Vernier caliper accuracy
Christopher Tidy
Can anyone tell me what the typical accuracy of a decent pair of vernier
calipers is (when used carefully and correctly)? I just bought a dial
bore gauge and am wondering if I can reasonably set the zero point with
vernier calipers instead of having to buy a 3" to 4" micrometer. The
calipers are Etalon brand, heavy and made in Switzerland. The bore gauge
dial is graduated down to 0.0005". I'm not reboring an engine, I just
want an idea of the wear, so I don't need the highest accuracy. From
using these calipers, my own guess is that they're probably good to
0.002", but some of that might be my errors in reading them. I'd be
interested to hear what other people think.
Best wishes,
Chris
Ed Huntress
"Christopher Tidy" <cdt22...@cantabgold.net> wrote in message
news:4B7F56B7...@cantabgold.net...
Keep them within a few degrees of some standard (68 deg. F being the most
common one); keep your hands off of them and wear gloves; develop a good
feel, and you should be able to measure to +/- 0.001" with them. Your Etalon
probably is like my TESA Swiss vernier caliper, which is my best one.
You can easily check it with a few gage blocks. If you're going to do
precision work, you need at least a few of them. A set is great but for
checking mikes, calipers, height gages and so on, you only need a few in
different sizes. Then you won't be chasing your tail. You'll know with good
assurance how accurate your gages are. And they'll help you develop a feel
for how much force to use when you're measuring. A couple of gage pins, like
1/4" and 5/8", or something like that, are also useful for that purpose. I
have some precision toolmaker's buttons that serve that purpose.
--
Ed Huntress
Christopher Tidy
> Keep them within a few degrees of some standard (68 deg. F being the most
> common one); keep your hands off of them and wear gloves; develop a good
> feel, and you should be able to measure to +/- 0.001" with them. Your Etalon
> probably is like my TESA Swiss vernier caliper, which is my best one.
This is the kind I have, except mine are 15" calipers and they don't
carry the P. Roch name:
http://cgi.ebay.com/ETALON-P-ROCH-ROLLE-SUISSE-12-30-mm-VERNIER-CALIPER_W0QQitemZ270531036164QQcmdZViewItemQQptZLH_DefaultDomain_0?hash=item3efce80404
Why do you suggest wearing gloves? To keep the calipers cool? Right now
the workshop is more like 40 deg. F.
> You can easily check it with a few gage blocks. If you're going to do
> precision work, you need at least a few of them. A set is great but for
> checking mikes, calipers, height gages and so on, you only need a few in
> different sizes. Then you won't be chasing your tail. You'll know with good
> assurance how accurate your gages are. And they'll help you develop a feel
> for how much force to use when you're measuring. A couple of gage pins, like
> 1/4" and 5/8", or something like that, are also useful for that purpose. I
> have some precision toolmaker's buttons that serve that purpose.
I don't have any gauge blocks. Do you think they'd be a better
investment than a 3" to 4" micrometer?
Best wishes,
Chris
Richard J Kinch
> I don't have any gauge blocks. Do you think they'd be a better
> investment than a 3" to 4" micrometer?
The import 1-2-3 blocks are $10 at Enco. Then you have standards for whole
inches 1 through 6 to better than 0.001". Add a granite slab and you have
the beginnings of some serious metrology for all of $35. Just buy $15 more
worth of anything and they'll ship it free, including the 61 lb stone!
http://www.use-enco.com/CGI/INSRIT?PMAKA=630-4010
http://www.use-enco.com/CGI/INSRIT?PMAKA=640-0120
This spacer set is only $30 and does 0.050" to 4 or 5 inches in most
0.001" increments:
Don Foreman
No. Calipers are not as accurate as micrometers because they spring
more and provide less "feel". Their value is wide range and quick
operation for ballpark measurements to within a coupla thou. That's
often quite sufficient, certainly in roughing or intermediate
operations.
I've found the Asian import micrometers to be very good value, and
some come with a reference with which to check them. Don't know about
sources in the UK but an Asian 3-4" mike can be bought here for about
$35. I have some good mikes (Etalon, Starrett, Fowler) and some Asian
imports. They agree to well under .001", usually to better than half
a thou. The better mikes do have better ratchets or friction
thimbles. I prefer a friction thimble, not found on cheap mikes.
If you want to measure accurately to .0002" or better, go name brand
from Switzerland, Japan or USA in that order. If .001" is close
enough, about any mike with carbide faces will suffice.
Ed Huntress
> Ed Huntress wrote:
>
>> Keep them within a few degrees of some standard (68 deg. F being the most
>> common one); keep your hands off of them and wear gloves; develop a good
>> feel, and you should be able to measure to +/- 0.001" with them. Your
>> Etalon probably is like my TESA Swiss vernier caliper, which is my best
>> one.
>
> This is the kind I have, except mine are 15" calipers and they don't carry
> the P. Roch name:
> http://cgi.ebay.com/ETALON-P-ROCH-ROLLE-SUISSE-12-30-mm-VERNIER-CALIPER_W0QQitemZ270531036164QQcmdZViewItemQQptZLH_DefaultDomain_0?hash=item3efce80404
>
> Why do you suggest wearing gloves? To keep the calipers cool? Right now
> the workshop is more like 40 deg. F.
The reason is thermal expansion of the caliper. If you're pushing for the
best accuracy, and particularly if the room is cold, holding it in your hand
for a few minutes can make a difference. If you warm a three-inch-long piece
of steel by 30 deg. F, from your 40 deg. shop to 70 deg., it will expand by
almost 0.001". If the piece you're measuring is the same temperature as the
caliper, and if they're both steel, the actual temperature matters little,
but a difference in their temperatures, if it's more than a few degrees, can
result in inaccurate measurement.
But it's time to back up. I suppose you realize that a slide caliper is not
a high-accuracy gage. It's usually used for moderate-accuracy work. You were
asking what accuracy you can achieve with it, and the answer is, using a
good caliper, cotton gloves, and calibration with gage blocks, you probably
can achieve +/- 0.001". But not everyone does. Some people have the touch
and some don't.
You'll be able to achieve that more easily with a micrometer. With good
technique and a good mike, you can cut that range in half. You won't need
gloves if you just hang your pinky or two fingers over the bridge part, or
if you work quickly.
>
>> You can easily check it with a few gage blocks. If you're going to do
>> precision work, you need at least a few of them. A set is great but for
>> checking mikes, calipers, height gages and so on, you only need a few in
>> different sizes. Then you won't be chasing your tail. You'll know with
>> good assurance how accurate your gages are. And they'll help you develop
>> a feel for how much force to use when you're measuring. A couple of gage
>> pins, like 1/4" and 5/8", or something like that, are also useful for
>> that purpose. I have some precision toolmaker's buttons that serve that
>> purpose.
>
> I don't have any gauge blocks. Do you think they'd be a better investment
> than a 3" to 4" micrometer?
That's a good question. If you need a mike, you'll want both. There will be
differences of opinion about this but my vote, if you already have a
caliper, would be for the gage blocks first.
