Hypothesis: Pressing causes long-term damage to comics 2 2

[bronze_rules]

Thanks for your support, buttock.

 

I found an old Nedor funny animal (have no recollection of where I got it from) that I'd feel comfortable experimenting with. I was thinking of trying to bake the individual pages for different durations and then see if the bend-test works at all. If it does, a next step could be to gather a sample of books and send them to FFB for cutting and pressing.

 

I am sure lots of people will be laughing if this doesn't work. All I am saying is this is the best _I_ can think of, and as of yet I have not seen any alternative suggestions. I will be very busy over the next few days so there is plenty of time to contribute ideas before my Nedor gets baked.

 

 

tb,

 

Don't go sticking the Nedor into the Viking oven at home just yet. The aging ovens used to conduct these tests are special pieces of equipment that are temperature and humidity controlled. If we're going to destroy books, let's do it properly, in the name of science!

 

Since a few people expressed interest in this, I decided to sacrifice a few tedious hours on trying the bending experiment that I had read about. This is about as low tech as it gets: I found two very worn Mickey Mouse Magazines from the late 1930s: one with brittle pages and one with cream pages. I picked books with relatively bad pages to avoid the need for an aging oven. I then cut the books in half, removed the staples, and folded each sheet of paper around the spine until it broke. I tried to do the folding as carefully and uniformly as possible, but, obviously, significant variations in the numbers are to be expected. My hope was that the results would fall into a sufficiently narrow range that any significant difference between pressed and unpressed books would show up given a sufficiently high number of samples. None of these samples were pressed: I did this only to get an idea of whether it would make sense to proceed with the cutting-in-half experiment that FFB suggested.

 

The first book was so brittle that only one interior wrap was usable:

Upper half: 79 folds

Lower half: 72 folds

 

For the second book, I tried 5 interior wraps.

Page 1: Upper: 164 Lower: 239

Page 2: Upper: 339 Lower: 297

Page 3: Upper: 315 Lower: 374

Page 4: Upper: 299 Lower: 180

Page 5: Upper: 212 Lower: 271

 

Though these only are preliminary results, my own conclusion is that this brain dead technique, which everyone with enough patience can perform at home with no fancy equipment, just might be able to detect whether pressing a Golden Age book causes measurable damage to the spine. It would take a large number of samples, though.

 

I'd be curious to hear your thoughts, FFB?

-----------------------------------------------------------

It's generally a good idea to have a minimum of about 30 samples for any statistical

test. There's no statistical difference between the upper and lower set of the 2nd book

(as expected). The first and 2nd book (brittle vs. non) are definitely statistically different at <1%, but it is pretty much meaningless to compare only the two samples. It would be nice to have this applied to say 30-100 samples (books) with and without pressing (hopefully, of similar type, or brittle vs. non-brittle, but classification attributes need to be sorted), then we could say something more useful.

 

 

Also, I assume the goal would be to be to run the pressing on the bottom half, with top half unpressed, then apply folding tests and tally data? One other thing is that there might likely be differences in the immediate pressed data; it's important to take data over

a long period of time I think (as in reality, most pressed books might retain their pre-pressed elasticity given enough time to dissipate any heat or moisture induced in the short term?) They might even show the opposite, as steam and/or moisture could make the pressed book more pliable under this test, and thus make it appear better with regards to bending/folding strength.

Sorry, if I'm chiming in as I'm on the computer. I will be glad to help with any data analysis if you guys want.

 

Your folding endurance for the 2nd (cream)book comes out to an average of 2.12, which seems inline with some professional testing (haven't looked at the details of what the paper type they were testing). 1st book FEavg = 1.88.

 

http://www.pira-testing.com/uploads/public/documents/Downloads/Paper/PPS%20Dec%2008.pdf

 

One more idea. Ideally, we should take as many pages as possible for each book (top 1/2 pressed vs. bottom non-pressed), use that as one sample set to get an average fold endurance for that book. Then, take 30-100 more books and generate a sampling distribution of pre-and post pressed, based on the averages of each FE/book. Lastly, compare the sampling distributions using the same t-test to look for statistical significance in differences.

