Standardizing the Wetness of Pens

[nkk]

Ok, all of this was added as an Edit to the first post to keep my thought centralized, but here is my response:

 

Ok, it seems I may have misstated things, or at least said them in a way that led to misunderstanding. My system would not be so that someone could say "model X is wet". My system would be able to say that "This specimen of model X has a wetness of N". Thus, different factors that were brought up as changing wetness, i.e. nib material, feed design, and feed material, do not need to be looked at. They will all be taken as given when I refer to a pen. Same with flow adjustments. My end goal is to be able to have this conversation:

 

Me:""Hello. My Pelikan is currently a B on he wetness scale, but I find that too dry. Could you make it a C?

Nibmeister: "Yes, I could. Let me find my standard pen of wetness C an compare, and then I can alter your pen"

 

The point would come where the nibmeister would not need a standard, as they would have a feel for what each wetness was. Mr. Binder, for example, probably knows his scale by heart. And he has gotten good enough at it that if I order 10 Pelikan M200 nibs in M with a flow of 6, they will all be very similar, because he inherently knows what to do from practice and repetition. The same goes for nibs. It can be hand polished, whatever. No matter what it is, you can say that the nib you are holding has a wetness of N.

 

Also, please lay into me and point out errors in logic, or an overlooked variable, or something. That is what makes me reevaluate my design and make it better. Or feel free to evaluate it yourself and gives some thoughts.

 

On the issue of different papers, this scale should be durable among all of them given that they do not bleed. I chose to standardize it for now, but that can always change.

 

For inks, I know about surface tension and whatnot. For Heaven's sake, I am in a fluid mechanics class now. I sort of just wrote one and not the other, because viscosity will change things, too. So thanks for pointing that out. I also know about the wetness of the inks being different. That is why I want to standardize the ink I use. When this is done, maybe I will move on and add an ink wetness term to it, and I will find the equation governing that term based on various variables. But that is for another day.

 

Pen pressure and angle points have been noted, and are being thought about. Thanks for pointing those out.

 

Gravity changes are not only negligible, but can be accounted for. The acceleration due to gravity is known to extreme accuracy for very many places on Earth. Additionally, if the feed and capillary action can fight gravity and write on a ceiling, they should in the same way stop it (they do) when writing the correct way, on a table. Thus, I am thinking the gravity term is actually going to just not be included. Again, please argue, as I may be wrong.

 

-Nkk

[Pfhorrest]

.

Edited February 7, 2010 by Pfhorrest

[OldGriz]

.... it is the same as someone stating that this chili is too hot and the other person stating it is not hot enough....

Wetness is an individual situation.....

 

Funny that you should mention chilli hotness, OldGriz. There is an actual scale for measuring that - the Scoville scale (devised by William Scoville in 1912). The unit of measurement is the SHU (Scoville Heat Unit).

 

Different individuals can tolerate / like different levels of chilli hotness. Similarly, different pen-users can enjoy / dislike different degrees of nib wetness. However, the wetness of a pen should be measureable objectively. Just my 2 cents' worth.

 

I know all about the scoville scale... I was talking about a personal perception of what is hot and what is not....

Personally, if I do not break out into a sweat it is not hot enough.... I am one of those masochists who add hot chili oil to the dipping sauce for fried dumplings in a Chinese restaurant and then drop some more onto the dumplings...

 

As for objective measurement of wetness, If you come up with an accepted wetness scale you will need to test every pen and nib variation available in the world today, including vintage pens to classify them according to the scale. And this will not work since even amongst pens with the same nib... i.e. Parker Sonnet with a Medium nib... there will be differences in wetness.

[hari317]

NKK, what will be the unit you will use to measure the wetness? litres/minute?

[nkk]

The units do not matter. As I said before, the units are trivially converted within an given dimension. The dimensions will probably be Volume/Area x length, which reduces to no dimensions nor units. It is a unitless, i.e. scalar, quantity. The volume will be volume of ink, the Area that of the nib (or line, I am deciding which is more practical to use as a standard), and the length that of the line drawn.

 

Than being said, units for the individual quantities will probably be mL for volume, cm or mm or length, and mm^2 or cm*^2 for area.

 

-Nkk

 

PS Just to clear something up, there is a fine distinction between units and dimensions. Not everyone here has made this mistake, but it is a cool thing to know IMHO:

 

Units are something like meter, foot, gram, second, pound, etc. They are quantifiable and can be measured against a standard or in some other way they are consistent for every use.