But it depends on whether you're measuring *relative* dimensions or
*absolute* dimensions. If it's the former, you don't need a well-calibrated
gage. If it's the latter, you'll never know for sure how accurately your
gages measure unless you check them from time to time against a gage block
(or a stack) that's somewhere in the middle of the gage's range, and another
one near its largest opening. For a 3" - 4" mike, I'd want at least a 3"
block (or a stack to make that dimension) and a 4" block or stack. I'd like
to have a 1/2" block, too, to wring with the 3" block to test the midrange.
If any of this is unclear, ask, and I or someone else will explain. FWIW,
most home-shop work doesn't require accurate measurement of absolute
dimensions. Usually we're trying to make two things fit together, and what
you need to know for that is their *relative* dimensions. If that's the
case, forget the gage blocks for now.
--
Ed Huntress
John
You can buy a cheap set of B grade gauge blocks for about 85 bucks.
They are accurate enough for setting up and calibration any gauge you
will normally use. When you set your gauges make sure they are all at
the same temp. Gloves will keep the metal from getting warm and also
keep your fingerprints off the gauge blocks. Also if you have sine plate
to set up angles, you will definitely like to have a set of gauge blocks.
John
Snag
> Christopher Tidy writes:
>
>> I don't have any gauge blocks. Do you think they'd be a better
>> investment than a 3" to 4" micrometer?
>
> The import 1-2-3 blocks are $10 at Enco. Then you have standards for
> whole inches 1 through 6 to better than 0.001".
And you just answered the question I was about to ask . Thanks !
--
Snag
Still learnin'
after all
these years .
Jim Wilkins
wrote:
>
> Can anyone tell me what the typical accuracy of a decent pair of vernier
> calipers is (when used carefully and correctly)? I just bought a dial
> bore gauge and am wondering if I can reasonably set the zero point with
> Chris
I have good 0.0001" micrometers with a 1" setting standard for 0-1 and
1-2", cheaper second-hand ones for the larger ranges.
At home I almost never need to measure larger sizes with absolute
accuracy. I have both the inside and outside pieces and only need to
cut them to fit together. I use an unzeroed bore gauge to check for
taper while I'm boring the cylinder, and compare that to the bore
gauge's reading of the micrometer set to the piston's OD.
I've had Starrett and Phase 2 dial calipers checked for accuracy. The
Starrett was considerably better than 0.001" over its range, nearly as
good as I could read by interpolation. The Phase 2 barely held 0.001".
When I checked the cal standards myself I noticed that the Phase 2
reading was sensitive to how I held it.
So I bought an incomplete set of gage blocks cheap at an auction and
practiced on them. After a while I could get the 0.0001" mikes to read
to one division but don't believe my dial and vernier calipers to
better than 0.001", if that.
This Enco 1-2-3 block here shows 2.0002" at both ends, at 14C (my
normal room temperature). That's good enough to set a 2-3" micrometer
that reads to 0.001".
jsw
Grumpy
I would suggest that you calibrate it against a new bearing bore. The
accuracy that those are ground to is amazing.
Wes
I belive Christopher is a bit far away for free shipping.
Wes
John
Yup, I use bearings as standards after checking them, usually for
setting dial snap gauges and bore gauges. An ABEC 1 bearing can be
smaller than the nominal size but if you can check it against known
standards they are an inexpensive was to have a working standard for
checking and setting measuring instruments.
John
DoN. Nichols
> Hi folks,
>
> Can anyone tell me what the typical accuracy of a decent pair of vernier
> calipers is (when used carefully and correctly)?
How new are the calipers? I would expect 0.001" accuracy, if
zeroed properly, if the gibs are set right so the head does not rotate,
and if you have sufficient light and magnification to read the vernier
properly.
I'm assuming that you are not calling dial calipers "vernier".
Just to be sure, I'll cover them too.
Digital calipers typically have a resolution of 0.0005", but
check the actual accuracy against gauge blocks -- in particular at the
size you want to set.
Dial calipers can have cyclic errors a function of the gear
pinion which engages the rack.
> I just bought a dial
> bore gauge and am wondering if I can reasonably set the zero point with
> vernier calipers instead of having to buy a 3" to 4" micrometer. The
> calipers are Etalon brand, heavy and made in Switzerland. The bore gauge
> dial is graduated down to 0.0005".
Ideally, it should be set to a ring gauge, not a micrometer.
These will typically be accurate to 0.000050". You really want the
standard to be more accurate than what is being set.
If you use a 4-5" micrometer, you will need a 4" standard to
zero it before you use it. (And I'm assuming that the micrometer has a
tenths vernier around the barrel so you can set it to 0.0001".
Also -- you really don't want to hold the micrometer's C-frame,
as it can expand with heat from the hand. Put it in a micrometer bench
clamp so you can set it and not worry about it drifting as it cools off.
> I'm not reboring an engine, I just
> want an idea of the wear, so I don't need the highest accuracy. From
> using these calipers, my own guess is that they're probably good to
> 0.002", but some of that might be my errors in reading them. I'd be
> interested to hear what other people think.
A lot of the variation in calipers comes from a loose fit of the
head to the bar, allowing it to tilt when pressure is put on the jaw
tips. If you have it pretty snug, you can keep this from being a
problem. And the closer to the bar you do the measurement, the less the
problem from a tilting head and jaws.
Enjoy,
DoN.
--
Email: <dnic...@d-and-d.com> | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---
DoN. Nichols
>
>> Keep them within a few degrees of some standard (68 deg. F being the most
>> common one); keep your hands off of them and wear gloves; develop a good
>> feel, and you should be able to measure to +/- 0.001" with them. Your Etalon
>> probably is like my TESA Swiss vernier caliper, which is my best one.
>
> This is the kind I have, except mine are 15" calipers and they don't
> carry the P. Roch name:
> http://cgi.ebay.com/ETALON-P-ROCH-ROLLE-SUISSE-12-30-mm-VERNIER-CALIPER_W0QQitemZ270531036164QQcmdZViewItemQQptZLH_DefaultDomain_0?hash=item3efce80404
O.K. That looks like a very well made one -- except that there
seems to be no provision for sliding the vernier scale to zero it as
there is on some. So -- it is either quite accurate, or permanently
inaccurate, and it was originally zeroed by lapping the jaws until it
reached zero.
> Why do you suggest wearing gloves? To keep the calipers cool? Right now
> the workshop is more like 40 deg. F.
To keep the heat of your hand from changing the dimensions of
the metal. Note that he suggests that you keep it within a few degrees
of 68F (20C) since this is probably the temperature at which it was
originally calibrated.
>> You can easily check it with a few gage blocks. If you're going to do
>> precision work, you need at least a few of them. A set is great but for
>> checking mikes, calipers, height gages and so on, you only need a few in
>> different sizes. Then you won't be chasing your tail. You'll know with good
>> assurance how accurate your gages are. And they'll help you develop a feel
>> for how much force to use when you're measuring. A couple of gage pins, like
>> 1/4" and 5/8", or something like that, are also useful for that purpose. I
>> have some precision toolmaker's buttons that serve that purpose.
>
> I don't have any gauge blocks. Do you think they'd be a better
> investment than a 3" to 4" micrometer?
They would be a good investment even if you *have* a 3-4"
micrometer, as it allows you to check the micrometer at several points
through its range. At best, the micrometer came with a single standard
rod for the zero position 3" for a 3-4" micrometer. The same advice
about handling with gloves applies here, because both the gauge blocks
and the micrometers also change dimensions with temperature.