The books should be around the same time frame, but we can empirically observe if they are close enough in FE response, so that might not be an issue.

[tb]

BR, thanks for taking this seriously. I should have made it more clear that I certainly am not making any claims that this proves anything. I merely found it encouraging that the samples for the second book were as close as they were and think this potentially could yield some interesting answers given enough samples. Not least since it is the first numerical result that anyone ever has tried to produce to my knowledge(?). Even the handful of tests that I did took several hours of immensely tedious (and painful) work and I just don't have more time to put into this. Hopefully there will be easier methods.

[bronze_rules]

BR, thanks for taking this seriously. I should have made it more clear that I certainly am not making any claims that this proves anything. I merely found it encouraging that the samples for the second book were as close as they were and think this potentially could yield some interesting answers given enough samples. Not least since it is the first numerical result that anyone ever has tried to produce to my knowledge(?). Even the handful of tests that I did took several hours of immensely tedious (and painful) work and I just don't have more time to put into this. Hopefully there will be easier methods.

 

You are on the right track by collecting data, and your numbers seem reasonable compared to some of the numbers I've seen reported (in lab example I attached, for example). They have some types of instrumentation to run these tests much faster than hand.(200 folds/minute) I believe they are called schopper folders and/or M.I.T testers.. I'm just learning all the jargon for paper testing as well, so bear with me as I become familiarized.

 

Also, there are some rules of thumb for emulating accelerated aging via temperature in this paper: http://www.naa.gov.au/images/batterham-rai_tcm2-13043.pdf

3days @ 100C ~ 25yrs normal library storage conditions.

 

It would take a lot of data, but if you (or FFB or others) could get the temp/acc. aging emulation going pretty well, this would be a nice type of graphical analysis to have the pre and post pressing best line fits for comparison.

[tb]

Thanks for the very interesting link. It seems like it should be very doable to find the answers with these techniques.

[namisgr]

Just keep in mind these tests are extrapolations of what might happen to comic paper after a pressing and decades of aging. This is much like the testing of potential carcinogens, performed using higher doses of test substance than are found in nature in order to extrapolate to real-world conditions of low dose-long term exposure. The data that emerge from such extrapolations are informative, but never definitive.

[bronze_rules]

Just keep in mind these tests are extrapolations of what might happen to comic paper after a pressing and decades of aging. This is much like the testing of potential carcinogens, performed using higher doses of test substance than are found in nature in order to extrapolate to real-world conditions of low dose-long term exposure. The data that emerge from such extrapolations are informative, but never definitive.

 

Agreed. That's why I mentioned it is important to take it over some long period of time to be truly accurate. However, according to the studies in the paper I mentioned, it is a close approximation to reality (under typical storage conditions). Also, it is good enough that they use it to simulate shelf life for medicine and food products.

 

But, there are a lot of issues to be thought through such as the effect the oven has on the h20 content of a freshly pressed item (i.e. does it remove the additional h20 content? etc.,).

It's also important to determine if there is enough similarity between items(dates/conditon/etc) to be used in the overall test set. We could run a few tests like tbs to make sure they are all statistically similar enough (pre-pressed) to begin with.

 

I was also thinking, that it would be a good idea to come up with a proposal of the most feasible tests and attributes, including order of test, for each observation up front, so that we don't waste comics in the process. Ideally, it would be nice to have a table of attribute properties (fold endurance being one), so that we could run cross validation tests to see if ultimately, we could classify out of sample pressed vs. non-pressed items with better accuracy than guessing . Although, I still think these (mechanical) types of tests are crude compared to modern spectral type tests. Another nice qualitative (nominal) attribute could be to note the OWL color of the iterm pre and post test (although, again very crude, but possibly enough to add to classifier decision?).

 

It is also important to make sure the 'same' pressing parameters are applied to each sample (i.e. const heat, const moisture or same apparatus/settings). Since we are limited on samples, I would suggest a press that has steam or moisture (since FFB said most don't technically use steam) applied, since it is likely a good start to find distinguishing numbers vs. cold/dry.

