Powered by Invision Power Board


Forum Rules Forum Rules (Please read before posting)
  Reply to this topicStart new topicStart Poll

> Mesurement Fubars
Jimthecopierwrench
Posted: January 02, 2012 03:45 am
Reply to this postQuote Post


Moderator
Group Icon

Group: Moderators
Posts: 10,139
Member No.: 561
Joined: January 10, 2004




Man, gone through a 4 x 8' sheet of aluminum this weekend trying to build an experimental waveguide.

As I don't have any trustworthy finished edges, and my framing square isn't 'good enough' i've been doing the following:

An arbitrary line.
With a compass set to an arbitrary distance, scribing a circle at a point on the line.
Scribing a second circle from the point where the firsts' circumference meets the line.
Scribing a straight line across the jesus fish created by the two circles, which should be 90 degrees from the first.
All other measurements are then taken from this cross point.

The pieces I'm trying to make are not huge - couple of square feet. Just finished scribing and cutting the third version, and yup - two 'identical' pieces are 2mm different.

The compass is sharp, I'm scribing with an x-acto blade, and my primary straight edge is true - that is the two I have appear to meet evenly on edge in all possible (4) ways that the edges can meet - by my reasoning they must be true or one configuration or another would show up.

And yeah, have my glasses on and I'm not liquored laugh.gif

Rechecking my scribe line intersections show them to all be dead eyed - all coming to a single point as far as can be seen by the human eye.

Failure to sacrifice a virgin to some obscure ruler god? Just can't figure why every attempt has gone south so far. Meh. Taking a break, will crack a sud (sometimes you have to walk away before a temper tantrum slash sloppy frustrated work ensues) and see what's waht here for a few minutes.


--------------------
All children are equal in the eyes of dog :)
PMEmail PosterUsers Website
Top
telomere
Posted: January 02, 2012 04:04 am
Reply to this postQuote Post


Forum Addict ++
*******

Group: Spamminator Taskforce
Posts: 4,300
Member No.: 9,502
Joined: May 18, 2007




Is your compass "slipping" and getting wider as you measure?


--------------------
"Sometimes I can't tell the difference between thinking and being quiet, but I'll try."
PMEmail Poster
Top
johansen
Posted: January 02, 2012 04:14 am
Reply to this postQuote Post


Forum Addict ++
*******

Group: Trusted Members
Posts: 3,114
Member No.: 10,912
Joined: September 06, 2007




which side of the line did you cut on, and with what?


--------------------
Can the Ethics Committee have a closed door meeting without imploding from cognitive dissonance?
PMEmail PosterUsers Website
Top
Jimthecopierwrench
Posted: January 02, 2012 05:30 am
Reply to this postQuote Post


Moderator
Group Icon

Group: Moderators
Posts: 10,139
Member No.: 561
Joined: January 10, 2004




Naw, compass is nice and tight - I go back and (lightly) rescribe to make sure the last is the same as the first. Not a kerf issue either - just using some thin siding material, couple of gentle bends and it breaks right on the line scribed with a knife. Marks are oh, maybe .1mm wide or so. Swear, just the way it is some days.


--------------------
All children are equal in the eyes of dog :)
PMEmail PosterUsers Website
Top
Geek
Posted: January 02, 2012 06:20 am
Reply to this postQuote Post


Moderator
Group Icon

Group: Moderators
Posts: 9,943
Member No.: 62
Joined: July 23, 2002




It's that damn 4th-dimensional property dry.gif

Seriously... I had several pieces cut by a shear at a local machine shop.... shear is true, right? I mean they build truck bodies all day and nothing's out.

I cut the piece as per clients order for chassis plate straight as per square (confirmed square at Home Dept) and it **always** ends up 3mm trapezoidal! wacko.gif

Cheers!