Probably for your purpose, the cheap Chinese sets are good
enough. They are supposedly accurate within 0.000050" (half a tenth)
which is enough beyond the accuracy you expect even from the micrometer
so it should be a good way to check it.
BTW For precise checking of bores, you want a three-point contact
bore micrometer, which is self centering in the bore. They are
*expensive* -- say about $600.00 for a 1.000" to 1.200" one
which reads in divisions of 0.0002". And each pair in adjacent
sizes should come with a setting ring which is at the high side
of the range of one and the low side of the range of the other.
I've collected most of mine from eBay auctions over the years,
and mostly have Tesa/Brown & Sharpe ones.
Good Luck,
Richard J Kinch
> I belive Christopher is a bit far away for free shipping.
Oh, right.
Christopher Tidy
Hi Don,
Sorry to everyone for the slight delay in replying.
> No. Calipers are not as accurate as micrometers because they spring
> more and provide less "feel". Their value is wide range and quick
> operation for ballpark measurements to within a coupla thou. That's
> often quite sufficient, certainly in roughing or intermediate
> operations.
That was about the accuracy I was guessing. As calipers go, these are
about as stiff as you can get. Probably weigh between 1 and 2 lbs. But
obviously there's no mechanism to ensure a constant closing force, like
a micrometer thimble.
I figure that as I'm not reboring an engine (I just want to know how
severe the wear on each part is, so I can decide what to replace), a
measurement to a couple of thousandths is good enough for me. After all,
the smallest oversize piston I've seen is +0.010". But if anyone thinks
this is a bad plan, do let me know.
> I've found the Asian import micrometers to be very good value, and
> some come with a reference with which to check them. Don't know about
> sources in the UK but an Asian 3-4" mike can be bought here for about
> $35. I have some good mikes (Etalon, Starrett, Fowler) and some Asian
> imports. They agree to well under .001", usually to better than half
> a thou. The better mikes do have better ratchets or friction
> thimbles. I prefer a friction thimble, not found on cheap mikes.
I've never seen a micrometer without a thimble. My 0.0001" Tesa
micrometer (Swiss I think) has a friction thimble and my cheap Draper
metric micrometer still has a ratchet thimble.
What do people think of micrometers with interchangeable anvils, to give
several measurement ranges? So for example you have four anvils which
fit a 4" micrometer frame to give measuring ranges of 0" to 1", 1" to
2", 2" to 3" and 3" to 4"? My gut reaction is that they won't be as
consistent as a single range micrometer, but for brand new ones,
Mitutoyo's website claims this isn't the case.
If calipers are good to a couple of thousandths, my inclination is to
use them for the moment.
Best wishes,
Chris
Christopher Tidy
Hi Ed,
> The reason is thermal expansion of the caliper. If you're pushing for the
> best accuracy, and particularly if the room is cold, holding it in your hand
> for a few minutes can make a difference. If you warm a three-inch-long piece
> of steel by 30 deg. F, from your 40 deg. shop to 70 deg., it will expand by
> almost 0.001". If the piece you're measuring is the same temperature as the
> caliper, and if they're both steel, the actual temperature matters little,
> but a difference in their temperatures, if it's more than a few degrees, can
> result in inaccurate measurement.
Right. I rarely wear gloves except for particularly dirty jobs. I prefer
barrier cream.
> But it's time to back up. I suppose you realize that a slide caliper is not
> a high-accuracy gage. It's usually used for moderate-accuracy work. You were
> asking what accuracy you can achieve with it, and the answer is, using a
> good caliper, cotton gloves, and calibration with gage blocks, you probably
> can achieve +/- 0.001". But not everyone does. Some people have the touch
> and some don't.
Thanks. That's the figure I wanted. These are good calipers and I think
my touch is fair, but I'm not being careful about temperature and the
scale on the calipers cannot be moved to calibrate them. Sounds like my
guess of +/- 0.002" is probably fair.
> But it depends on whether you're measuring *relative* dimensions or
> *absolute* dimensions. If it's the former, you don't need a well-calibrated
> gage. If it's the latter, you'll never know for sure how accurately your
> gages measure unless you check them from time to time against a gage block
> (or a stack) that's somewhere in the middle of the gage's range, and another
> one near its largest opening. For a 3" - 4" mike, I'd want at least a 3"
> block (or a stack to make that dimension) and a 4" block or stack. I'd like
> to have a 1/2" block, too, to wring with the 3" block to test the midrange.
What's a stack?
> If any of this is unclear, ask, and I or someone else will explain. FWIW,
> most home-shop work doesn't require accurate measurement of absolute
> dimensions. Usually we're trying to make two things fit together, and what
> you need to know for that is their *relative* dimensions. If that's the
> case, forget the gage blocks for now.
In this case, I think it's the absolute dimensions. The engine's piston
wobbles noticeably in the bore. It's an aluminium piston in a cast iron
bore, so I suspect there could be much more wear on the piston than the
bore (the bore looks good, without a ridge at the top). So I want to
know how well a brand new piston will fit in the existing bore, before I
buy one.
Best wishes,
Chris
Christopher Tidy
Hi Don,
> How new are the calipers? I would expect 0.001" accuracy, if
> zeroed properly, if the gibs are set right so the head does not rotate,
> and if you have sufficient light and magnification to read the vernier
> properly.
Probably 30+ years old, but in very good condition. A few scratches here
and there, but no noticeable wear to the jaws and no sign of having been
dropped.
> I'm assuming that you are not calling dial calipers "vernier".
> Just to be sure, I'll cover them too.
They're genuine verniers.
> Ideally, it should be set to a ring gauge, not a micrometer.
> These will typically be accurate to 0.000050". You really want the
> standard to be more accurate than what is being set.
I know this is essential for a 3-point bore gauge, but is it so
important for a 2-point gauge? Mine is a 2-point gauge with a centring
device.
I looked at ring gauges. Any idea whey they're made in weird sizes
instead of round numbers? Such as this one, 16.8349.mm:
http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&item=310146851030
> A lot of the variation in calipers comes from a loose fit of the
> head to the bar, allowing it to tilt when pressure is put on the jaw
> tips. If you have it pretty snug, you can keep this from being a
> problem. And the closer to the bar you do the measurement, the less the
> problem from a tilting head and jaws.
Good tip. But fortunately the jaws on my calipers are snug.
Best wishes,
Chris
Ed Huntress
Interchangeable anvils are fine if you keep them very clean, and if you
check them against gage blocks.
Christopher Tidy
Hi Don,
>>This is the kind I have, except mine are 15" calipers and they don't
>>carry the P. Roch name:
>>http://cgi.ebay.com/ETALON-P-ROCH-ROLLE-SUISSE-12-30-mm-VERNIER-CALIPER_W0QQitemZ270531036164QQcmdZViewItemQQptZLH_DefaultDomain_0?hash=item3efce80404
>
>
> O.K. That looks like a very well made one -- except that there
> seems to be no provision for sliding the vernier scale to zero it as
> there is on some. So -- it is either quite accurate, or permanently
> inaccurate, and it was originally zeroed by lapping the jaws until it
> reached zero.
