This to That | Hacker News Reader

This to That

thistothat.com

512 pointsrobin_reala83 comments

Couldn’t not surface this gem from their primary plastic to plastic recommendation:

> If you live in the U.S. there is a different formula for Goop than in Canada. The Canadian formula contains perchloroethylene which is a known carcinogenic. The U.S. Goop formula contains toluene, which although is a dangerous solvent, it is not carcinogenic. It is however more flammable. Because the Canadian standards are quite rigid regarding flammables and explosive, Eclectic manufactures a formula for Canada, replacing the toluene with perchloroethylene. Apparently, the people who set these standards in Canada are more concerned with the dangers of fires than cancer!

https://www.thistothat.com/glue/hgoop.shtml

For what it’s worth, my experience has proven “Lexel” solid. Strong enough to keep the top window panel on my roof top tent locked down across countless thousands of miles, and it sticks to wet surfaces. Meaning when you’re caught lacking in a PNW rainstorm getting soaked in the middle of the night, you can apply the glue then and there to resolve the issue immediately.

Of course you and your stuff will still be soaked. For that I recommend Mexico.

Perchloroethylene is used by almost every dry cleaner in the United States.

Good thing we don't use toluene in dry cleaning then!

Who still uses a dry cleaner?

I'm not spending time to wash, dry and iron 30 shirts. I'd rather pay $70 at the dry cleaners, once in a while. Now that cancerogenic substance is an interesting topic. I'll have to ask about it the next time I go there.

I'm sure they will be up on the science.

Washing - drying 1 or 30 shirts is only one load. The time to pull out a board and iron 30 shirts is less than the time to load them up drive to a place and repeat when you pick them up.

Some people love ironing. My brother irons while he watches baseball. He says it's Zen for him. I myself hate ironing. I used to stop by his place with my shirts and he'd iron them for me.

The best are those hanging steamer/dryer closets. Fancy stuff though.

Should I do this before or after mowing the lawn, general home maintenance, vacuuming, cleaning, shopping, cooking, showering. Mending work clothes (metal fabrication workshop is hard on work clothes). What am I complaining about, I don't even have children.

Or do you mind if I outsource some of these things occasionally?

No you can't do any of those things because you spent your time/energy at the drycleaner. Outsource only if it saves you time.

It would take me longer to properly iron just one shirt than to drop them all off at the cleaner.

I mean you could pay for a "wet" cleaner. It doesn't sound like your shirts need dry.

I dry clean my suits. I thought that was the only way to clean them... am I missing something?

What are you supposed to do with clothes that say "dry clean only"? Why are there still like 2 dry cleaners per every square city block in NYC? Are these just money laundering operations?

Went to Mexico. Tent is dry but I'm still soaked, now with a combination of rain and sweat. Please advise.

I should have been more specific! For each time PNW isn’t lovely, Baja is.

Reminds me of the infant mattresses and car seats covered in carcinogenic flame retardants.

Lets not pretend that Toluene is without health concerns. It's a volatile organic solvent with health concerns.

Toluene has a NFPA Hazard ID of: Health: 2 Flammability: 3 Reactivity: 0

Perchloroethylene has a NFPA Hazard ID of: HEALTH=3 FIRE=0 REACTIVITY=0

Either way these are nasty substances.

I love simple sites like this - it's a shame that google seems to punish sites like this that contain information that is not often updated (because it does not need to be).

It's weird, right? A wordpress blog that wrapped each combination of materials in a separate five page article and republished them every week with slight changes would do better.

Within engineering, there is this weird dominant culture that fresher information is more likely to be accurate.

Sometimes, that's right, because a lot of tech changes so quickly that a three year old article about is now hopelessly out of date since v5 deprecated v3 syntax.

For a lot of other articles, I'd rather read an authoritative, well-written article that was published 40 years ago. Nothing material has changed in the intervening period, so it doesn't matter.

Google just doesn't seem to know when I value quality over freshness, so it defaults to the latter most of the time.

Sheldon Brown's bike tuning comes to mind after reading your comment. https://www.sheldonbrown.com/articles.html (look at that, an html page!!!)

It's a type of prisoners dilemma with Google. They're having to out game the SEO gamers and pick a tactic that moves the web forward and doesn't over reward entrenched parties. I agree this site gives good useful information, but what if new glues come out on the market? What if a YouTube channel comes out comparing and testing glues? Authority is great until it's out of date, so I think it's logical to go with a bias towards fresh content.

I do wish there was more of a focus on attribution and citing sources on the web. Maybe Web3 will save us on that front...

> I agree this site gives good useful information, but what if new glues come out on the market? What if a YouTube channel comes out comparing and testing glues?