--------------------
-= Gregg =-
"Ratings are for transistors.....tubes have guidelines"
(please do not PM me for advice. Non-forum business messages will be ignored)
PMUsers Website
Top
Sch3mat1c
Posted: January 02, 2012 07:45 am
Reply to this postQuote Post


Forum Addict ++
Group Icon

Group: Moderators
Posts: 19,437
Member No.: 73
Joined: July 24, 2002




Meh, the shear at work is questionable -- it's the kind with a hand-adjustable backstop, with a scale and drive knob on each side of the backstop to set it. I believe the scales are true to the cutting edge; however, on the front, the right side margin that you push the metal in alongside isn't quite square to the backstop (although it could be the metal I was cutting was off by as much).

This is equivalent to those paper cutting "WHOMP" bar type shears, if the scale along the back edge were not perpendicular to the blade edge.

Jim: if you want to scribe by basic geometry, you want to minimize error. Most likely, you'll have to project the lines quite far from a mere compass circle to get to the edges of the sheet. This magnifies any small error (eyeballing a scribe mark, a few to maybe 10 thou at a 3" radius?) to a big offset at the end (fractional inch?). Make yourself a compass out of a metal strip with holes along it, placing the pointy bits in whichever holes you need for the length. Or if you're fancy, use a clamp-on pointy for adjustable length. Make the circles as big as you can make them, full sheet if possible. Now your eyeballing error goes from a few tenths or tens out of a 3" compass, down to a couple tens out of a four foot sheet.

As waveguide goes, aluminum probably isn't the best idea, as you'll probably have to bolt the seam pretty tightly. Or weld it (with no more than 1/10th wavelength long gaps between beads).

Or bend it so the seam ends up centered along the axis, parallel to the E-field of whichever mode you're building (this is usually right down the middle of the wider side, IIRC). Reason being, this axis is supposed to have no current flow, so it doesn't mind if you slot it axially.

Slotted waveguide is interesting because it lets you get inside -- at school, there's a section of ~10GHz waveguide (whichever band, it's about 0.5 x 1" inside dimension) with a slot along the top, with a 1/4 wave antenna poking in, attached to a slide with scale. The purpose: hook up a wattmeter to the antenna, then slide it along the gap, measuring the distance between peaks (= wavelength/2), the peak amplitude and the valley amplitude. The ratio is SWR, literally, standing wave ratio. If you operate it with a metal plate bolted to the opposite end (full reflected power), you get well defined peaks with hardly any power in the valleys (SWR >> 10); if you add a terminated load, you can hardly see anything (in practice, < 1.5 SWR is acceptable for circuitry, so with SWR < 1.1, the match is apparently quite excellent).

All the microwave hardware at school is brass or copper, sometimes silver plated. Standard flange is face mounted with four bolts in the corners, like you'd see a bolt-together pipe flange.

You might be interested to know cylinders carry waveguide modes just as well. Works just like coax, without the co-. A length of appropriately sized aluminum tube or extrusion, or better yet, copper, may provide you with better results than (literally) rolling your own.

Tim


--------------------
Answering questions is a tricky subject to practice. Not due to the difficulty of formulating or locating answers, but due to the human inability of asking the right questions; a skill that, were one to possess, would put them in the "answering" category.
PMEmail PosterUsers Website
Top
johansen
Posted: January 02, 2012 11:30 am
Reply to this postQuote Post


Forum Addict ++
*******

Group: Trusted Members
Posts: 3,114
Member No.: 10,912
Joined: September 06, 2007




the stuff i used to work on was 17 ghz and tig welded aluminum. none of it was silver plated, though i think the flexable couplings were silver plated.

4 by 8 foot sheet.. what band are you going for?


--------------------
Can the Ethics Committee have a closed door meeting without imploding from cognitive dissonance?
PMEmail PosterUsers Website
Top
CWB
Posted: January 02, 2012 02:30 pm
Reply to this postQuote Post


Forum Addict ++
*******

Group: Spamminator Taskforce
Posts: 21,225
Member No.: 15,154
Joined: May 15, 2008




did you try cross-measurement to ensure the dimensions ?
once you get two truly "squared" pieces (assuming a rectangular waveguide) use them as a pattern for marking out the remaining pieces .

