Bushwalk Australia

Hey all,

I’m in the process of replacing all my guy lines on my MLD tarp that has Line Loc 3’s all around. Ron recommends their 2.7mm cord and says that they specifically designed it for the LL3.

I’ve been in contact with Simon at Tier Gear (still stoked on my Quenda) and they only have 2 or 3mm. Being a DCF tarp I would like to use the lightest possible cord to keep weight as low as I know it can be.

So, questions are....
Who has a decent amount they would like to part with or
Who knows where I can get some in Australia? I would like to keep my money local in these times and postage from the US just shocks me anyway.
Who has experience with other cord and can recommend something?

Cheers.

Not really that much difference between 3mm and 2.7mm.

Heh, Ted, that’s 10% - bet you’d like a 10% discount on all purchases,

I got some nice 2.5ish mm reflective cord off eBay recently, but it came from China, and it wasn’t as good as the thinner stuff I found previously.

And, I need to sort through all the moral and ethical issues about buying stuff cheaply on eBay, whether from the Chinee, or anyone else....

Another night.

A

Well if you want to measure in percentages, but what then is the manufacturing tolerance on the lineloc mouldings themselves?

If you're really interested I could be persuaded to measure a bunch of Linelocs to get an idea. But I think you can be sure that it's a lot less than 10%.

For cord the weight ratio is the square of the diameter ratio so (assuming similar material/construction) a 3mm cord will be about 20% heavier than 2.7mm. For example, MLD 2.7mm (polyester) is 5.4g/m; whereas PMI 3mm (polyester/nylon) is 6.5g/m. With a DCF tarp that would be noticeable.

I like the Hilleberg 2mm cord (Vectran/polyester). It's a lighter material as well as thinner and weighs 2.5g/m (my measurement). It's kind of pricey but so is DCF. The 2mm cord has some texture to its weave which provides friction but I can't say exactly how much force is required for it to slip in a Lineloc 3 under different conditions. I prefer a different line lock for that cord, the "2mm Line Runners" that Hilleberg uses. As a bonus they are ever so slightly lighter than the Linelocs (0.9g vs 1.3g).


edit: Couldn't resist...
I weighed 10 Lineloc 3s on a scale that is repeatable to ±0.002g:

1.276 g
1.273
1.279
1.275
1.271
1.280
1.274
1.276
1.271
1.274

It looks like the tolerance is less than 1%.

I think I've got some. PM sent

Orion wrote:I weighed 10 Lineloc 3s on a scale that is repeatable to ±0.002g:

1.276 g
1.273
1.279
1.275
1.271
1.280
1.274
1.276
1.271
1.274

Real ultralight bushwalkers file them all down to a consistent 1.270.

The best thing I have found for use in LineLoc 3s is a light tape. The tape I have been using comes from the lanyards that came with a heap of USB sticks - about 7mm, 3.4g per metre. Unfortunately I only salvage 75cm per lanyard but I would expect a 6mm (1/4 inch) grosgrain tape would work just as well.

Also if you want to average out the weight of items it is usually better to weigh ten in a single weighing and then divide by ten to get the average weight. Especially with light items each weighing will add any inherent calibration error to the total you rather than just one.

ribuck wrote:

Orion wrote:I weighed 10 Lineloc 3s on a scale that is repeatable to ±0.002g:

1.276 g
1.273
1.279
1.275
1.271
1.280
1.274
1.276
1.271
1.274

Real ultralight bushwalkers file them all down to a consistent 1.270.

Or do they not use them at all? Is the amount of extra line used for a knot going to weigh less?

PedalRoll wrote:Or do they not use them at all? Is the amount of extra line used for a knot going to weigh less?

Probably. But not very much. And it's not going to be enough to offset giving up the ease of adjustment that a line lock provides.

Mark F wrote:The best thing I have found for use in LineLoc 3s is a light tape. The tape I have been using comes from the lanyards that came with a heap of USB sticks - about 7mm, 3.4g per metre. Unfortunately I only salvage 75cm per lanyard but I would expect a 6mm (1/4 inch) grosgrain tape would work just as well.

How cool! I've never seen this stuff. The very few USB sticks I have don't have a lanyard.

Mark F wrote:Also if you want to average out the weight of items it is usually better to weigh ten in a single weighing and then divide by ten to get the average weight. Especially with light items each weighing will add any inherent calibration error to the total you rather than just one.

Agreed that averaging over ten items gives a better average than averaging over one item. But I don't see how averaging will eliminate a scale calibration error. If your scale is reading high you'll just end up with an average that's high. I'm sure you know this so it must be that I am misunderstanding you.

Agreed that averaging over ten items gives a better average than averaging over one item. But I don't see how averaging will eliminate a scale calibration error. If your scale is reading high you'll just end up with an average that's high. I'm sure you know this so it must be that I am misunderstanding you.

