Science desk
< August 8	<< Jul \| August \| Sep >>	August 10 >

Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

August 9[edit]

timing belt idler wheels[edit]

I have a bunch of timing belt idler wheels that I want to mount onto a plate[1]. The ball bearings inside these have a inner diameter of 5mm. What's the "proper" way of mounting them?

If I use a screw to mount them (like the left side of [2]), then the screw will improperly contact the side of the wheel, and the wheel will grind against the plate.

I tried putting a 5mm diameter dowel pin inside the wheel and mounting the dowel pin instead (right half of [3]). But the problem is that dowel pins are always oversized, so the mere act of the putting in the dowel pin has managed to destroy the bearings. Mũeller (talk) 05:15, 9 August 2017 (UTC)[reply]

I think the idea of the screw, or a bolt, is the right way to go. You need two flat washers whose outer diameter is no greater than the outer diameter of the inner race. Place one washer between the idler and the plate, and the other under the head of the bolt. When you tighten up the bolt, the inner race will be placed under compression and will not turn, leaving only the outer race to turn around it. You need good quality washers to avoid uneven compression and deforming of the inner race. Akld guy (talk) 06:20, 9 August 2017 (UTC)[reply]

Thanks. I'm having a hard time finding the right washers. ID is 5mm, and the outer diameter of the inner race is only 6.3mm. The smallest M5 washer I can find is 8mm. Mũeller (talk) 07:49, 9 August 2017 (UTC)[reply]

Here's the cheapest nastiest solution. http://www.unitedfasteners.com.au/media/images/product/product_tmp/large/UN0104_2861351072354.JPG I suspect you will encounter problems with the the bolt working loose in the plate, really you should be supporting both ends of the bolt. You may get away with it depending on the usage profile. Greglocock (talk) 08:14, 9 August 2017 (UTC)[reply]

If I bolt it on without using washers as spacers then the wheel won't turn at all. Mũeller (talk) 09:27, 9 August 2017 (UTC)[reply]

The proper way would be an Axle with grooves for retaining rings at the right positions. If you want to fixate your belt wheel precisely you fix one ball bearing form both without any space for it to move. The other beraring will then only be secured with enough free space so it can run free but not fall off. If precise fixation is not that important you use an Axle with 2 groves and 2 rings that fixates both bearings with enough free space to run properly. You can manage something identical with a screw (put in from the opposite side of your plate, a fitting washer on the side of your bearing to get away from the plate and two nuts where one fixates you first bearing (you will have to get one of them out of your wheel and after reassembling the last nut should be a selfsecuring version at the end of the wheel so it has a tiny space to run free. --Kharon (talk) 10:26, 9 August 2017 (UTC)[reply]

(edit conflict) Those dimensions suggest the use of a ring instead of a machined washer. Try a shop that repairs small electric motors to see whether they have rings of the suitable size. Another approach would be to get a shop to machine a circular groove around the hole in the plate to provide clearance for the outer race, leaving just a small ring around the hole. It's a precision job, needs to be cut only a mm or two deep, and is probably the best solution because it removes the need for a washer or ring which introduce tolerance uncertainties. Akld guy (talk) 10:31, 9 August 2017 (UTC)[reply]

Hang on, your photo had 'standoff' inners on the ball bearings' or am I imagining things (I'm not i just checked)? If not, then yes, you need spacers. Vague badly defined questions get vague answers. Sad but true. Kharon is talking nonsense. Greglocock (talk) 10:35, 9 August 2017 (UTC)[reply]

@Greglocock Well i assume you are a professional (from your page discription). Guess here where i live (Germany) we do construction differently or i wrote it down to complicated. Have a look at the german wiki [4] a glimps on some drawings if you like. Couldnt find any english wiki articles. Because of the many differend bearings, applications and requirements for all possible constructions for all possible tasks the professional approach for bearings ofcourse is much more complicated and sophisticated. --Kharon (talk) 10:57, 9 August 2017 (UTC)[reply]

Yes I am a mechanical engineer. Your post is nonsense in English. Don't blub, i couldn't write a whole sentence in German, never mind using the right technical terminology. Greglocock (talk) 11:14, 9 August 2017 (UTC)[reply]

@Greglocock Nope, unfortunately no standoff or spacers came with the wheels. It's just two standard ball bearings press-fitted inside. The inner race of the bearing stick out maybe 0.2 mm compared to the outer race of the bearing (and the wheel itself) but that's about it.

My question is basically asking where to buy the right spacers. Washers won't work because their OD are too large. I can't find anything COTS that has a ID of 5mm and an OD that's less than 6.3mm. Mũeller (talk) 11:28, 9 August 2017 (UTC)[reply]

Will you consider rubber O-rings in compression? Blooteuth (talk) 13:13, 9 August 2017 (UTC)[reply]

Thanks! I'll give it a try. Mũeller (talk) 13:22, 9 August 2017 (UTC)[reply]

These bad boys may be appropriate http://www.tracgear.com/product/rc4wd/custom/conWasher/dd.jpg or you could try and find the right size of these https://dnwgit7zg7mqr.cloudfront.net/images/141020-UK/800/30475.jpg Greglocock (talk) 16:09, 9 August 2017 (UTC)[reply]

Thanks! Conical washers are exactly what I needed. Much obliged. Mũeller (talk) 02:37, 10 August 2017 (UTC)[reply]