@ tim :
slotted waveguides , etc ...
i was working with another engineer at a transmitter site ... he pointed out a sampling port in the 4" hardline and told me this story :
normally there was a probe bolted into the hole/flange but it was removed for repair (the replacement was sitting in a box) .
a dj from the station had come along to help out with some grunt work .
a new final had been installed , burnt in and was being ramped up .
the engineer was looking at the metering ... the transmitter settled down nicely and was running about 30KW up the pipe .
there was a ker-bang , a scream , the lights dimmed a bit and the transmitter shut itself down .

the engineer walked around the end of the transmitter and seen the dj/grunt boy laying in a pile on the floor .
he noticed that the end of his index finger was missing .
after getting the guy to "come around" and some first aid , he hauled him down off the mountain to meet the ambulance .
he went back up and dug the end of the dj's finger out of the port .




--------------------
"Know how to solve every problem that has been solved"
R. Feynman '88
PM
Top
AwesomeMatt
Posted: January 02, 2012 03:13 pm
Reply to this postQuote Post


Forum Addict ++
*******

Group: Trusted Members
Posts: 3,000
Member No.: 9,878
Joined: June 21, 2007




QUOTE
he went back up and dug the end of the dj's finger out of the port .


At least it was his finger he tried that with.

Jim: Whassis for?
PMEmail Poster
Top
Jimthecopierwrench
Posted: January 02, 2012 06:46 pm
Reply to this postQuote Post


Moderator
Group Icon

Group: Moderators
Posts: 10,139
Member No.: 561
Joined: January 10, 2004




Experimenting with directional 'cantenna' type stuff, making my own guides and horns - probably useless anyway given that the math might as well Greek ... Oh, wait. it is laugh.gif . Might as well call it dicking around without the equipment to even measure actual gain but hey, it's the holidays and I wanted to lighten up and play around a bit.

Actually I know that one of the rigs isn't even correctly polarized to the active elemnet but I'm the kind of clown who tends to learn more by trying to carry the cat home by it's tail (Mark Twian reference - which I now play in my mind with the voice of whoever played Clements on that two prt TNG episode)

QUOTE
did you try cross-measurement to ensure the dimensions ?
Yeah, there really wasn't anything wrong with the gross methodology - some days really are just like that. One would come out cut perfect and then one of the bends would go south. Better - finally got one I was happy with and dropped it on the floor during the drilling phase.

Last week I finished an equipment rack that I'd drawn out. Similar clown show. Designing with several factors in mind (road noise resonance, structural strength vs weight, tie points for the equipment). Steel rack, 5 carpeted (like speaker cabinet) plywood shelves supported by closed foam 'runners'. Came out exactly as I'd drawn (after having to find a different carpet material because the first buy had a backing that wouldn't take contact cement) but after standing back, lighting a smoke, and looking at my work - Suddenly felt like I'd just taken 10 hours and $160 to duplicate a high school AV cart. laugh.gif

Ya win, ya lose. c'est la vie.

This post has been edited by Jimthecopierwrench on January 02, 2012 06:48 pm


--------------------
All children are equal in the eyes of dog :)
PMEmail PosterUsers Website
Top
tekwiz
Posted: January 02, 2012 08:03 pm
Reply to this postQuote Post