All good questions, but shame that google answer to that is "surely a Wordpress blog called 2022-glue-boss.com made in the last year has all those answers!"

Web3 will save nothing for any of us, but funnel money in the pockets of investors. That’s all it is.

I'm sure Web4 will solve that.

And with ChatGPT, it would be really easy to generate...

This really explains so much. The it was much easier to find random facts on the "old" internet. Now, to get to a random fact, it needs to be wrapped in a life story explaining why they have a deep personal connection with that fact.

It's an ancient website, although it's both a bit too specific and not specific enough.

An example of insufficient specificity: "rubber". Both silicone rubber and SBR are rubbers, but they behave very differently (and almost nothing bonds silicone). Another example: "plastic". Again, this spans both easily-glued materials such as ABS, and hard-to-glue commodity plastics such as HDPE or PP, where almost nothing sticks.

An example of excessive specificity: pretty much everything else. Low surface energy materials aside, gluing ceramic to plastic isn't hugely different from gluing metal to ABS. The real question for most gluing projects is whether the joint needs to be flexible, how much of a gap you need to fill, and how good it needs to look.

Well, I think the target population is the general population. And most people probably don't know what kind of plastic they are using, and wouldn't know that gluing metal to ABS is similar to gluing ceramic to plastic.

I certainly wouldn't have known that.

This reminded me of a recent experience. I ordered a budget, homage Scout Knife, based on the Aitor Gran Capitan "Spanish Army Knife" pattern (produced sans corkscrew of course, which made me chuckle since I'd been a bit of a rebellious scout myself). The knife arrived and the scales promptly began to detach after some vigorous sharpening.

I was surprised to remove the scales and discover they were glued on, smooth plastic to smooth metal. (The factory glue was amber colored, bubbly, and rubbery, with a pizza-cheese effect when pulled cold)

I knew it was a budget tool, but somehow expected at least basic friction-fastening at the metal rivet points. Instead, there was this glue on the scales' inner plastic lattice work, plus some friction pressure around the scale edges.

I mixed up some epoxy, which as I was raised is something of a knife hobbyist's friend. I still have a zero-tang carving knife my dad made by epoxying a Victorinox blade into antler.

Just as I was getting materials ready to sand plastic and metal, I noticed the epoxy mix was already in the advanced stages of setting!

Turns out I had purchased "4 minute set time" epoxy from PC Epoxy. And it had been nearly that long since the mixing. Instead of my usual dilemma of waiting for things to set, the situation was suddenly more urgently coming together.

Usually I prepare surfaces with plenty of time left in the curing, so personal experience had led me astray on account of not typically needing to read the epoxy product label closely. I'd never seen 4-minute epoxy at the store before.

So, I really hope my unsanded but hastily- and liberally-coated surfaces hold together OK for a while.

I think it's a good idea to always do surface prep before mixing epoxy (regardless of the working time).

I find it distasteful that the answer is a brand name instead of an active ingredient. not all brands are available worldwide, and it's better to know what it is made of to check for actual options.

It also makes people think happen by magic instead of letting them learn about what makes the glue work.

Not only are many brands not universally available, even those that are more often than not have different recipes in different regions.

I think calling this “distasteful” is pretty dumb. it’s a website that has offered a free service for decades. If I wanted to glue some glass to metal, I wouldn’t want to which ratio of certain ingredients I’d need. That would just result in me having to check the description of dozens of packages in a shop, while I would forget the names of those ingredients as soon as I’d bought the glue and I have no idea what the effect of a different ratio will be. I’d be just a bunch of useless information, while knowing which product I need to buy is providing a service people actually use.

Too bad that it’s not super useful for people who don’t live in the US, but you can’t make everyone happy.

> I wouldn’t want to which ratio of certain ingredients I’d need. That would just result in me having to check the description of dozens of packages in a shop,

Glues are not a long list of ingredients like a recipe or something. There's only a few ingredients that actually matter in making things stick.

> That would just result in me having to check the description of dozens of packages in a shop,

Checking ingredients is a good thing in a store. Not sure why people find this kind of thing a problem. Aren't you curious with what things are made?

Sometimes I'm curious, sometimes I just want to glue two things together to finish my project. Do you meet many people who mix their own glues?

So wait, originally you are upset that the creator of the site didn't do extra legwork for you. Then immediately after you are chiding someone for not being curious and wanting to know what's inside their glues... Seems like the obvious solution would be for you to take your own advice and look up the product name to find out what's inside. Curious people can and will do that.

We don't have a problem with people checking ingredients. We have a problem with you coming in and telling people that they should check ingredients but also that the author's work is _distasteful_ because they should have provided the ingredients so you don't have to check yourself.

