By continuing to use this site, you agree to our use of cookies. Find out more
Forum sponsored by:
Forum sponsored by Forum House Ad Zone

Metric vs Imperial - Practical or Traditional?

All Topics | Latest Posts

Search for:  in Thread Title in  
JasonB25/10/2010 20:02:34
avatar
Moderator
22740 forum posts
2653 photos
1 articles
Your mic may have a resolution of 0.001mm but whats the accuracysome with the same resolution can be + or - 0.004mm so really you are only reliably reading to 0.01mm
 
Mine will read to 0.001 with 0.001mm accuracy but I do tend to prefer using my old Mitutoyo to the digital one.
 
Jason

Edited By JasonB on 25/10/2010 20:04:15

Mark Dickinson25/10/2010 20:26:23
48 forum posts
4 photos
I left full time education in the early 90's having been taught to use the metric system. Then I entered the real world...

I measure length in millimeters but distances in miles.
I used to buy fuel in gallons but now buy it by the litre, I measure fuel economy in mpg.
Fuel price used to go up a penny a gallon, now it's a penny a litre.
I buy drink by the pint.
My cars engine capacity is calculated to the nearest tenth of a litre. Its wheel size is 16" but its tyre width is 235mm.
My Land Rover uses metric, unf, unc, and whitworth threads.
I can visualise a gradient of 1:6, but a road sign saying 17% means nothing to me.

I work in IT where a byte could be 8 bits, but then again it could be more. A kilobyte to me is 1024 bytes, but should be 1000 according to the SI standard. My 500 gigabyte hard drive should actually be a 500 gibibyte hard drive. So that will be 500 gib then (is that pronounced gib or jib)

I'm sitting here in front of my 22" monitor typing this

There are only 10 types of people in the world — those who understand binary, and those who don't

Since taking up "model engineering" I'm having to deal with imperial fractions, imperial decimals, number and letter drill sizes.
 
People wonder about my vaguely confused expression......

Stub Mandrel25/10/2010 20:33:58
avatar
4315 forum posts
291 photos
1 articles
One thing people seem to ignore is that fractions make designing things very easy. Using squared or lined paper I find it very easy to design things using stock sizes, much easier than working in mm in which I end up needing non-preferred sizes.
 
Neil
John Olsen25/10/2010 21:45:09
1249 forum posts
94 photos
1 articles
Mark...A byte should always be 8 bits. A word will be a different length depending on the actual computer system in use, and need not be a  power of two although they usually are.
 
Neil, I don't know that I completely agree with you on that....since the material I can get tends to be a mixture of Imperial and metric sizes. The machines I own are also a mixture of Imperial and metric.  Luckily the CAD system I am using will accept input like :
3/16
3+13/32
17/32+9/64
5/25.4
25.5 * 3/16
 
It does the arithmetic for you. 
You can also draw in one unit and then change the whole drawing to the other, although this is not a very practical way of converting a design. Well, actually I guess it depends on the nature of the conversion you need...if you are working on an Imperial design, using Imperial stock, but machining on a metric machine it could be useful. (Or vice versa I guess)
 
regards
John
Wolfie25/10/2010 22:35:18
avatar
Moderator
502 forum posts
"There are only 10 types of people in the world — those who understand binary, and those who don't"
 
Ah yes. I'm an IT man too. That makes 10 of us
Mark Dickinson25/10/2010 23:13:04
48 forum posts
4 photos
Posted by John Olsen on 25/10/2010 21:45:09:
Mark...A byte should always be 8 bits. 
 
Pedant mode.
An octet will always be 8 bits. A byte's is the smallest amount of memory that a CPU can address individually and in the past computer systems could have used more or less than 8 bits per bytes.
Nowadays it is usually assumed that a byte is the same as an octet, but I was always taught never assume anything.
 
anyway its all geek to me 
John Olsen25/10/2010 23:41:42
1249 forum posts
94 photos
1 articles
If you Google for it you will find that the vast majority of sources define a byte as 8 bits.
 
eg
Eight bits; also called an octet.
 