Forum Addict ++
Group Icon

Group: Moderators
Posts: 28,711
Member No.: 5,746
Joined: July 24, 2006




You need to get some better equipment & some improvements to procedure.
Ending up only 2mm off isn't bad for what you were doing & what you were using.
However, measurement errors add up & add up quickly. The more steps you need, the more error.
A few points:
You can calibrate your framing square. First, run a file along the edges to make sure there are no bumps. Real squares are hardened to prevent this.
Next, place the square up against a wall, in contact with a piece of sheet metal resting on the same FLAT surface the horizontal leg of the square rests on. Scribe a vertical line. Flip the square over so that the horizontal leg moves 180. Scribe another line. The difference between the two is your error.
The square can be peened in a triangular pattern with it's vertex at the point where the two legs come together, or at the very corner of the square, depending on which way the square has to move to be at exactly 90.
Use this & the check method above to correct the square as necessary until both scribed lines are identical. This may take quite some time, depending on how much motion is necessary to fix things. You want to correct only half the overall error. Then you'll have an accurate square until the next time it's bent or dropped.
Next, all decent scales(rulers) have indented graduations. Using a divider(compass with two metal points) you place the divider on the scale & carefully adjust it until both legs drop into the indentations. This gives you consistent accuracy.
The same drop in action works for scribed lines...if you're careful, you can feel the 'click' when the point drops into the scribed line. One reason to always scribe deep.
The same method works for setting centerpunch marks, if you always start with a tiny sharp punch, then use a bigger one to get the punch mark to your desired size.
You also should be using layout dye or a substitute to increase contrast when working to a line. Then you can see & judge when a line is, approached, split in half, or just barely disappears. Exact work is difficult without it. Spray paint may be a decent substitute, if it doesn't flake away when scribed. KBC tools has the Dykem brand of layout fluid, which is what I use. This even comes in bingo dabbers, for easy application. These are ~$5 apiece & will cover several square feet.
A digital caliper makes an excellent divider for scribing lines. The most accurate method most of us have available. This can also be used to set regular dividers.
Most important: NEVER, never use the end of the scale for anything other than depth measurement. These wear from normal use. Always starting your divider setting on the one inch mark will eliminate any error from this source.

Using the correct methods, it's not difficult to produce work within a couple of thou just by eye working to scribed lines.


--------------------
Trouble rather the tiger in his lair, than the sage among his books.
For to you, kings & armies are things mighty & enduring.
To him, mere toys of the moment, to be overturned at the flick of a finger.

Fortuna favet fortibus.
PMEmail Poster
Top
Jimthecopierwrench
Posted: January 03, 2012 02:40 am
Reply to this postQuote Post


Moderator
Group Icon

Group: Moderators
Posts: 10,139
Member No.: 561
Joined: January 10, 2004




QUOTE
Ending up only 2mm off isn't bad for what you were doing & what you were using.
Downright horrible for the results I'm used to getting. Front frame on the beastie is 1/16" 'off' in 86", and square to the width of a ruler mark - Was intensely proud of those figures for a welded steel structure. I commonly hand file straight lines that don't allow light to pass between them and a striaght edge.

QUOTE
NEVER, never use the end of the scale for anything other than depth measurement. These wear from normal use. Always starting your divider setting on the one inch mark will eliminate any error from this source.
Yes, the two I have are used with kid gloves. never left on the bench, dropped, LOANED, used as paint lid pries, etc. They have their spot, hug when not in use, and wiped clean with a cotton rag - exactly because I'm a measurement nazi smile.gif

Framing square is just that. Good enough for wood and scribing lines along one edge, but it's not square (I do know how to bisect the error) and the error I've noticed depends on temperature.

QUOTE
get some better equipment
Can always use more of that, no? I've got a small stainless scale I use at work - surprise princess find of all places - that's just wonderfully accurate with very fine graticule etchings. It gets the same nazi treatment - no loan outs, don't touch it. Harder and harder to find good tools without premium prices though. The 'zero' settings on most 'handyman' tools are a sad joke.


--------------------
All children are equal in the eyes of dog :)
PMEmail PosterUsers Website
Top
AwesomeMatt
Posted: January 03, 2012 11:23 am
Reply to this postQuote Post


Forum Addict ++
*******

Group: Trusted Members
Posts: 3,000
Member No.: 9,878
Joined: June 21, 2007




QUOTE
Can always use more of that, no?