Indeed. I searched for the type of calipers which have a vernier scale
that can be moved to zero them, but I couldn't find any for sale (maybe
they're an American thing - what brands have this feature?). So I bought
these.
> BTW For precise checking of bores, you want a three-point contact
> bore micrometer, which is self centering in the bore. They are
> *expensive* -- say about $600.00 for a 1.000" to 1.200" one
> which reads in divisions of 0.0002". And each pair in adjacent
> sizes should come with a setting ring which is at the high side
> of the range of one and the low side of the range of the other.
I'm specifically interested in engine cylinders at the moment. The first
one I want to measure has a nominal bore of 80 mm. So that would likely
be a very expensive micrometer.
Before buying the gauge, I did some research and got the impression that
the standard tool for measuring engine cylinders is the dial indicator
bore gauge, with two measurement points and a centring device. That way,
you can tell the direction in which any ovality lies, which would be of
interest to me. So that's what I bought.
Best wishes,
Chris
Christopher Tidy
> Interchangeable anvils are fine if you keep them very clean, and if you
> check them against gage blocks.
Sounds like I should avoid them unless I buy the gauge blocks then.
Chris
Ed Huntress
Multiple gage blocks "wrung" together, which add up to their combined
dimension. If you wring three 1" gage blocks together, they'll equal a
single, 3" block. The wringing technique excludes air between them, and adds
less than 2 millionths of an inch to the stack height.
This only works with good-quality gage blocks. Some of the other setting
tools that have been recommended in this thread can't be stacked up with
that kind of accuracy.
The ideal accuracy of dimensional standards, such as gage blocks, is ten
times or more the accuracy of what you're trying to measure. In other words,
if you're trying to determine the accuracy of your micrometer to +/- 0.0001
in., you ideally should have gage blocks that are accurate to +/- 0.000,01
in. That's a workshop-quality gage block, grade A or a good quality grade B,
which today is called a "grade 3" (or AS-1) in new blocks.
If you're going metric, the picture is a little different.
Having said all that, a cheap workshop-grade gage block will be more
accurate than you need for most shop work. But the troubles with cheap sets
are twofold: they have parallelism problems, as well as looseness in
absolute dimensions; and they often won't wring.
Back to Earth: For what you want, almost anything will do. <g> Any decent
gage standard is a whole lot better than nothing. But you have at least one
mike, and you might want to use it for more demanding work at some time. For
that, get a few gage blocks.
>
>> If any of this is unclear, ask, and I or someone else will explain. FWIW,
>> most home-shop work doesn't require accurate measurement of absolute
>> dimensions. Usually we're trying to make two things fit together, and
>> what you need to know for that is their *relative* dimensions. If that's
>> the case, forget the gage blocks for now.
>
> In this case, I think it's the absolute dimensions. The engine's piston
> wobbles noticeably in the bore. It's an aluminium piston in a cast iron
> bore, so I suspect there could be much more wear on the piston than the
> bore (the bore looks good, without a ridge at the top). So I want to know
> how well a brand new piston will fit in the existing bore, before I buy
> one.
>
> Best wishes,
>
> Chris
It would really help to have a good engine man chime in here. There are
several around; try a new thread with a title like "Measuring engine bores"
or something like that. You may drag one up.
Many of us can help you with handling gages but engine cylinders have other
issues, like taper and ovality, etc. If it's an old engine, the pistons may
be cylindrical. If it's a newer one, they're probably elliptical ("oval").
And they're tapered along their lengths, too.
In engine work, you have to know not only how to measure, but what it is
you're really measuring. It's not difficult but you do have to know what
you're doing. You want to ask someone who really knows his stuff.
Good luck!
--
Ed Huntress
Ed Huntress
I'd avoid them unless you really need them. Full disclosure -- Mitutoyo was
my client for many years, and I wrote all of their articles during those
years and some of their instructional materials. They're quite honest about
what they claim. Just be aware that things like interchangeable-anvil
micrometers sacrifice some assurance for the sake of convenience. They can
be as accurate as any mike; you just have to be a bit more careful.
--
Ed Huntress
Pete Snell
accuracy of the inner and outer jaws. On one set cheapie Chinese digital
calipers I have, the difference between an inside and outside
measurement (of identical dimension) was about 0.003"
Pete
--
Pete Snell
Department of Physics
Royal Military College
Kingston, Ontario,
Canada
-----------------------------------------------------------------------
Either this man is dead or my watch has stopped.
Groucho Marx (1895�1977)
Ed Huntress
"Pete Snell" <sne...@rmc.ca> wrote in message
news:4b829c5e$1...@win9.rmc.ca...
> One thing to be very careful of with all calipers, is the comparative
> accuracy of the inner and outer jaws. On one set cheapie Chinese digital
> calipers I have, the difference between an inside and outside measurement
> (of identical dimension) was about 0.003"
>
> Pete
Save that one for woodwork. <g>
--
Ed Huntress
Larry Jaques
<cdt22...@cantabgold.net> scrawled the following:
>Ed Huntress wrote:
>
>Hi Ed,
>
>> The reason is thermal expansion of the caliper. If you're pushing for the
>> best accuracy, and particularly if the room is cold, holding it in your hand
>> for a few minutes can make a difference. If you warm a three-inch-long piece
>> of steel by 30 deg. F, from your 40 deg. shop to 70 deg., it will expand by
>> almost 0.001". If the piece you're measuring is the same temperature as the
>> caliper, and if they're both steel, the actual temperature matters little,
>> but a difference in their temperatures, if it's more than a few degrees, can
>> result in inaccurate measurement.
>
>Right. I rarely wear gloves except for particularly dirty jobs. I prefer
>barrier cream.
Now that I'm no longer doing greasy mechanical work, I tend to wear
gloves more often, to keep my finnernails clean. Barrier cream is
great, except for that.
>> But it's time to back up. I suppose you realize that a slide caliper is not
>> a high-accuracy gage. It's usually used for moderate-accuracy work. You were
>> asking what accuracy you can achieve with it, and the answer is, using a
>> good caliper, cotton gloves, and calibration with gage blocks, you probably
>> can achieve +/- 0.001". But not everyone does. Some people have the touch
>> and some don't.
>
>Thanks. That's the figure I wanted. These are good calipers and I think
>my touch is fair, but I'm not being careful about temperature and the
>scale on the calipers cannot be moved to calibrate them. Sounds like my
>guess of +/- 0.002" is probably fair.
Fair to high. 2 thou to start, 1 thou after you get used to it, 5
tenths once you're good and comfortable with your dial calipers. But
for critical measurements, micrometers are the way to go.
>> But it depends on whether you're measuring *relative* dimensions or
>> *absolute* dimensions. If it's the former, you don't need a well-calibrated
>> gage. If it's the latter, you'll never know for sure how accurately your
>> gages measure unless you check them from time to time against a gage block
>> (or a stack) that's somewhere in the middle of the gage's range, and another
>> one near its largest opening. For a 3" - 4" mike, I'd want at least a 3"
>> block (or a stack to make that dimension) and a 4" block or stack. I'd like
>> to have a 1/2" block, too, to wring with the 3" block to test the midrange.
>
>What's a stack?