Your definition of a byte is what I would call a word...eg the first microprocessor had a word length of 4, later ones had 8 and I have worked with machines with word lengths of 10, 16, 18, and 32 bits.  Your definition would also render the quoted sizes of hard drives and files completely meaningless, since different computers can have different minimum word lengths.
 
 regards
John
Mark Dickinson26/10/2010 00:53:00
48 forum posts
4 photos
John,
I only mentioned bytes to highlight the ambiguities of measurement systems, I wish I hadn't bothered.
However

From the Oxford English Dictionary

byte

Pronunciation:/bʌɪt/
noun
Computing
a group of binary digits or bits (usually eight) operated on as a unit
a byte considered as a unit of memory size

another definition 
http://www.wordiq.com/definition/Integer_(computer_science)#Bytes_and_octets

I didn't say it wasn't 8 bits.


Non 8 bit byte architectures include CDC 6000 mainframes, DEC PDP-10 mainframes(yes obsolete now)

Hard drive sizes, would that be using manufacturers stated sizes where 1GB is 1000^3 or Microsoft's reported sizes where 1GB is 1024^3
Sam Stones26/10/2010 04:13:03
avatar
875 forum posts
326 photos

This thread about standards, looks like it will go on forever.

So here’s another six penn’orth of stuff to add to the mix.
This time, it's about thread standards, Metric and/or Imperial. Not the normal run-of-the-mill types though. Around the early 60's, while working from a technical service laboratory, I noticed that there was a regular stream of requests from designers and manufacturers addressing the following problem.
 
Caps and closures, moulded in recently formulated polyolefins were working loose in service. The problem was fairly clear, but while trying to get the message across, I acquired a booklet which described thread profiles especially suitable for glass `finishes’, and their counterpart - (rolled) metal closures. And here's my point:- 

There were about 125 different thread profiles!

These profiles had little to do with pitch or diameter, so you can guess how many permutations there could be. Fortunately, I’m happy to report, after spreading the word over a considerable period, buttress threads are now almost the standard.

If you want more on this latter topic, please - start another thread. Pun intended.

Sam

Edited By Sam Stones on 26/10/2010 04:14:32

Andrew Johnston26/10/2010 09:32:14
avatar
6601 forum posts
701 photos
The only axiomatic truth about standards is that they are non-standard.
 
Regards,
 
Andrew
Stub Mandrel26/10/2010 20:53:51
avatar
4315 forum posts
291 photos
1 articles
But John, are your words big or little endian? (that's 11 of us Wolfie)
 
Neil
John Olsen26/10/2010 22:28:02
1249 forum posts
94 photos
1 articles
I thought the big endian thing only applied to numbers? I leave the programming to my sons these days. I liked programming in assembler on the old 68000 series, I don't want to get into x86 in any shape or form. One thing about the byte though. I doubt if the meaning is legally defined anywhere, unlike say the metre or the foot. In English, that means that the only authority for the meaning of the word is usage.
 
The same probably applies to the "kilo = 1024" convention, which would mean that the hard drive manufacturers are strictly in the right when they quote in thousands  and millions...however, why a marketing person would want to leave so many customers feeling they have been short changed is a mystery to me.
 
The other thing about standards is the the old ones never really go away. So although here in NZ we metricated many years ago, you can still buy Imperial sizes, and indeed industry would grind to a halt if you could not. Actually lately it seems to be doing so anyway.
 
regards
John
 
 
Andrew Johnston26/10/2010 23:31:30
avatar
6601 forum posts
701 photos
Ah, but words here means numbers. Now is that English or binary?
 
For those unencumbered by the minutiae of microprocessor design I'll explain the in-joke. When microprocessors came out that used 16 or 32 bits internally instead of 8, the problem arose of how to transfer those bits to a peripheral, say a UART, that only accepted data in blocks of 8 bits. The data can be transferred most significant bits first, known as big endian, or least significant bits first, known as little endian, in a salute to Swift.
 