 

Another idea for a very cheap and easy crude PH test, would be to use one of those gardening soil PH meters .

[tb]

[...]

 

But, there are a lot of issues to be thought through such as the effect the oven has on the h20 content of a freshly pressed item (i.e. does it remove the additional h20 content? etc.,).

It's also important to determine if there is enough similarity between items(dates/conditon/etc) to be used in the overall test set. We could run a few tests like tbs to make sure they are all statistically similar enough (pre-pressed) to begin with.

 

I was also thinking, that it would be a good idea to come up with a proposal of the most feasible tests and attributes, including order of test, for each observation up front, so that we don't waste comics in the process. Ideally, it would be nice to have a table of attribute properties (fold endurance being one), so that we could run cross validation tests to see if ultimately, we could classify out of sample pressed vs. non-pressed items with better accuracy than guessing . Although, I still think these (mechanical) types of tests are crude compared to modern spectral type tests. Another nice qualitative (nominal) attribute could be to note the OWL color of the iterm pre and post test (although, again very crude, but possibly enough to add to classifier decision?).

 

It is also important to make sure the 'same' pressing parameters are applied to each sample (i.e. const heat, const moisture or same apparatus/settings). Since we are limited on samples, I would suggest a press that has steam or moisture (since FFB said most don't technically use steam) applied, since it is likely a good start to find distinguishing numbers vs. cold/dry.

 

I agree that the folding experiment is crude, outdated, and not conclusive. In other words, it is a perfect match for the pressing discussion itself. I deliberately don't use the word "debate" since, without scientific data, it is really a philosophical discussion.

 

My hope is that some of the other initiatives will lead to state-of-the art results conducted at a professional lab. However, these test have been around since before CGC was founded and no one in the hobby has taken the initiative to make that happen yet.

 

My own feeling is that a low tech experiment could be of value in the sense that it might raise the urgency of the matter if, for example, the folding experiment consistently were to show a statistically significant difference between pressed and unpressed books. The numbers I got last night suggest that the technique might be accurate enough to detect that. One danger is that, if the experiment were to show no difference, it would be interpreted to give a false sense of security that pressing does not cause damage. That would be an argument for waiting to get professional data. I am just concerned this could be a chicken and egg problem?

 

Personally, I am not too concerned about destructive testing. There are plenty of cheap books from the 1930s in uncollectible condition, like the ones I used yesterday. That risk has to be weighed against the (small) chance that pressing indeed causes long term damage: in that case we are currently witnessing the destruction of a large fraction of surviving Golden Age comics.

 

I do strongly agree with BR that any such testing should be carefully thought through so that books are not destroyed unnecessarily. If we can agree to a set of rules for the experiments, I would be willing to contribute some time to help with the work, whether it be done mechanically or manually.

 

 

 

 

[bronze_rules]

exhibit A:

 

one stanky smelling donald duck silver age gets immersed in the water (and gets sacrificed in the name of science; sorry, donald). Under one minute, water ph drops from 6.9 to an astounding ~6.2. To put that in perspective, I measured off the shelf balsamic vinigar at 6.5.

 

 

edit: 15 min later it drops all the way to 5.5. So there is a time dynamic that can also be collected and plotted that tells something about the rate of reaction property, in addition to the final settle value of the ph of the sample.

[namisgr]

For the record, the pH of vinegar (which is 5% acetic acid) is about 2.5.

 

Also, for testing the acidity of different comic papers, I'd suggest you soak each paper in water overnight before attempting to measure the pH. Finally, as for the rate of change and what it might reflect: it reflects more on the way the paper has been crunched up and the volume of water being used for the soaking than it does anything about the comparative state of reactivity of the paper.

[bronze_rules]

For the record, the pH of vinegar (which is 5% acetic acid) is about 2.5.

 

Also, for testing the acidity of different comic papers, I'd suggest you soak each paper in water overnight before attempting to measure the pH. Finally, as for the rate of change and what it might reflect: it reflects more on the way the paper has been crunched up and the volume of water being used for the soaking than it does anything about the comparative state of reactivity of the paper.