When a gage block isn't thick enough, you stack another on top of it.
>> If any of this is unclear, ask, and I or someone else will explain. FWIW,
>> most home-shop work doesn't require accurate measurement of absolute
>> dimensions. Usually we're trying to make two things fit together, and what
>> you need to know for that is their *relative* dimensions. If that's the
>> case, forget the gage blocks for now.
>
>In this case, I think it's the absolute dimensions. The engine's piston
>wobbles noticeably in the bore. It's an aluminium piston in a cast iron
>bore, so I suspect there could be much more wear on the piston than the
>bore (the bore looks good, without a ridge at the top). So I want to
>know how well a brand new piston will fit in the existing bore, before I
>buy one.
Remember that the COE is different for steel and aluminum, with steel
expanding less by half. Aluminum pistons are wobblier by nature.
Don't go by the piston, go by the bore. It sounds like your engine is
in fairly good shape, but check for elongation perpendicular to the
crankshaft. Bores become oval from wear, where the connecting rod
pushes 'em up and drags 'em down the opposite sides. If you bore it
out, you'll need all new pistons by default.
--
"Just think of the tragedy of teaching children not to doubt."
-- Clarence Darrow
Larry Jaques
<cdt22...@cantabgold.net> scrawled the following:
>DoN. Nichols wrote:
>
>Hi Don,
>
>> How new are the calipers? I would expect 0.001" accuracy, if
>> zeroed properly, if the gibs are set right so the head does not rotate,
>> and if you have sufficient light and magnification to read the vernier
>> properly.
>
>Probably 30+ years old, but in very good condition. A few scratches here
>and there, but no noticeable wear to the jaws and no sign of having been
>dropped.
>
>> I'm assuming that you are not calling dial calipers "vernier".
>> Just to be sure, I'll cover them too.
>
>They're genuine verniers.
Then your 2 thou is likely more hopeful. Get a pair of dials, boy!
They're less easy to misread. I forgot you were talking about
verniers. Got a good magnifying glass? ;)
Pete Snell
Yes, or plumbing. I think that all calipers should be guilty of this
shortcoming until proven innocent. It's something to beware of anyhow.
Pete
--
Pete Snell
Department of Physics
Royal Military College
Kingston, Ontario,
Canada
-----------------------------------------------------------------------
" The first instance of novel principle is invariably defeated
by the developed example of established practice."
Lawrence Pomeroy (1883-1941)
Ned Simmons
<cdt22...@cantabgold.net> wrote:
>DoN. Nichols wrote:
>
>Hi Don,
>
>>>This is the kind I have, except mine are 15" calipers and they don't
>>>carry the P. Roch name:
>>>http://cgi.ebay.com/ETALON-P-ROCH-ROLLE-SUISSE-12-30-mm-VERNIER-CALIPER_W0QQitemZ270531036164QQcmdZViewItemQQptZLH_DefaultDomain_0?hash=item3efce80404
>>
>>
>> O.K. That looks like a very well made one -- except that there
>> seems to be no provision for sliding the vernier scale to zero it as
>> there is on some. So -- it is either quite accurate, or permanently
>> inaccurate, and it was originally zeroed by lapping the jaws until it
>> reached zero.
>
>Indeed. I searched for the type of calipers which have a vernier scale
>that can be moved to zero them, but I couldn't find any for sale (maybe
>they're an American thing - what brands have this feature?). So I bought
>these.
>
The Starrett 123 series calipers have adjustable scales and are as
good as verniers get. The 50 division scales with the vernier and main
scale in the same plane make them very easy to read. I can read them
without glasses in a pinch, though I don't recommend it.
http://www.starrett.com/download/62_123_caliper.pdf
BTW, I have a pair of 8" verniers that, other than the length and
Mauser brand marked on them, appear to be identical to yours.
--
Ned Simmons
Ned Simmons
wrote:
>
>
>I would suggest that you calibrate it against a new bearing bore. The
>accuracy that those are ground to is amazing.
>
It is pretty amazing what you get for your money when purchasing
bearings. But even though the vast majority of bearings will likely be
better, the bore of a bearing in this size range could be close to
.001 low and another .001 out of round. As you might expect, thin
section bearings have a wider roundness tolerance than the standard
62xx and 63xx series. So, trust, but verify. <g>
--
Ned Simmons
DoN. Nichols
> Don Foreman wrote:
>
> Hi Don,
>
> Sorry to everyone for the slight delay in replying.
>
>> No. Calipers are not as accurate as micrometers because they spring
>> more and provide less "feel". Their value is wide range and quick
>> operation for ballpark measurements to within a coupla thou. That's
>> often quite sufficient, certainly in roughing or intermediate
>> operations.
>
> That was about the accuracy I was guessing. As calipers go, these are
> about as stiff as you can get. Probably weigh between 1 and 2 lbs. But
> obviously there's no mechanism to ensure a constant closing force, like
> a micrometer thimble.
Indeed -- you need to develop a feel for the sliding force on
the caliper jaws once closed. (And, of course, the jaws are not carbide
faced, so they will wear more rapidly than most micrometer jaws.)
> I figure that as I'm not reboring an engine (I just want to know how
> severe the wear on each part is, so I can decide what to replace), a
> measurement to a couple of thousandths is good enough for me. After all,
> the smallest oversize piston I've seen is +0.010". But if anyone thinks
> this is a bad plan, do let me know.
Probably adequate -- especially as you are closing the calipers
to a given setting (using the adjustment screw with the extra traveling
head clamped down) and then using it as a reference for the dial gauge.
If you had a set of gauge blocks, you could close it to a light sliding
fit on those (to free yourself from the possible errors in the vernier
and in reading it) and hold it in a vise to eliminate the problems from
hand heat causing expansion of the beam.
>> I've found the Asian import micrometers to be very good value, and
>> some come with a reference with which to check them. Don't know about
>> sources in the UK but an Asian 3-4" mike can be bought here for about
>> $35. I have some good mikes (Etalon, Starrett, Fowler) and some Asian
>> imports. They agree to well under .001", usually to better than half
>> a thou. The better mikes do have better ratchets or friction
>> thimbles. I prefer a friction thimble, not found on cheap mikes.
>
> I've never seen a micrometer without a thimble. My 0.0001" Tesa
> micrometer (Swiss I think) has a friction thimble and my cheap Draper
> metric micrometer still has a ratchet thimble.
My first one from about 1960 (long gone, and I don't remember
the brand) had only a small diameter spinner, not a ratchet or friction
thimble, and I developed a feel for letting my fingers slip on the
knurled portion of the man thimble.
I've got others with various designs, including a 0-6" Brown &
Sharpe set which still have no torque limiting features, along with
others with very good friction thimbles.
> What do people think of micrometers with interchangeable anvils, to give
> several measurement ranges? So for example you have four anvils which
> fit a 4" micrometer frame to give measuring ranges of 0" to 1", 1" to
> 2", 2" to 3" and 3" to 4"? My gut reaction is that they won't be as
> consistent as a single range micrometer, but for brand new ones,
> Mitutoyo's website claims this isn't the case.
Remember that you have to zero them against a standard each time
you change the anvils -- or for that matter if you pick it up a week
later and want to measure -- the room temperature might have changed
between the zeroing and the time of the intended measurement.