 

Dimensions are much more general. A dimension is something like length, volume, area, time, etc. There are many units that can be used for any given dimension. For example, area can be measured in sq. inches, sq. centimeters, inch x mile, etc.

Edited February 7, 2010 by nkk

[Eldan]

I don't think it's practical to introduce the area of the nib or the line. Both are difficult to measure accurately. Besides, nib area is one of those "characteristics of the pen" you intend to leave out. Dimensions of volume or mass per length are appropriate.

 

I think you'll have to take measurements at two different normal forces and average the results. This will help take nib flex into account.

 

PS. "Scalar" quantities can have units too.

[nkk]

I don't think it's practical to introduce the area of the nib or the line. Both are difficult to measure accurately. Besides, nib area is one of those "characteristics of the pen" you intend to leave out. Dimensions of volume or mass per length are appropriate.

 

I think you'll have to take measurements at two different normal forces and average the results. This will help take nib flex into account.

 

PS. "Scalar" quantities can have units too.

 

No, because if we just do volume per length, we get a wet fine being the same wetness as a dry medium. I could do that, but it seems much better to have wetness standardized to the area. Actually...there could be two wetness scales, absolute and normalized. Absolute wetness would be what you prescribed, and normalized would be absolute over area. They would only differ by a 1/Area term, so that is not bad. And they would provide more comparison avenues to be explored in between models, nib sizes, etc.

 

As for two different normal forces, that is also a good idea. Thanks.

 

As for scalar, yeah, I know. I was tired and had previously erased some derivation with a dot product in it (or angle of nib to paper. I have decided that that derivation was wrong, but I may have a new one). The scalar was me missing a remnant of my old statement and istead of erasing it, working it into a new statement where it is wrong.

 

Thanks to all those that have helped. Making me rethink and justify my choices has made me rethink a coule of things, so this is only getting better.

 

-Nkk

[Eldan]

No, because if we just do volume per length, we get a wet fine being the same wetness as a dry medium.

 

I don't understand why that's a problem. Compare F with F, M, with M.

[crowdog]

I think it would be easier and less variables to just measure pen flow. For example. Take a precise ink measure and add it to a pen. Then draw a measured line continuously, and then measure the amount of ink . The variable would be the ink left in various nibs, but it would give you a standard. I just think that you have way to many variables to have any type of standard measure.

 

If you do go with your way, try to make an objective criteria, that someone else could follow, and have them duplicate it. If your stats come close, I think you have a usable scale.

 

Marty

[nkk]

No, because if we just do volume per length, we get a wet fine being the same wetness as a dry medium.

 

I don't understand why that's a problem. Compare F with F, M, with M.

 

Compare a Japanese XF nib with a Lamy 2000 XF. They are not the same, and to compare their wetnesses jut because they are both XF is misleading. The Lamy has a larger surface area on the part that hits paper, and that needs to be accounted for. Now, I did say that we could use an absolute measure of wetness, i.e. vol/length, and that would compare them. But it does not standardize the results to any area. Because both could prove useful, I suggest having both absolute wetness (your suggestion) and a standardized wetness (my original idea). As I said, your idea and my idea only differ by a factor of 1/Area, and thus are not that different. Yours may be more useful in comparing two pens on the fly, but mine may have more uses in comparing wetness across a certain nib size (area, not rated size like F, M, etc.). The comparison is what led me to this quest, so it made sense to me to add the 1/A term.

 

-Nkk

[Eldan]

Why is it misleading? It's no secret that Japanese nibs run finer. If my question is, what's the wettest XF, an absolute scale has the answer. If my question is, what's the wettest Japanese XF, again an absolute scale has the answer.

 

Could you pose a question that can be answered only by your area normalized number?

 

I'd also be interested to see how you propose to standardize the measurement of nib "area" in contact with paper. And what if flow is not linearly dependent on nib "area?" If flow is more dependent on spread or flex your dimensionless number will be unreliable.

[nkk]

Why is it misleading? It's no secret that Japanese nibs run finer. If my question is, what's the wettest XF, an absolute scale has the answer. If my question is, what's the wettest Japanese XF, again an absolute scale has the answer.

 

Could you pose a question that can be answered only by your area normalized number?

 

I'd also be interested to see how you propose to standardize the measurement of nib "area" in contact with paper. And what if flow is not linearly dependent on nib "area?" If flow is more dependent on spread or flex your dimensionless number will be unreliable.