I am not a scientist but I am home bored. If your scale has an error of + 1 g and if your lanyard weighs 10 g then the error on 1 measurement will b e1g per lanyard. Multiply that by 10 and you have an error of 10 g for 10 lanyards.

If you weigh 10 lanyards a 1g error gives you a 1/100g error or 1g for 10 lanyards which works out to be an average error of 0.1 g per lanyard.

(How the *&%$#! does the quote function work?)

Some of you people aren't coping with ISO very well.

slparker wrote:(How the *&%$#! does the quote function work?)

Make sure that what you want to quote goes between [qu ote] and [/qu ote]; if you want attribution, start with [quote="slp arker'] (remove the extra spaces).
If you're interpolating comments between various sections of a quoted post, each section needs to be quoted separately:
[qu ote] Their bit. [/qu ote]
Your bit.
[qu ote] Their next bit. [/qu ote]
Your next bit. etc

It can be a little fiddly. Either copy/paste the bits you want individually, making sure each one has it's own [qu ote] [/qu ote] set, adding your comments between. Or quote the whole thing multiple times, deleting the irrelevant sections, and adding your own bits after each end quote marker. And then double and triple check it and edit when it goes wrong.

stry wrote:Some of you people aren't coping with ISO very well.

What is "ISO" in this context?

slparker wrote:I am not a scientist but I am home bored. If your scale has an error of + 1 g and if your lanyard weighs 10 g then the error on 1 measurement will b e1g per lanyard. Multiply that by 10 and you have an error of 10 g for 10 lanyards.

If you weigh 10 lanyards a 1g error gives you a 1/100g error or 1g for 10 lanyards which works out to be an average error of 0.1 g per lanyard.

What you say is true but it's not a typical case. Usually a scale calibration error is closer to being linear, where the error is proportional to the weight. So you might have a +1% error in the weight. In that case each item would weigh 10.1g and ten of them would weigh 101g.

Complicating this is that scales have limited resolution and precision. Perhaps your scale doesn't even read out to 0.1g, only to the nearest gram. In that case weighing a single 10g item would have a resolution uncertainty of ±0.5g in addition to the precision, which could be worse than 1g. In that case it is clearly better to weigh a whole bunch of items and then divide.

But if there is +1% calibration error that would still mean you'd read high. It doesn't matter if you weigh 1,000,000,000 items.


But I'm still not sure what Mark F meant.

"ISO" Isolation. As in Covid inspired mandatory staying at home away from other people.

Australians are notorious for slangifying words

Orion wrote:What is "ISO" in this context?

Complicating this is that scales have limited resolution and precision. Perhaps your scale doesn't even read out to 0.1g, only to the nearest gram. In that case weighing a single 10g item would have a resolution uncertainty of ±0.5g in addition to the precision, which could be worse than 1g. In that case it is clearly better to weigh a whole bunch of items and then divide.

But if there is +1% calibration error that would still mean you'd read high. It doesn't matter if you weigh 1,000,000,000 items.

There is an ISO (17052) standard to calculate measurement uncertainty.

To claim a device with a resolution of 1 has a 'resolution uncertainty of ±0.5' you would have to know that the device has some dither ... We are getting into NATA (https://www.nata.com.au/) territory here. I was a NATA signatory for some things. I never ever reduced a resolution from 1 to 0.5 even where the manufacture claims they have built in dither as you would need to prove it at each calibration interval and for the entire range of the instruments measurement capabilities. The overheads are simply too high for that to be worthwhile.

In metrology 'precision' means 'repeatability' it is not accuracy nor error though many think it implies the same, it does not.

I have spent months resolving measurement uncertainties ... on a domestic device subject to the weather variations .. the uncertainties will be high.

I thought I had it right, at least for typical consumer scales that round to the nearest displayable value. But maybe I'm overlooking something.

So what is the uncertainty contributed by the 1g display resolution of my inexpensive kitchen scale? It is actually more precise than 1g but is incapable of displaying sub-gram values directly.

Orion wrote:So what is the uncertainty contributed by the 1g display resolution of my inexpensive kitchen scale? It is actually more precise than 1g but is incapable of displaying sub-gram values directly.

As the resolution is 1 g it would have to be subjected to lots of tests to find out if it had greater accuracy that 1 g. The cost of professional tests for that would be many times the cost of the instrument. Certainly the average user is not going to weigh something with it and get resolutions less than 1 g.

I have 4 weight scales here at home.
One to weigh me .. I forget what its resolution is.

I have one kitchen scale 1 g resolution to 5 kg max weight.

I have one fine scale 0.01 g resolution to 500 g max. It will drift with temperature changes .. putting it in the sun is a disaster.

I have one luggage scale with 1 g resolution to 9,999 g then 10 g resolution to 50kg max.

The resolution is one contributor to the uncertainty of the instrument. There are tests call 'Pattern Approval' to ensure the instrument will maintain its expected performance with expected temperature, humidity etc conditions. This tests the design. The on top of that each instrument has to be 'calibrated' on a regular basis. Pattern approved instruments will cost more than what we buy from retail stores. Then you pay for 'calibrations' on a regular basis.