My main consideration of the problems with the multi-range
micrometers is that you have the large C frame so there is more metal to
expand as your hand warms it, thus more error from the thermal
expansion.
If you can prevent this (including hand warming of the standards
you use when you zero the micrometers) you should be OK. Often a bench
clamp stand for the micrometer is indicated.
> If calipers are good to a couple of thousandths, my inclination is to
> use them for the moment.
And potentially good for better than that as you will be using
them in zeroing the bore gauge. If you use gauge blocks to set the
calipers, you will be better than the caliper.
But then, if you have a stack of gauge blocks, and two longer
blocks at the ends, and you have a setting reference which won't need
the calipers at all. There are devices designed for just this purpose
-- a rectangular tube to hold the blocks, a pair of long blocks for the
ends, and provisions for clamping it down firmly.
DoN. Nichols
They *should* come with standards for zeroing the micrometer
after changing the anvils -- as many standards as you have anvils.
These should be enough to assure accuracy at the zero point at least,
and you can use gauge blocks to check for errors in mid spindle travel,
but this should not be a problem with new micrometers anyway -- only
after a lifetime of use by you, or someone else.
Enjoy,
DoN. Nichols
>
> Hi Don,
>
>>>This is the kind I have, except mine are 15" calipers and they don't
>>>carry the P. Roch name:
>>>http://cgi.ebay.com/ETALON-P-ROCH-ROLLE-SUISSE-12-30-mm-VERNIER-CALIPER_W0QQitemZ270531036164QQcmdZViewItemQQptZLH_DefaultDomain_0?hash=item3efce80404
>>
>>
>> O.K. That looks like a very well made one -- except that there
>> seems to be no provision for sliding the vernier scale to zero it as
>> there is on some. So -- it is either quite accurate, or permanently
>> inaccurate, and it was originally zeroed by lapping the jaws until it
>> reached zero.
>
> Indeed. I searched for the type of calipers which have a vernier scale
> that can be moved to zero them, but I couldn't find any for sale (maybe
> they're an American thing - what brands have this feature?). So I bought
> these.
Well ... the ones which I have are quite old. These days, most
people buy either dial or digital calipers. Dial is nice if you can
keep chips out of the rack, but Digital (if good) is far more
convenient, including the ability to zero at any point (such as on what
you are fitting to) and then measure how much more needs to be removed
to reach the proper dimensions. Also, measuring the distance between
the centers of two holes of identical size by zeroing it while measuring
the ID of one hole, then shift to measuring the distance between two
opposite sides of the two holes in question. This subtracts the
diameter of one hole from the final measurement which is the sum of the
center distance and half of each hole's diameter, giving the actual
center-to-center distance.
>> BTW For precise checking of bores, you want a three-point contact
>> bore micrometer, which is self centering in the bore. They are
>> *expensive* -- say about $600.00 for a 1.000" to 1.200" one
>> which reads in divisions of 0.0002". And each pair in adjacent
>> sizes should come with a setting ring which is at the high side
>> of the range of one and the low side of the range of the other.
>
> I'm specifically interested in engine cylinders at the moment. The first
> one I want to measure has a nominal bore of 80 mm. So that would likely
> be a very expensive micrometer.
>
> Before buying the gauge, I did some research and got the impression that
> the standard tool for measuring engine cylinders is the dial indicator
> bore gauge, with two measurement points and a centring device. That way,
> you can tell the direction in which any ovality lies, which would be of
> interest to me. So that's what I bought.
O.K. But it misses three-lobed bores, which can be in a shape
which actually measures the same between any two opposite points, but
still is potentially seriously out of round. (This is more likely to be
formed by certain machining operations -- including centerless grinding
for OD shapes, and in a used automobile engine, the ovality is more
likely, and for that your bore gauge is probably quite adequate.
To show that kind of error (the three-lobed one) with the
three-point micrometer, you would want to do at least two measurements
one rotated 60 degrees from the other -- and ideally yet another
half-way between them just to be sure.
For measuring this sort of thing for outside diameters, there
are V-anvil micrometers which duplicate the three-point contact.
(Actually, there are two styles -- one for the three-lobed measurement,
and another for measuring the diameter of five-flute milling cutters.)
DoN. Nichols
>
> Hi Don,
>
>> How new are the calipers? I would expect 0.001" accuracy, if
>> zeroed properly, if the gibs are set right so the head does not rotate,
>> and if you have sufficient light and magnification to read the vernier
>> properly.
>
> Probably 30+ years old, but in very good condition. A few scratches here
> and there, but no noticeable wear to the jaws and no sign of having been
> dropped.
O.K. And from the image which you posted and I saw after my
question, it looks as though it has a very long bearing surface on the
moving jaw assembly, so there should be very little jaw rotation,
especially if the gibs are set properly.
>> I'm assuming that you are not calling dial calipers "vernier".
>> Just to be sure, I'll cover them too.
>
> They're genuine verniers.
As your later posted image showed. (Or was that someone else's
image? I lose track from time to time. :-)
I asked because in other discussions -- here and elsewhere --
I've been led astray by people using "vernier" as a term for any slide
bar calipers. I just wanted to be sure that we were talking about the
same thing. :-)
>> Ideally, it should be set to a ring gauge, not a micrometer.
>> These will typically be accurate to 0.000050". You really want the
>> standard to be more accurate than what is being set.
>
> I know this is essential for a 3-point bore gauge, but is it so
> important for a 2-point gauge? Mine is a 2-point gauge with a centring
> device.
Well ... having the setting device more accurate than what is
being set is always good practice. As I just mentioned a bit ago (and
you of course had not seen when you posted this), a stack of gauge
blocks with two long blocks sticking out to one side will make a more
accurate setting reference than your calipers.
Note that with the calipers, or with the gauge block stack, if
the angle of the gauge is a bit off, the reading will be too high (thus
the setting will be too small), while with a ring gauge, error in
following the diameter will be too small, while error in keeping the
probe shaft parallel to the axis of the bore will cause it to read too
large.
> I looked at ring gauges. Any idea whey they're made in weird sizes
> instead of round numbers? Such as this one, 16.8349.mm:
> http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&item=310146851030
Well ... this one is for setting an air gauge, so it has to be
pretty close to the diameter of the probe. An air gauge is hollow with
both ends sealed, and air feed in through one end, with a hole of a
specific size (or maybe more holes? -- I haven't actually seen one --
just read about them in books). The air goes out the hole, and flows
out between the probe walls and the ID of the bore being checked, and
the pressure built up in the interior of the gauge is measured and is a
very good indication of the difference between the OD of the gauge and
the ID of the bore -- over a fairly narrow range. So -- this ring was
for setting something quite close to 16.8349 mm --- maybe 16.8350 mm is
the actual target size. Anyway -- the rings are made too small,
hardened, and then ground to a diameter near what is needed, and then
marked as measured. They may be lapped as a final finishing pass.
>> A lot of the variation in calipers comes from a loose fit of the
>> head to the bar, allowing it to tilt when pressure is put on the jaw
>> tips. If you have it pretty snug, you can keep this from being a
>> problem. And the closer to the bar you do the measurement, the less the
>> problem from a tilting head and jaws.