 

My scale would allow me to look at a pen whith wetness 4 and say it is wetter than a pen with wetness 2. The 4 may be a fine, and the 2 a Bold, so the 2 probably puts out more ink. But my scale represents how the pen feels when you write with it--i.e. your gut reaction (at least I hope) as to whether it is wet or not. That is what I wanted to do. A bold that puts down the ink of a M will be dry, bu a F that does the same will be wet. I wanted to correct for that. Your scale will have an associated normal curve for every nib size. You can say a pen has a wetness of 6, and that is fine. But if I want to buy 10 pens, from XXXXF to BBBB, they will all have a different wetness on your scale. But on mine, I could say that they all should have a wetness of 1.414, and that could be done easily. I could then tell everybody that I prefer every nib I own to have a wetness of 1.414, and we could compare. With yours, I will have to say that I want all fines of mine to be 1.732, all mediums to be 2.73, and all B to be 3.1415. I feel my scale makes it easier to talk about. That may be a preference of mine, though, and you would find it easier to talk in your scale. I think it makes more sense the way I am doing it, but I very much see how yours would be helpful.

 

As for variance with flex, I am still trying to figure that one out. Honestly, I do not have a flex pen to try anything with, so I am at a loss how to even do that. I do want a nice vintage flex (or even modern semiflex), so this may be a good excuse...

 

If it is not linearly dependent on area, I will figure that out later. I have yet to do the dimensional analysis on this, and possible apply the Buckingham-Pi theorem.

 

If you do not mind me asking, what science/engineering credentials do you have? You seem to be asking some of the right questions to make me think about this, which tells me you either have some experience, or are just randomly really good at asking questions.

 

-Nkk

 

EDIT: lol, I just realized that my numbers are all significant in math to some degree. I did not do that purposely, they just popped into my head as I wrote;

1.732 is sqrt(3)

1.414 sqrt(2)

3.1415 pi

2.73 e

Edited February 8, 2010 by nkk

[crowdog]

So, I am not a mathematician or engineer, but I am a human behavioral expert. Humans have huge variables in determining behavior and being able to measure changes in behavior, means we try to limit and or account for as many variables as possible. If it is not observable and measurable it didn't happen.

 

I was intrigued by your "study" at first, and I thought there would be some real data, but I am coming to the conclusion that you just want to have a little engineering/math project to show how smart you are.

 

I am not trying to be demeaning, but you are putting wayyyyy to much analytical thought into outcomes, without first doing simple variable analysis, of limited trails, with very limited expectations.

 

If you want to do all your cool engineering stuff, that is great, but this doesn't seem able to be replicated, or even a standard that anybody but you can use. So even your comparison example will be about as good as saying.."I like mine more wet". There is nothing quantitative about it, even with your math.

 

And, the reason that I put what I did for a living, is that some of us have done experimental analysis with multiple variables, and we are not engineers.

 

Marty

[nkk]

So, I am not a mathematician or engineer, but I am a human behavioral expert. Humans have huge variables in determining behavior and being able to measure changes in behavior, means we try to limit and or account for as many variables as possible. If it is not observable and measurable it didn't happen.

 

I was intrigued by your "study" at first, and I thought there would be some real data, but I am coming to the conclusion that you just want to have a little engineering/math project to show how smart you are.

 

I am not trying to be demeaning, but you are putting wayyyyy to much analytical thought into outcomes, without first doing simple variable analysis, of limited trails, with very limited expectations.

 

If you want to do all your cool engineering stuff, that is great, but this doesn't seem able to be replicated, or even a standard that anybody but you can use. So even your comparison example will be about as good as saying.."I like mine more wet". There is nothing quantitative about it, even with your math.

 

And, the reason that I put what I did for a living, is that some of us have done experimental analysis with multiple variables, and we are not engineers.

 

Marty

 

I will sort of ignore the accusation of being a show off. The truth is, I will have to probably buy the things to measure all the variables. Option 2 is somehow getting into a lab on campus, but that may prove difficult. I can weigh paper discreetly during regular class/lab sessions, but for everything else (measuring nib/line width), I will need precise instruments that I do not have and cannot pay for. Before I go asking for any favors, I want to know exactly what I am looking for. My conversation with Eldan is off topic a little, in that we are getting ahead of ourselves. If you look at my OP, I am still looking at which variables to examine.

 

Also, in a way, you may be right. I want a governing equation, and that means taking everything into account. So perhaps this will not be realistic. However, if you ignore all the negligible variables (changes in gravity and pressure), you will reduce me equation into a widely usable standard, which seems to be what you want.