 

Good clarifications, namisgr. I'm no expert on chemistry and hated the subject, but just starting some crude experiments. Maybe the balsamic viniger test needed time to settle.

I just observed the initial reading. (thumbs u

 

Would you suggest to roll a bottom half of one book (in consistent manner)with same volume of water content for each sample to be a fair and objective setup?

 

Also, do you have a table with typical Ph values (inc. the balsamic vinegar) you could share?

I used to have one of those CRC science constant references, but gave it away. You never have things when you need them.

[tb]

BR: It might be interesting to correlate the pH-value with the result of other experiments, such as the folding I did? If you think it would be worthwhile, I would be glad to mail you samples from the two books I used (I attached pictures above) for comparison.

[slym2none]

exhibit A:

 

one stanky smelling donald duck silver age gets immersed in the water (and gets sacrificed in the name of science; sorry, donald). Under one minute, water ph drops from 6.9 to an astounding ~6.2. To put that in perspective, I measured off the shelf balsamic vinigar at 6.5.

 

edit: 15 min later it drops all the way to 5.5. So there is a time dynamic that can also be collected and plotted that tells something about the rate of reaction property, in addition to the final settle value of the ph of the sample.

 

I am going to have to invest in some micro-chamber paper.....

 

 

 

-slym

[bronze_rules]

BR: It might be interesting to correlate the pH-value with the result of other experiments, such as the folding I did? If you think it would be worthwhile, I would be glad to mail you samples from the two books I used (I attached pictures above) for comparison.

 

Hi tb,

 

That sounds like a good idea, although I think I need to work with the experiment a bit more to make it consistent across samples. So far, the PH has moved back up to 6.0 and settled a few hours later, but also much of the pages have expanded outside of the water. I need to find a jar big enough to contain all the papers so they do not exceed the containing water volume after absorption. Also, I plan to roll half the comic up and insert into the jar for consistency of form according to namisgr's comments. Another thing I'm not certain about is whether leaving the jar open is proper. Next I will likely try your folds test on two halves of some comic, press half with a steam iron, then run the both (ph and fold) tests on each half to look for any significant differences for a start.

 

Also, would like to hear some feedback from FFB and others on the thread.

 

 

Interesting similar experiment on newsprint from 1980, shows similar values for untreated newsprint. 2.017 FE vs. your 2.12 FE measured on the cream sample.

http://www.e-wpp.com/download/pdf/5pal6.pdf (note fig 1. is similar to one of my earlier ideas).

Untreated PH measured at 5.34, compared to my drop to 5.5 (although it currently moved up to 6 again, i need to work on the setup a bit more).

 

---------------------------------------------

more food for thought:

referring to post-washed paper pressing.

 

 

we should see some kind of evidence, either way on the pre/post folds tests.

[ImageMask]

exhibit A:

 

one stanky smelling donald duck silver age gets immersed in the water (and gets sacrificed in the name of science; sorry, donald). Under one minute, water ph drops from 6.9 to an astounding ~6.2. To put that in perspective, I measured off the shelf balsamic vinigar at 6.5.

 

 

edit: 15 min later it drops all the way to 5.5. So there is a time dynamic that can also be collected and plotted that tells something about the rate of reaction property, in addition to the final settle value of the ph of the sample.

 

What instrument are you using to measure these pH values?

[slym2none]

 

 

 

-slym

[bronze_rules]

exhibit A:

 

one stanky smelling donald duck silver age gets immersed in the water (and gets sacrificed in the name of science; sorry, donald). Under one minute, water ph drops from 6.9 to an astounding ~6.2. To put that in perspective, I measured off the shelf balsamic vinigar at 6.5.

 

 

edit: 15 min later it drops all the way to 5.5. So there is a time dynamic that can also be collected and plotted that tells something about the rate of reaction property, in addition to the final settle value of the ph of the sample.

 

What instrument are you using to measure these pH values?