>
> Good tip. But fortunately the jaws on my calipers are snug.
If the image which I saw was yours, the traveling body is long
enough to minimize the jaw tilt which I was worrying about. I've seen
others which were much shorter (including the digitals which I have)
which make the gib setting more critical.
DoN. Nichols
>
> Hi Ed,
>
>> The reason is thermal expansion of the caliper. If you're pushing for the
>> best accuracy, and particularly if the room is cold, holding it in your hand
>> for a few minutes can make a difference. If you warm a three-inch-long piece
>> of steel by 30 deg. F, from your 40 deg. shop to 70 deg., it will expand by
>> almost 0.001". If the piece you're measuring is the same temperature as the
>> caliper, and if they're both steel, the actual temperature matters little,
>> but a difference in their temperatures, if it's more than a few degrees, can
>> result in inaccurate measurement.
>
> Right. I rarely wear gloves except for particularly dirty jobs. I prefer
> barrier cream.
Gloves are a *very* bad idea when operating machine tools, but
when measuring with micrometers or calipers are a good way to reduce
thermal transfer resulting in errors. (Of course -- the thermal errors
can be in the workpiece as well -- if you have just turned a steel
workpiece in the lathe, it will measure significantly larger than after
it has cooled for a couple of hours -- which can lead to fit problems
when fitting a backplate to a lathe chuck for example. Finish the rough
turning, go eat dinner and read a book or watch a TV show, and come back
later to measure it and take the final passes to get the desired size.
[ ... ]
>> But it depends on whether you're measuring *relative* dimensions or
>> *absolute* dimensions. If it's the former, you don't need a well-calibrated
>> gage. If it's the latter, you'll never know for sure how accurately your
>> gages measure unless you check them from time to time against a gage block
>> (or a stack) that's somewhere in the middle of the gage's range, and another
>> one near its largest opening. For a 3" - 4" mike, I'd want at least a 3"
>> block (or a stack to make that dimension) and a 4" block or stack. I'd like
>> to have a 1/2" block, too, to wring with the 3" block to test the midrange.
>
> What's a stack?
Multiple gauge blocks wrung together to make up a precise size.
A good set of blocks will have something like:
1.000"
2.000"
3.000"
4.000"
0.100"
0.200"
0.300"
0.400"
0.500"
0.600"
0.700"
0.800"
0.900"
0.110"
0.120"
0.130"
0.140"
0.150"
0.160"
0.170"
0.180"
0.190"
0.101"
0.102"
0.103"
0.104"
0.105"
0.106"
0.107"
0.108"
0.109"
0.1001"
0.1002"
0.1003"
0.1004"
0.1005"
0.1006"
0.1007"
0.1008"
0.1009"
And -- in a separate box, I also have:
0.10000"
0.10001"
0.10002"
0.10003"
0.10004"
0.10005"
0.10006"
0.10007"
0.10008"
0.10009"
"Wringing" two blocks together involves cleaning the surfaces
about to mate, crossing them at right angles, and rotating them until
the two blocks are parallel. A good block will cling to its neighbor,
and you can progress to a fairly long stack of blocks which can be
lifted by the top-most block and they will all stay together. (The
cheap Chinese sets, supposedly accurate to 0.000050", tend to not have a
good enough finish to bond together properly -- at least my set does
not. But I have a good B&S/"Jo"hansen block set which does this well,
and which comes with a certificate of calibration listing the actual
size of each block.
The combination of multiple blocks will allow you to build up
almost any dimension as long as it is shorter than the multiplicity of
ones which are just above '0.10000"' (they don't often make thinner ones
because they can warp more easily). As a result, some sine plates have
an extra section ground 0.20000" or 0.30000" below the main surface, so
you can build a stack of blocks which will achieve the small angle you
want to make.
>> If any of this is unclear, ask, and I or someone else will explain. FWIW,
>> most home-shop work doesn't require accurate measurement of absolute
>> dimensions. Usually we're trying to make two things fit together, and what
>> you need to know for that is their *relative* dimensions. If that's the
>> case, forget the gage blocks for now.
>
> In this case, I think it's the absolute dimensions. The engine's piston
> wobbles noticeably in the bore. It's an aluminium piston in a cast iron
> bore, so I suspect there could be much more wear on the piston than the
> bore (the bore looks good, without a ridge at the top).
Then again, softer materials tend to embed abrasive particles so
they wear away the harder material against which they are run. This is
the principle of a "lap".
At a guess, I would expect the measurement down into the bore at
right angles to the crankshaft axis to measure larger than the dimension
at that height parallel to the axis. (The forces from the crankshaft
and connecting rod would tend to apply more force along that cross axis.)
> So I want to
> know how well a brand new piston will fit in the existing bore, before I
> buy one.
Understood -- and also to discover whether you need to bore
0.100" oversized to fit the next size piston up.
Note that the skirts of some pistons are designed to move with
temperature to fit well at operating temperature. The solid part near
the top remains more constant in dimensions than the bottom of the
skirts.
Good Luck,
Christopher Tidy
> It would really help to have a good engine man chime in here. There are
> several around; try a new thread with a title like "Measuring engine bores"
Might do that, but I'm going to try taking the measurements first. I
have the urge to get into the workshop :-).
> Many of us can help you with handling gages but engine cylinders have other
> issues, like taper and ovality, etc. If it's an old engine, the pistons may
> be cylindrical. If it's a newer one, they're probably elliptical ("oval").
> And they're tapered along their lengths, too.
Interesting. Why are pistons intentionally made elliptical? I've heard
of pistons where the top land is of a smaller diameter (apparently it
reduces wear because the lubrication is poorer at the top), but I've not
heard of an elliptical piston. I'm surprised it doesn't increase the
wear on the sides due to the reduced surface area, and I'd have thought
it could allow the piston to vibrate in an angular sense about the
gudgeon pin. Any more information, Ed?
Best wishes,
Chris
Christopher Tidy
> I'd avoid them unless you really need them. Full disclosure -- Mitutoyo was
> my client for many years, and I wrote all of their articles during those
> years and some of their instructional materials. They're quite honest about
> what they claim. Just be aware that things like interchangeable-anvil
> micrometers sacrifice some assurance for the sake of convenience. They can
> be as accurate as any mike; you just have to be a bit more careful.
Thanks. I'll remember you're the guy to ask about Mitutoyo gear! The
interchangeable anvil micrometer I saw advertised doesn't include all
the calibration standards Don mentioned, so I'll avoid it.
Best wishes,
Chris
Christopher Tidy
> Don't go by the piston, go by the bore. It sounds like your engine is
> in fairly good shape, but check for elongation perpendicular to the
> crankshaft. Bores become oval from wear, where the connecting rod
> pushes 'em up and drags 'em down the opposite sides. If you bore it
> out, you'll need all new pistons by default.
The piston is wobblier perpendicular to the crankshaft that it is in the
parallel direction, which suggests to me that there's some significant
wear. Also, the compression isn't great. But the bore itself looks fine,
hence the bore gauge purchase.
