to them both, is to make a note of the date I fit them to the bikes and record the distance they last before I replace the bearings.
I then thought I might as well extend it to chains and other components.

I think you, and a few other people, are missing the point as well, Drac.
I'm not looking for something to tell me when to replace a component.

But that's what this thread (which you've now commandeered) is all about, your own thread was about logging mileage, but the OP's idea here is some sort of notification system.

My idea, once I've fitted the Ward titanium spindles to them both, is to make a note of the date I fit them to the bikes and record the distance they last before I replace the bearings.
I then thought I might as well extend it to chains and other components.

But what are you going to do with that data? Assume that every other set of Egg Beaters will last the same length of time and replace accordingly?

what would you like displayed (ie replace part now, this part has done x miles...)

But what are you going to do with that data? Assume that every other set of Egg Beaters will last the same length of time and replace accordingly?

Again... You may not be, but this thread was about just that.

If you're not using it for that what use is the data?

You should develop an add-on whereby users can record the numbers of trains that they see while on their bikes too, you'll be on to a sure winner there !

Yup I was talking about the OP and a bit if Midlands. You may have noticed that there's more than one of you taking this rather odd idea.

Ok, it seems TQG was the only one who actually read my original post properly.

I asked what people wanted, I gave some suggestions that is all.

So Graham, at the very top level you want to be able to select a component and know how much it has been used. It would make sense to do this in miles/km.....

Is there anything else that you would want to know about the part? Would you for instance want to specifically about "rear wheel bearings" and how long a particular set has lasted, or would you want to know about a rear hub and how many times the bearings have been replaced and at what interval? Very subtley different, or the option to choose between the two?

Interestingly I am just going through bike journal and it would appear they are starting to add this sort of functionality.

So far they have the ability to add chain and tyres.

Dom

At the very top level, what would you like displayed (ie replace part now,

I think it's been made quite clear by those that read it that this is not necessary of realistically achievable.

1. You have a wish list for a program to help you monitor your bike components. At the very top level, what would you like displayed (ie replace part now, this part has done x miles...) and how would you like it displayed (Graphical, diary reminder...)

Drac - if you are going to quote something quote it properly.

I used the term "ie", apparently incorrectly, to mean "for example". So I apologise for that!!!

I was offering suggestions of what people may want rather than dictating what they get. I actually completely agree with you, but one thing I have learned is NEVER to throw out an idea until it has been fully investigated.

I was once involved in looking at different methods for vehicle recovery, I said no idea is a bad idea, someone amusingly suggested we use the A-Team. So we spent 5 or so minutes in the next round investigating whether this was possible or not. Obviously it wasn't.

"I was once involved in looking at different methods for vehicle recovery, I said no idea is a bad idea, someone amusingly suggested we use the A-Team. So we spent 5 or so minutes in the next round investigating whether this was possible or not. Obviously it wasn't."

I could have told you that in 5 seconds ! Do you spend your life pondering pointless things? Sound slike you should work for the civil service 🙂

[i]one thing I have learned is NEVER to throw out an idea until it has been fully investigated.[/i]

I'd like my new parts delivered to me on the trail, by flying pigs 0.1 of a second before the old one fails. Said pigs to be accompanied by a phalanx of monkeys fully equipped to remove the old part and fit the new one. Whilst I'm waiting I'd like a double espresso and a piece of fruitcake.

Oh, and if they can wash the bike before handing it back too, please.

it's all about data collection. Data is valuable stuff. I like the idea of logging the use of a bike component. However, I would be loathed to detail each component on my bike before or after I rode. Therefore your application should allow very easy data entry (i.e. a checklist of what you are using).....and some sort of mechanism to relog if you make an error.

Drac - if you are going to quote something quote it properly.

I did I quoted the part I was answering.

I learnt along time ago don't waist time on something that isn't needed or necessary.

I learnt along time ago don't waist time on something that isn't needed or necessary

what you doing here then?

Haha @ Whippersnapper.

Wwwaswas - it's in the pile of things to look into, although I fear getting hold of the pigs and monkeys may be the downfall of that idea.

what you doing here then?

Waiting time whilst I repair my laptop. 🙂

DrRomRob - I'm prepared to compromise on the pigs and the monkeys if you think it'll help.

Might donkeys and ostriches be a goer?

wwaswas - I'll have a look into it.

I'll start putting together something which defines the requirements, I'll post it up as soon as possible... probably tomorrow or sometime at the weekend when the missus is back on nightshifts and I can unleash the inner geek without fear of reprisals!

I haven't read all the thread so I might be getting wrong end of t'stick but, I think there may be some mileage in some form of online community thing where everyone logs there rides (with suitable variables like weather conditions, hardcore XC or light pootle down a railway line etc) but also logs what components and when they are changed/repalced etc. You could also put in a description of the failure mode etc.

Then the clever bit would be for all this info to be analysed to give realtime, real world, genuine reviews or ratings for components. You could then use this info to make informed decisions about upgrades, service intervals etc etc.

I would be interested in something that told me that a (on average) a pair of Avid pads lasts about 20hrs riding in the Peaks in the winter but a pair of Superstar ones last about 15hrs, for example.

The sample size would have to be vast for that to mean anything though.

Not really, you can get statistical significance from relatively small samples, 10 would probably show some sort of relationship, just with quite a wide tolerance.

njee20 - You have to start somewhere!!

Poisonspider - it would be quite difficult to understand what the width of the error margin might be since if one rider who drags their brakes puts information in 10 times, and 1 rider who doesn't only inputs 9 times, then the first rider will hugely affect the spread of data.

However, you could be very cunning and only input the riders average for anyone place, ie although the first rider has done 10 rides, they only enter into the global statistics once, using their average.

In which case the sample size would only be 2.

Anyway, this is for a bit later down the line. Good idea though.

Also, from what has been mentioned so far, it would appear an online application would be more beneficial to people than a standalone application...?

I know Python is used to drive some websites, but I don't know how effective it is considering the amount of data that MAY be generated adn therefore the size of the database required....

The sample size would have to be vast for that to mean anything though

I've had another look at BikeJournal and there's the option to add tyres or chains.
That's pretty much what I'm looking for so I think I'll have a go at cheating a bit and using those categories to enter pedals and wheels.
In fact, I'll email them and ask if they can add User Defined components as an option.

Cool,

I am going to carry on with this anyway cause I need to learn the language.

Dom

I think there's still room for something aimed solely at component monitoring. BikeJournal is more of a training log.
Being able to import directly from a Garmin or .gpx file would be handy.
User customisable drop down boxes listing any number of bikes and components and the ability to calculate the total distance of any component between any two dates would be a start.

Hi Guys,

sorry, had a really hectic weekend. Will try and get something up before I head off to Rome on Thursday.

Dom