> should have provided the ingredients so you don't have to check yourself.

I provided a rationale as to why it matters: the same glues are not available under the same brand name everywhere, and the brand name is irrelevant. If a brand disappears for some reason, it does not mean that this type of glue is gone.

Couldn't you get the recipe based on the brand name if not available in your region?

There are complete lists of ingredients on everything in my grocery store. But NOT in my hardware store.

A large majority of items don't tell you what's in them at all. Sometimes Googling will tell you the main ingredient in a product, but often it won't or you'll get conflicting answers from random forum posts.

> NOT on my hardware store

Not true - because the regulations of almost every country require to mention the ingredients that compose more than a few percents of the formulation. There is no long list on ingredients on certain glues because they can as simple as 100% ethyl 2-cyanoacrylate.

Well, those regulations sure don't apply in the US then.

Here's the first organic result for "glue" on Amazon, a well-known brand:

https://www.amazon.com/Gorilla-Super-Glue-gram-Clear/dp/B082...

First two photos are the front and back of the package. Care to point out where it says what the ingredients are? Because I sure can't find it.

I could trivially find a hundred other examples. So I don't know where you get the idea that hardware ingredients must be listed on packaging.

It's probably because they wanted to be useful to lay people. Something us hackers often forget. Be useful.

I disagree. If I'm in the middle of a project, I'd rather get a product name and go directly to the store. I don't want to search for multiple products. I can always look for the chemical in the product if needed

None of the glues I have list ingredients. I have no idea what's the "active ingredient".

You just need to read the back of the package.

https://smokewrap.com/pub/media/catalog/product/cache/74c105...

Here for crazyglue: standard cyanoacrylate, like in most "super glues".

Okay, I checked my glues and some of them do list ingredients.

I am pleased that this is about materials rather than microservices.

Much easier than glueing microservices together.

I first used that site in the 90s... amazing it is still there

Maybe you don't mind me asking. I have a small hole in my top-loading washing machine's rubber seal and I have been wondering whether it could be fixed (a new seal costs 100$ and it'd feel nuts to throw it away for environmental reasons alone). I know there are rubber glues, and I have made tests with one, but the question is can they be used inside a washing machine. Does anyone know (a solution to this problem)? The hole is 4cm x 1cm.

Edit: As a side note I have wondered also whether I could use two pieces (of metal, let's say) that I'd screw/press together tightly to prevent the leak. Does that make any sense?

Depending on the size and shape of this seal(gasket), you can go one of two paths; (1) use a gasket compound, it comes out liquidish and then sets into a rubber. (2) buy sheet gasket material similar to your gasket, and trace and cut the gasket out. Depending on size the sheet cost could very between a few dollars to like $10-15.

Hard to know whether a metal gasket would work here, but it's extremely unlikely as you need tight tolerances and high clamping force. Rubber is used here probably because those things are absent, and there is probably movement involved. Other commenter has the right idea, you can probably make a new gasket.

It's hard to say what kind of forces are involved. I could probably open the side panel and be able to see if I run it without water. I was hoping the rubber seal would mostly be flapping along mildly as a whole.

How about a bicycle inner tube patch? Those are generally resilient to the elements.

Or an air mattress patch, or a pool liner patch, or a jetski seat repair patch...

I was thinking of a truck wheel patch, but I'd like to be sure there won't be any stains, and that the glue won't react with the detergent, for example. I don't know if it's possible.

In that case I think you want something that will 'vulcanize' with the material in the gasket, rather than being adhered... I don't know if that's feasible for your exact material but I know there are methods of rubber patching that are more akin to welding than glueing.

Yes, something like that crossed my mind also, but it could be that it's not feasible in small diy work. Have to check.

A little surprised to see that the recommendation for wood to wood isn't the ubiquitous Titebond line of wood glues. I've never seen a bottle of Elmer's carpenter's glue in any shop I've ever set foot in.

There are other adhesives used to glue wood together depending on the application (epoxy is big in the boat world AIUI), but for everyday wood gluing, Titebond is pretty much king in the US.

From a brand popularity and professional use point of view, you are 100% correct. The various Titebond formulations are, I believe, the most widely used by carpenters in the US. However, all PVA glues are going to have extremely similar strength and overall properties, and they do have an excellent summary here:

https://www.thistothat.com/glue/pva.shtml

"Be wary of over priced PVAs that claim to be for a specific use. There is very little difference from one PVA to the other, and nothing that should increase the cost."

Titebond does have range of products, well described here:

https://thewoodwhisperer.com/articles/differences-between-ti...

but Titebond original would be considered to be nearly identical to the Elmers that they recommend

Titebond III creates bonds stronger than the hardest woods I use and they’re even highly water resistant. It’s also easy to apply, clean, store, etc.