Blindness to important information because of deluge of irrelevant emails[edit]

Is there any research suggesting that someone whose received emails are <90% irrelevant to them are liable to start missing information? It happens to me and I'd like to make the suggestion that my college start providing more mailing lists so that people that are meant to be receiving the e-mails do and I and everyone else don't. I imagine that the cost of wasted person-hours is considerable. --129.215.47.59 (talk) 11:22, 9 August 2017 (UTC)[reply]

I would imagine so. If you wanted to start your research in this field, the place to start is Information overload, which is the exact concept you describe, applied to any source of information, not just email. It is a well trodden field of psychology. That article has an entire section on information overload through email. --Jayron 32 11:29, 9 August 2017 (UTC)[reply]

Agreed, it's a real problem. I don't see it as information overload though, because there's no information. It's more of a technical problem (how do I not even see the rubbish), rather than the psychological problem of trying to study a huge table of numbers and extract relevance. The trouble is that it's an arms race - as fast as trimming waste becomes more practical, so does generating more subtle waste that's harder to identify. Andy Dingley (talk) 11:37, 9 August 2017 (UTC)[reply]

If that is the case, then what you're going to have to do is delete the entire section from the Wikipedia article I cited above that specifically discusses how the concept works with email, because according to you the Wikipedia article is wrong and misleading. --Jayron 32 11:58, 9 August 2017 (UTC)[reply]

(EC) This is from a survey by a marketing company so isn't the best quality research and doesn't directly relate to missing information but does suggest when it comes to marketing emails (which I presume aren't what you're referring to), receiving too many or too many irrelevant emails is a common reason for unsubscribing, actually the most common reason of those surveyed. If you look at their report, [5] it's actually also barely the most common reason for marking emails as spam. Nil Einne (talk) 11:39, 9 August 2017 (UTC)[reply]

I'm not referring to marketing emails but e-mails talking about people retiring or meetings that don't even involve anyone in my gosh-darn building or someone selling their ***** sofa (not meant to happen but still does) and so on. There are loads of e-mails intended for specific groups of people but there's no mailing list for them so they just go ahead and e-mail the whole college without even bothering to specify whom they're trying to reach so you're left struggling to determine if and how this e-mail has any relevance to you. — Preceding unsigned comment added by 129.215.47.59 (talk) 14:03, 9 August 2017 (UTC)[reply]

As a practical method I suggest having at least 2 email addresses, one you only give out to trusted parties who won't SPAM you, and another you do give to them, such as when they require an email to join a web site. Read every email carefully under the trusted email address, but just skim the titles under the non-trusted address. To get there from where you are, you might make your current email address the non-trusted one, and send out a new address to those you do trust not to SPAM you. (I'd actually prefer to have a separate email address for each person or business I communicate with, so I can tell who is selling my name, but then I'd need an email system that can collectively view some combo of dozens of email addresses at once, rather than having to log into each individually.) StuRat (talk) 11:55, 9 August 2017 (UTC)[reply]

The questioner specifically mentions college. I have six college email accounts. All of them collect more junk from the college itself than important content. Only one is nice enough to send out a single daily "This is your junk for the day" email. (It is technically called the Campus eConnection.) What I do is use Thunderbird to read all email in one program. That allows me to use Thunderbird's junk email filters to get rid of a lot of the junk. It also allows me to move email from a college account to my personal account if I want to keep track of it better. 209.149.113.5 (talk) 12:13, 9 August 2017 (UTC)[reply]

I wrote a reply here, but decided to move it to Wikipedia:Reference desk/Computing#E-mail deluge instead, as it is more technical, as opposed to the OP's question, which was about the psychological effects of the deluge. — Sebastian 12:43, 9 August 2017 (UTC)[reply]

Antibiotic resistance by production of beta-lactamase[edit]

I read a bit about antibiotic resistance and beta-lactamase and I wonder how much antibiotic could be inactivated by bacteria which produce beta-lactam. I guess the rates vary a lot between different bacteria strains and antibiotics and they will probably depend on thousands of conditions. I would be happy with examples just to get an idea of the order of magnitude. --134.76.38.158 (talk) 18:36, 9 August 2017 (UTC)[reply]

It varies enormously. In some bacteria the minimal inhibitory concentrations (MIC) of antibiotics are only modestly increased while other bacteria may become absolutely resistant at least for any concentration that is safe for humans. As an example you can look at Neisseria gonorrhoeae which in 1940s had MIC for penicillin less than 0.05 μg/mL but in 1970s acquired a couple of plasmids and become virtually untreatable with penicillin. Ruslik_Zero 20:55, 9 August 2017 (UTC)[reply]

Which Gazelle species is this article referring to?[edit]

This article talks about the "Acacia gazelle", but I can't figure out which species it's referring to - there are lots of gazelles, and none of the ones I saw seemed to match the range and population the article describes. Ariel. (talk) 20:58, 9 August 2017 (UTC)[reply]

It is referring to Gazella arabica acacia, a subspecies of the Arabian gazelle (Gazella arabica) found in Israel. Some theories say that it is going through a speciation process and will eventually become a distinct species. You can read further about it in the article "Wild Gazelles of the Southern Levant: Genetic Profiling Defines New Conservation Priorities". --SuperJew (talk) 22:21, 9 August 2017 (UTC)[reply]