 

You suggested earlier (and I missed it then) about just drawing a line with a known amount of ink. That may work, and actually was a testing method I was considering. The difference was I was going to weight the ink, and find the change in weight of the paper, and then using density find the volume of ink put out, instead of measuring it before I put it in the pen. As for an objective criteria, well....any suggestions? Contrary to what you think I am trying to do, I am not trying to just mask me bragging about how smart I think I am. Thus, I am open to any suggestions you have on that, as I am at a loss. When you ask me for something objective, I think hard numbers. Which leads me to a formula, which gets me to my OP.

 

And do not shortchange yourself. I said scientist or engineer, and by any metric I can think of (other than only physical sciences allowed), behavior analysis is a science.

 

-Nkk

 

PS As a side note, all the variables that you think are superfluous probably are. I think they will drop out as negligible, and thus the equation will essentially reduce itself. But why not start with the most complexity, so that nothing is left out?

 

Also, the more I think about it, the more I am irked by the fact that you think I am just showing off. I hate bragging, and the though that this is coming off as me saying "look at me, I can do the maths and physics really well" (said in snotty little "look ma, no hands" voice) really gets to me. To anyone out there, please accept me sincerest apologies if this seems to be bragging. I assure you, it is not.

Edited February 8, 2010 by nkk

[Eldan]

It is not my credentials that should matter to you but the weight of my argument. I wonder how many pertinent questions I need to ask before you reject random chance!

 

You can't "worry about area" later. The area assumption is central to your hypothesis that one can effectively normalize by "area." The very first thing you should be doing is determining how and with what accuracy area can even be measured. Why not choose diameter? Is it a fetish for dimensionless values?

 

It seems you want to be able to speak of the "wetness" of a nib size relative to the universe of nibs of that size. The way to do that in a dimensionless way is take large samples of pens, determine a sample distribution of flow for each nib size, then score the pens by percentile or Z-score.

[Eldan]

and find the change in weight of the paper

A little thing called evaporation is going to bedevil your efforts. If you weigh the pen on a milligram scale both after filling and after writing some standard length sufficient to consume approximately 0.5 ml of ink you will have enough resolution for your purposes.

[crowdog]

I wasn't trying to be mean, or demean you. So how about instead of "showing off" that you seemed to want to talk about science, more then you wanted to start your experiment. That is what it seemed to me.

 

As I said with kids, we do the most simple, basic limited variable test we can do. A probe, is what we call it, then we add variables, to see if the behavior will increase in the future. Next time, we do a probe again to see if any change has occurred. We take massive amounts of data. Antecedent, Behavior and consequence data.

 

So my concern with your experiment was one of outcomes. To start with a small experiment to test your hypothesis. If that proved to be promising, then you could start adding other variables (size of nibs, etc all the other factors) to then see if you hypothesis of ink "wetness" was valid.

 

As others have mentioned, it then becomes more a problem of definition then calculation. If you don't account fully for at least the amount of ink on paper, as "wetness" then even other variable factors will be suspect.

 

Also I think you really have two different things here. On the one hand your high end experiments could be really cool for ink saturation, and flow and fluid dynamics, all that.

 

On the other hand, to compare pens as a whole, you are really looking at basic statistical analysis, samples,populations and t-curves. If your sample size is broad enough, you can then establish a significance in the population.

 

Marty

[hari317]

The dimensions will probably be Volume/Area x length, which reduces to no dimensions nor units. It is a unitless, i.e. scalar, quantity.

 

How did you arrive that the dimension for wetness will be Volume/(Area x length)? area and length of what?

 

The first task is to define wetness. More wet = more ink output per given time?, keeping rest of the variables same? e.g. I remember a post where Binder draws a circle on a piece of paper towel, and the time taken to fill the circle with absorbed ink might give him a measure of wetness of the nib-feed-ink-paper combo.

[troglokev]

Eldan has the right idea: you need a suitable standard measurement for wetness, rather than a theoretical description.

 

The machine in the link is for ballpoints, but I'm sure a similar arrangement could be made that exerts a more apropriate force at a reasonable writing angle.

 

The hard part then begins: getting the manufacturers to use them, and to publish the results.

[ImolaS3]

NKK

 

I think you will be unable to do what you are setting out to do. I am an academic and scientist of 30 years experience and would have to say that there are too many variables you cannot consider in your formula which are specific to a user such as angle held (vertically), angle of rotation of the nib, user pressure, frequency of use of the pen etc.

 

Also, it is a trivial and very easy task to make a dry pen write wetter - a sharp craft knife and 10 seconds work with no nib expert being called upon (I have done this on several pens)