Maybe Elmer’s is good but in 15+ years of gluing wood, I’ve never seen it recommended or used, either.

Elmer's got high scores on Project Farm(and their testing methodology is, to a layman's understanding, reasonably structured). But their actual conclusion in that video was that if you were testing most of these glues at their limit, the wood would fail first.

I love project farm. That’s good to know, I’ll have to look that one up.

This makes sense too, considering the type of glue will be chemically very similar across brands. If the bonds are stronger than wood, then the only qualities worth comparing would be unrelated to strength.

I suppose set time, ease of storage, ease of application, etc will remain almost identical across brands as well. But maybe not?

I’ll have to watch that project farm. He’s a knowledgeable guy so I imagine I’ll come away from it a little better educated.

I use Elmer's normal white glue all the time for wood. It has a MUCH longer open time (I am not fast), and is still as strong or stronger than the wood.

My home-made bench made of southern yellow pine boards ripped and laminated is exclusively that and will outlive me, and my son if he decides to not throw it out.

NB: I also use Titebond, but not because it's better at holding, just cures/dries quicker. And is better for wet/damp applications.

I couldn't find the specific combo I spent quite some time researching last year so I tried ChatGPT. Within seconds it recommended the exact solution I ended up using based on my searches back then.

"What adhesive would work well for affixing a mylar sheet to foam board?"

> A good adhesive for affixing a mylar sheet to foam board would be a spray adhesive, such as 3M Super 77. This type of adhesive can provide a strong bond between the mylar and foam board, and it dries clear, so it won't be visible once the mylar is attached. You could also use a glue stick, but it may not provide as strong of a bond. It's important to use an adhesive that is compatible with both the mylar and the foam board, so be sure to read the labels carefully before making a purchase.

Interesting how the very first sentence went for a brand name. Naively I thought it would at least start with compounds, even if give a few examples of products in the end.

An absolutely brilliant idea. I frequently use the wrong glue. A little more info about what qualities each choice has would be interesting.

Remembering to let glues which "cure" have stress free time, and glues which are "apply to each surface, wait, and then contact press" dog my attempts too.

I've been using superglue (cyanoacrylate) to attach a small metal L-bracket to some IP65-rated ABS plastic boxes, then hanging them on a nail or screw using the metal bracket. The boxes contain battery-powered temperature sensors. This has worked great for several years for the attic and crawlspace. But the outside boxes fall off every winter. I think the metal and plastic expand and contract at different rates through the temperature swings (40C/104F summer, -10C/14F winter) causing the bond to break. Amusingly, superglue is NOT one of the recommendations given on this website (LePage's Metal Epoxy, J-B Weld, Faststeel Epoxy Putty) for Metal to Plastic.

It's great design, straight to the point, but I have 2 big issues with it:

- It's VERY US centric: I can't get even half of the recommended products where I live.

- I'd love to see recommendations for more specific materials, e.g. different types of plastic: it's not the same trying to stick 2 pieces of acrylic (PMMA / Plexiglass, you'd just use a solvent like Chloroform), or ABS (many common adhesives work fine), but gluing PE or PP is nearly impossible for all practical purposes.

Discussion from a few years ago: https://news.ycombinator.com/item?id=21110712

My dishwasher tub is made of a kind of plastic that’s very good for not absorbing foods/liquids but therefore is very bad at having anything adhere to it.

This is a problem as a tiny hole was burned through it. I’ve had a hard time finding something that plugs the hole. Various epoxies. Silicone caulk. JB weld for plastics. Everything peels off in weeks or months.

You can use mechanical fasteners (rivets) and a rubber gasket possibly?

Or, make the hole uniform to accept a plug (again, mechanical attachment)

And if you want to know whether the recommended glue is any good, watch a video from Project Farm https://www.youtube.com/@ProjectFarm/search?query=glue

I like his videos but adhesives are tough to test quickly. Strength over time is important for a lot of applications. It's also really strange that he used end grain glue joints when testing wood glues. No woodworker would use wood glue for that application.

Hug of death from HN?

Thank you! One of the items on my todo list today was to find a glue suitable to attach plastic to glass. I’ve tried 4 different glues and non have held. Here’s hoping!

Try double-sided tape (there's many different kinds). Unfortunately, the site only recommends glues.

Bookmarked forever, thank you!

One of the original microsites, dating to 1999!

I can't believe that a website that is practically static goes down like this.

cgi-bin : What year is it?!

This is a so pre 2000 behaving website. They could turned the glue recommendation engine to a SaaS so you can pay mothly for glue related advices. And they have no heavy JS front-end, I mean it's 2022, right?

Reply on news.ycombinator.com