Warin wrote:

Orion wrote:So what is the uncertainty contributed by the 1g display resolution of my inexpensive kitchen scale?

The resolution is one contributor to the uncertainty of the instrument.

Yes, and my question was what do you believe it is?

I googled "resolution uncertainty" and the very first search result linked to a credible paper on the subject: "How to Determine Resolution Uncertainty", Richard Hogen, Cal Lab: The International Journal of Metrology (2017). Mr. Hogen consults with laboratories on the subject of uncertainty analysis and ISO accreditation.

The paper says what I said above, that in a case like this the resolution uncertainty contribution would be half the resolution.

Warin wrote:As the resolution is 1 g it would have to be subjected to lots of tests to find out if it had greater accuracy that 1 g.

I was talking about the precision specifically, not the accuracy. Determining the precision would take a lot of work, true. But demonstrating that the precision is better than 1g isn't difficult.

Orion wrote:

Warin wrote:

Orion wrote:So what is the uncertainty contributed by the 1g display resolution of my inexpensive kitchen scale?

The resolution is one contributor to the uncertainty of the instrument.

Yes, and my question was what do you believe it is?

"it" ? Meaning what?

* The resolution? 1 g
* The resolution uncertainty contribution? 1 g with a rectangular distribution and level of confidence 99%. That would be what would be feed into the uncertainty calculation for the total system. There is no proof of any "rounding" of this cheap instrument. So no reduction of the resolution fro uncertainty calculation could be applied with any confidence.
* The uncertainty of a reading? How long is a bit of string? Not possible to estimate without more information.
* The precision of the device? I don't know - there is no information on how repeatable it is over time, temperature etc.

On an inexpensive instrument there will be no manufactures statement on any information of the resolution rounding, dithering nor any other thing relevant to the resolution uncertainties. I have used many laboratory instruments, all with manufactures statements on their applications obtain better performance than the resolution. In all cases the only entry to the systems uncertainty calculation the full resolution was used with no reduction. All these calculations have been reviewed both internally and by external professionals with expertise in the relevant field. What you chose to do is up to you.

Orion wrote:

PedalRoll wrote:Or do they not use them at all? Is the amount of extra line used for a knot going to weigh less?

Probably. But not very much. And it's not going to be enough to offset giving up the ease of adjustment that a line lock provides.

Spoken by someone who is not a true Ultralighter!

Warin wrote:"it" ? Meaning what?

The resolution uncertainty.

What other means besides rounding would be employed to convert the internal estimate to the display resolution? I suppose it could be truncated instead but truncation is just another form of rounding. In any case, there is clear evidence of rounding with this inexpensive device.

Your answer of 1g is at odds with the author of the document that I linked above.

ChrisJHC wrote:Spoken by someone who is not a true Ultralighter!

Guilty as charged. Although I have been UL in the past. I've gone without tent. Or stove. I've used knots instead of line locks. In fact, I usually have a few short extra tie-offs on tents that I only occasionally use and those have no line locks.

Ultralight is not that hard anyway, what with modern materials and construction. The even lighter levels are harder to do.

But more recently I don't care so much. I figured out how to make my pack really light and when I went as far as I could the game stopped being as much fun anymore. I started adding certain items back. I think that's a common path.

Orion wrote:

Warin wrote:"it" ? Meaning what?

The resolution uncertainty.

What other means besides rounding would be employed to convert the internal estimate to the display resolution? I suppose it could be truncated instead but truncation is just another form of rounding. In any case, there is clear evidence of rounding with this inexpensive device.

Your answer of 1g is at odds with the author of the document that I linked above.

I sate again.. I would use the instruments displayed resolution, in this case 1 g. Note 'resolution' in metrology is "the smallest thing you can see". Can you see 0.5 g on your device?

The measurement element is in its pure from analog. The analog is converted to digital, the resolution of that conversion is what you are claiming to know. How do you know it?

The document has an example of a stop watch .. the resolution is not divided in that example. Why not in that case any yet others are? To me the document does not clarify how to determine if resolution of a digital instrument should be divided.

--------------
I would rather overstate an uncertainty than understate it. That is far safer.

I think the logic is that for a "counted" device the internal state could be anywhere between two digits and would always report the lower value. For example, if you truncate a real number to the nearest integer the difference could be as much as (almost) a whole integer. But if you round a real number the integer is never more than 0.5 off the real value. So what I said before about truncation being the same as rounding is incorrect.

Is my 7000x1g scale more precise than 1g? Is it rounding? I think there is evidence that both of those are true.

I used another scale and a set of calibration weights to examine this some time ago and I added some more measurements just now. My other scale has a resolution of 0.01g and is repeatable (my estimate) to ±0.02g. Here's what it looked like:



Now granted, this is a limited set of measurements and over a very small part of the scale's range. But I found it persuasive.