Motorola processors, eg. the 68000 and 68020, used big endian, while the Intel x86 processors were little endian. If IBM had gone with the 68000 for their PC the world might have been a very different place.
 
Regards,
 
Andrew
Sam Stones27/10/2010 00:03:16
avatar
875 forum posts
326 photos
There seems to be no endian in sight!
 
Or should that be insight?
 
Sam

Edited By Sam Stones on 27/10/2010 00:04:32

Andrew Johnston27/10/2010 00:17:38
avatar
6601 forum posts
701 photos
Some microprocessors felt the same; they were bi-endian.
 
Regards,
 
Andrew
Bill Pudney27/10/2010 04:39:11
611 forum posts
24 photos
All the talk about measuring things in fractions of a horsehair made me think of when I had to work for a living....only a couple of years ago.  I was sort of responsible for buying machinery for a small machine shop.  One of the ongoing jobs had...and required very tight tolerances.  Generally tolerances were only tight where required, but there were lots of inter-related features with tight tolerances, in the order of +/- 0.01mm for instance. It could end up having a tolerance of "0" in fact if all the tolerance range was used up earlier in the process.  The various suppliers of co-ordinate measuring machines (CMM) were contacted with details of the spec required.  They were all really interested to start with but gradually said "...its too much for us".
 We ended up having to make a laboratory standard enclosure, with its own air conditioning system controlled to +/- 1 degree Centigrade 24 hours a day, each CMM mounted on its own concrete base shock/vibration insulated from the machine shop floor.  The CMMs (we ended up with 3) were CNC and took about 3 hours to fully inspect about 90% of the features they also generated a report which we had to keep.  The items to be inspected were left in the CMM room for 24 hours to thermally adjust before inspecting, and were handled as little as possible....those hot handed inspectors!!
The first items that we made took about 50 hours to machine and another 70 or 80 to inspect....it was very very fiddly!!
When I retired the machining time was about 7 hours and inspection time was about 3.5 hours total.

Edited By Bill Pudney on 27/10/2010 04:40:00

Wolfie27/10/2010 13:23:11
avatar
Moderator
502 forum posts
"that's 11 of us Wolfie"
 
Ahh 1 more and it'll be 100
DMB27/10/2010 20:54:32
1312 forum posts
1 photos
Hi all,
This is a response to Mark Dickinsons post of 25th; "In the 1960`s we took `acid` to make the world seem weird now the world is weird and we take Prozac to make it `normal`!"
V8Eng27/10/2010 22:16:09
1701 forum posts
1 photos
I notice that UK based TV programmes seem to be trying to metricate us by giving distances in Kilometers. 
Andrew Johnston27/10/2010 22:44:42
avatar
6601 forum posts
701 photos
I would hope that they would at least have the decency to brainwash us with kilometres.
 
Regards,
 
Andrew

All Topics | Latest Posts

Please login to post a reply.

Magazine Locator

Want the latest issue of Model Engineer or Model Engineers' Workshop? Use our magazine locator links to find your nearest stockist!

Find Model Engineer & Model Engineers' Workshop

Latest Forum Posts
Support Our Partners
Dreweatts
cowells
Rapid RC
Eccentric Engineering
Eccentric July 5 2018
Subscription Offer

Latest "For Sale" Ads
Latest "Wanted" Ads
Get In Touch!

Do you want to contact the Model Engineer and Model Engineers' Workshop team?

You can contact us by phone, mail or email about the magazines including becoming a contributor, submitting reader's letters or making queries about articles. You can also get in touch about this website, advertising or other general issues.

Click THIS LINK for full contact details.

For subscription issues please see THIS LINK.

Digital Back Issues

Social Media online

'Like' us on Facebook
Follow us on Facebook

Follow us on Twitter
 Twitter Logo

Pin us on Pinterest