Depends, do you want the new place looking like the old kitchen? tongue.gif

You still have the cave btw, or did the landlord decide he wanted it back?
PMEmail Poster
Top
GPG
Posted: January 03, 2012 11:30 am
Reply to this postQuote Post


Forum Addict ++
*******

Group: Spamminator Taskforce
Posts: 6,219
Member No.: 792
Joined: April 25, 2004




What are you building? Waveguides are reasonably forgiving as long as joints are done well. Google choke joint microwave
PMEmail Poster
Top
CWB
Posted: January 03, 2012 02:18 pm
Reply to this postQuote Post


Forum Addict ++
*******

Group: Spamminator Taskforce
Posts: 21,225
Member No.: 15,154
Joined: May 15, 2008




dykem ... good stuff .
i've used a "magic marker" in a pinch .


--------------------
"Know how to solve every problem that has been solved"
R. Feynman '88
PM
Top
tekwiz
Posted: January 03, 2012 09:25 pm
Reply to this postQuote Post


Forum Addict ++
Group Icon

Group: Moderators
Posts: 28,711
Member No.: 5,746
Joined: July 24, 2006




QUOTE (Jimthecopierwrench @ January 02, 2012 05:40 pm)
QUOTE
Ending up only 2mm off isn't bad for what you were doing & what you were using.
Downright horrible for the results I'm used to getting. Front frame on the beastie is 1/16" 'off' in 86", and square to the width of a ruler mark - Was intensely proud of those figures for a welded steel structure. I commonly hand file straight lines that don't allow light to pass between them and a striaght edge.

QUOTE
NEVER, never use the end of the scale for anything other than depth measurement. These wear from normal use. Always starting your divider setting on the one inch mark will eliminate any error from this source.
Yes, the two I have are used with kid gloves. never left on the bench, dropped, LOANED, used as paint lid pries, etc. They have their spot, hug when not in use, and wiped clean with a cotton rag - exactly because I'm a measurement nazi smile.gif

Framing square is just that. Good enough for wood and scribing lines along one edge, but it's not square (I do know how to bisect the error) and the error I've noticed depends on temperature.

QUOTE
get some better equipment
Can always use more of that, no? I've got a small stainless scale I use at work - surprise princess find of all places - that's just wonderfully accurate with very fine graticule etchings. It gets the same nazi treatment - no loan outs, don't touch it. Harder and harder to find good tools without premium prices though. The 'zero' settings on most 'handyman' tools are a sad joke.

That scale sounds like the common 6" machinist scales..I've got a few around, but haven't seen any in months. Those things cost ~$1 per inch of length.
My main straightedge is an old Starett 48" machinist's scale from the early 40's.
Main square reference is a mid 40's 18" Starett 2 piece solid square, that was checked with a coordinate measuring machine & is within 19 millionths of an inch of exactly 90, at the end of the 18" side. This stays inside of it's oil soaked wooden storage case & only comes out when I need extreme accuracy, or to calibrate lesser squares. It was a gift from the first shop I ever worked in when they closed up. They treated the square in a similar fashion.
Want a shock? Look up the price for a new one. wink.gif
I mainly use it for setting my vise up on the mill table these days, as an alternative to indicating the vise. This is good enough for anything not needing tenth thou accuracies.

One piece of layout equipment that is still on my acquisition list is a surface plate. This with a height gauge is invaluable for all kinds of checking, setup & layout operations. I'm planning to acquire a granite sink cutout for an acceptable substitute. Even though real granite Chinese surface plates are dirt cheap these days.
For those that aren't familiar with them, surface plates are slabs of granite that exist only to present a known flat surface...flat within a few millionths of an inch over the entire surface. These start at 6" x 8", & go up to 4' x 8' or larger...4x8 is the biggest I've ever seen.
Granite is the preferred material both because of it's stability & because it doesn't dent. Any oopsies can at worst leave a chip, which won't interfere with accuracy like a dent would because there is no raised edge around them.