 

The crude and cheap analog ph meter shown in the picture, calibrated to ~7.0 ph for tap water at the moment.

[bronze_rules]

P.S. A very comprehensive battery of tests and analysis of statistical pre and post pressing can be accomplished from this lab for a price est. ranging from a few hundred to 2k max. We can do the preliminary stuff here, but if we find significant differences, maybe that's the next step. You telling me CGC can't afford 2k for preliminary research?

 

http://www.ipst.gatech.edu/testing_services/paper_physical_testing/pricing_policy.html

[tb]

That looks very interesting.

 

Googling "folding endurance tester" gives 2,740 hits, including this one. If the equipment can be bought used for < $1,000, perhaps we could even buy one for the prelim testing? Just another option.

 

 

 

[slym2none]

How far do these tests go? Like for the folding, does the paper have to completely give itself up or can it be that it tears under the lightest of pressure like a wiping action???

 

 

 

 

-slym

[bronze_rules]

tb, Can you explain your folding technique a little more explicitly? When you folded each sheet along the spine back and forth, how exactly did you apply the pressure? Did you just fold 180 degress each way back and forth and place your hand over a table to lightly exert some force along the spine for each fold? Or more forcefully, run your finger along the spine to flatten any air gaps in each fold? The MIT machine says it only folds 135 deg. It seems (intuitively) that if you are using the spine, there is already more tension resistance built in in one direction, so that it is asymmetrical in terms of folding resistance (as it has already been pressed in one direction along the spine). However, that must also be the case for file folders that are subject to the test.

 

I tried with a piece of xerox printing paper, even running my finger along the fold each time and I went past 350 with no break. I can see how you said it was tedious.

Although your FE numbers are very close to professional tested numbers, so I think your approach seems fine.

 

Thanks for your support, buttock.

 

I found an old Nedor funny animal (have no recollection of where I got it from) that I'd feel comfortable experimenting with. I was thinking of trying to bake the individual pages for different durations and then see if the bend-test works at all. If it does, a next step could be to gather a sample of books and send them to FFB for cutting and pressing.

 

I am sure lots of people will be laughing if this doesn't work. All I am saying is this is the best _I_ can think of, and as of yet I have not seen any alternative suggestions. I will be very busy over the next few days so there is plenty of time to contribute ideas before my Nedor gets baked.

 

 

tb,

 

Don't go sticking the Nedor into the Viking oven at home just yet. The aging ovens used to conduct these tests are special pieces of equipment that are temperature and humidity controlled. If we're going to destroy books, let's do it properly, in the name of science!

 

Since a few people expressed interest in this, I decided to sacrifice a few tedious hours on trying the bending experiment that I had read about. This is about as low tech as it gets: I found two very worn Mickey Mouse Magazines from the late 1930s: one with brittle pages and one with cream pages. I picked books with relatively bad pages to avoid the need for an aging oven. I then cut the books in half, removed the staples, and folded each sheet of paper around the spine until it broke. I tried to do the folding as carefully and uniformly as possible, but, obviously, significant variations in the numbers are to be expected. My hope was that the results would fall into a sufficiently narrow range that any significant difference between pressed and unpressed books would show up given a sufficiently high number of samples. None of these samples were pressed: I did this only to get an idea of whether it would make sense to proceed with the cutting-in-half experiment that FFB suggested.

 

The first book was so brittle that only one interior wrap was usable:

Upper half: 79 folds

Lower half: 72 folds

 

For the second book, I tried 5 interior wraps.

Page 1: Upper: 164 Lower: 239

Page 2: Upper: 339 Lower: 297

Page 3: Upper: 315 Lower: 374

Page 4: Upper: 299 Lower: 180

Page 5: Upper: 212 Lower: 271

 

Though these only are preliminary results, my own conclusion is that this brain dead technique, which everyone with enough patience can perform at home with no fancy equipment, just might be able to detect whether pressing a Golden Age book causes measurable damage to the spine. It would take a large number of samples, though.

 

I'd be curious to hear your thoughts, FFB?