--------------------
Trouble rather the tiger in his lair, than the sage among his books.
For to you, kings & armies are things mighty & enduring.
To him, mere toys of the moment, to be overturned at the flick of a finger.

Fortuna favet fortibus.
PMEmail Poster
Top
Sch3mat1c
Posted: January 04, 2012 02:39 am
Reply to this postQuote Post


Forum Addict ++
Group Icon

Group: Moderators
Posts: 19,437
Member No.: 73
Joined: July 24, 2002




Average granite plate I've seen is about 2-3 feet across... every machine shop worth its swarf I've seen has several. They aren't light, like most machine tools.

Tim


--------------------
Answering questions is a tricky subject to practice. Not due to the difficulty of formulating or locating answers, but due to the human inability of asking the right questions; a skill that, were one to possess, would put them in the "answering" category.
PMEmail PosterUsers Website
Top
GPG
Posted: January 04, 2012 03:06 am
Reply to this postQuote Post


Forum Addict ++
*******

Group: Spamminator Taskforce
Posts: 6,219
Member No.: 792
Joined: April 25, 2004




QUOTE
Average granite plate I've seen is about 2-3 feet across.
Yep, me too.
PMEmail Poster
Top
johansen
Posted: January 04, 2012 11:45 am
Reply to this postQuote Post


Forum Addict ++
*******

Group: Trusted Members
Posts: 3,114
Member No.: 10,912
Joined: September 06, 2007




any granite counter top material stock should offer reasonable flatness vs. dollar cost average but by using such a material.. you would be essentially be doing what crane550 was doing when he build cnc machines from less than perfect stock tongue.gif

I for one, used the 3$/sq foot granite that was on sale at home depot last month...


--------------------
Can the Ethics Committee have a closed door meeting without imploding from cognitive dissonance?
PMEmail PosterUsers Website
Top
Geek
Posted: January 04, 2012 12:18 pm
Reply to this postQuote Post


Moderator
Group Icon

Group: Moderators
Posts: 9,943
Member No.: 62
Joined: July 23, 2002




Got a Liquidation World out there, Jim?

They had 6-8" x 2' granite scraps for $2/ea.

Cheers!


--------------------
-= Gregg =-
"Ratings are for transistors.....tubes have guidelines"
(please do not PM me for advice. Non-forum business messages will be ignored)
PMUsers Website
Top
tekwiz
Posted: January 04, 2012 08:30 pm
Reply to this postQuote Post


Forum Addict ++
Group Icon

Group: Moderators
Posts: 28,711
Member No.: 5,746
Joined: July 24, 2006




QUOTE (Sch3mat1c @ January 03, 2012 05:39 pm)
Average granite plate I've seen is about 2-3 feet across... every machine shop worth its swarf I've seen has several.  They aren't light, like most machine tools.

Tim

I had a small shop, so I had only one 18x24" unit. I thought about a bigger one, but that was the biggest size in the 2.5" thickness. After that, the thickness went up to 3" & the weight to over 300lbs. I didn't have an easy way to get one up onto my bench.

I'm a bit leery of using granite counter scraps, as they aren't thick enough to support heavy items without losing accuracy. This means that I would have to carefully bed one on epoxy into a specially constructed steel frame for strength.

I sure miss having a height gauge. Layout takes so much longer without one. sad.gif

Soon...


--------------------
Trouble rather the tiger in his lair, than the sage among his books.
For to you, kings & armies are things mighty & enduring.
To him, mere toys of the moment, to be overturned at the flick of a finger.

Fortuna favet fortibus.
PMEmail Poster
Top
0 User(s) are reading this topic (0 Guests and 0 Anonymous Users)
0 Members:

Topic Options Reply to this topicStart new topicStart Poll

 


:: support us ::




ElectronicsSkin by DutchDork & The-Force