Chris
Christopher Tidy
> "Wringing" two blocks together involves cleaning the surfaces
> about to mate, crossing them at right angles, and rotating them until
> the two blocks are parallel. A good block will cling to its neighbor,
> and you can progress to a fairly long stack of blocks which can be
> lifted by the top-most block and they will all stay together. (The
> cheap Chinese sets, supposedly accurate to 0.000050", tend to not have a
> good enough finish to bond together properly -- at least my set does
> not. But I have a good B&S/"Jo"hansen block set which does this well,
> and which comes with a certificate of calibration listing the actual
> size of each block.
Thanks for the explanation. Won't a stack be somewhat less accurate than
a single block manufactured to the right size? Or is the tolerance on
gauge blocks so small that it doesn't matter?
> The combination of multiple blocks will allow you to build up
> almost any dimension as long as it is shorter than the multiplicity of
> ones which are just above '0.10000"' (they don't often make thinner ones
> because they can warp more easily). As a result, some sine plates have
> an extra section ground 0.20000" or 0.30000" below the main surface, so
> you can build a stack of blocks which will achieve the small angle you
> want to make.
Just briefly, what's a sine plate used for?
> Understood -- and also to discover whether you need to bore
> 0.100" oversized to fit the next size piston up.
I've not seen +0.100" oversize pistons. For this engine the options are
+0.020" or +0.040".
> Note that the skirts of some pistons are designed to move with
> temperature to fit well at operating temperature. The solid part near
> the top remains more constant in dimensions than the bottom of the
> skirts.
I haven't removed the piston yet. The wobble I can see is adjacent to
the solid part, at the top.
Best wishes,
Chris
Christopher Tidy
> The Starrett 123 series calipers have adjustable scales and are as
> good as verniers get. The 50 division scales with the vernier and main
> scale in the same plane make them very easy to read. I can read them
> without glasses in a pinch, though I don't recommend it.
>
> http://www.starrett.com/download/62_123_caliper.pdf
Thanks. I'll bear them in mind if I ever need another set.
Chris
Christopher Tidy
> Well ... the ones which I have are quite old. These days, most
> people buy either dial or digital calipers. Dial is nice if you can
> keep chips out of the rack, but Digital (if good) is far more
> convenient, including the ability to zero at any point (such as on what
> you are fitting to) and then measure how much more needs to be removed
> to reach the proper dimensions. Also, measuring the distance between
> the centers of two holes of identical size by zeroing it while measuring
> the ID of one hole, then shift to measuring the distance between two
> opposite sides of the two holes in question. This subtracts the
> diameter of one hole from the final measurement which is the sum of the
> center distance and half of each hole's diameter, giving the actual
> center-to-center distance.
I chose the vernier calipers because I wanted both imperial and metric
on the same tool, and I wanted them to last. I hate replacing tools.
> O.K. But it misses three-lobed bores, which can be in a shape
> which actually measures the same between any two opposite points, but
> still is potentially seriously out of round. (This is more likely to be
> formed by certain machining operations -- including centerless grinding
> for OD shapes, and in a used automobile engine, the ovality is more
> likely, and for that your bore gauge is probably quite adequate.
I've not seen a three-point bore gauge intended for measuring engine
cylinders. All the gauges I found on the market were two-point with a
centring device. I picked the one I mentioned because it's complete with
all the anvils, locking nuts and the spanner, and is in good condition.
There were more accurate gauges available (measured down to 0.0001"
instead of 0.0005"), but they were much more worn and had missing parts.
Also, I figured that for investigating wear as opposed to reboring,
having a 40 thousandths range on the dial might be more useful than a 10
thousandths range. If people are interested, I can take a picture of the
bore gauge and calipers.
Best wishes,
Chris
Christopher Tidy
> Well ... this one is for setting an air gauge, so it has to be
> pretty close to the diameter of the probe. An air gauge is hollow with
> both ends sealed, and air feed in through one end, with a hole of a
> specific size (or maybe more holes? -- I haven't actually seen one --
> just read about them in books). The air goes out the hole, and flows
> out between the probe walls and the ID of the bore being checked, and
> the pressure built up in the interior of the gauge is measured and is a
> very good indication of the difference between the OD of the gauge and
> the ID of the bore -- over a fairly narrow range. So -- this ring was
> for setting something quite close to 16.8349 mm --- maybe 16.8350 mm is
> the actual target size. Anyway -- the rings are made too small,
> hardened, and then ground to a diameter near what is needed, and then
> marked as measured. They may be lapped as a final finishing pass.
Thanks. That's interesting. I wasn't sure what an air gauge was.
>>> A lot of the variation in calipers comes from a loose fit of the
>>>head to the bar, allowing it to tilt when pressure is put on the jaw
>>>tips. If you have it pretty snug, you can keep this from being a
>>>problem. And the closer to the bar you do the measurement, the less the
>>>problem from a tilting head and jaws.
>>
>>Good tip. But fortunately the jaws on my calipers are snug.
>
>
> If the image which I saw was yours, the traveling body is long
> enough to minimize the jaw tilt which I was worrying about. I've seen
> others which were much shorter (including the digitals which I have)
> which make the gib setting more critical.
That picture isn't actually my set of calipers, but is a very similar
set. Same brand, but shorter and perhaps made a little earlier.
Chris
Christopher Tidy
> Then your 2 thou is likely more hopeful. Get a pair of dials, boy!
> They're less easy to misread. I forgot you were talking about
> verniers. Got a good magnifying glass? ;)
Looks like I might regret buying verniers as I get older :-).
Chris
Ed Huntress
> Ed Huntress wrote:
>
>> It would really help to have a good engine man chime in here. There are
>> several around; try a new thread with a title like "Measuring engine
>> bores"
>
> Might do that, but I'm going to try taking the measurements first. I have
> the urge to get into the workshop :-).
>
>> Many of us can help you with handling gages but engine cylinders have
>> other issues, like taper and ovality, etc. If it's an old engine, the
>> pistons may be cylindrical. If it's a newer one, they're probably
>> elliptical ("oval"). And they're tapered along their lengths, too.
>
> Interesting. Why are pistons intentionally made elliptical?
Because the thicker sections, which are the boss areas for the wrist pins
(piston pins) expand with much more force than the thin sections. So the
pistons have a smaller diameter across the boss area.
> I've heard of pistons where the top land is of a smaller diameter
> (apparently it reduces wear because the lubrication is poorer at the top),
> but I've not heard of an elliptical piston.
Almost all production automobile pistons made today are elliptical. In fact,
they're often elliptical with the major axis in one direction at the top of
the piston, and in the other direction at the bottom. The bottom ellipse is
for better sealing, to meet emission requirements. It has to do with
differential friction and heating between the neutral axis, which is
parallel to the crankshaft, versus the other axis.
When I was at Wasino we had some drawings from Ford that actually had three
different ellipses along their length, from top to bottom, and they had to
blend into each other.
> I'm surprised it doesn't increase the wear on the sides due to the reduced
> surface area, and I'd have thought it could allow the piston to vibrate in
> an angular sense about the gudgeon pin. Any more information, Ed?
If you don't get an expert to chime in here, I'll see what I can dig up for
you. There is one guy who stops in here from time to time who is an engineer
for one of the world's top piston manufacturers; you won't get any better
info than that from him.
If you're eager to search on it yourself, try both "elliptical piston" and
"oval piston." They're often, incorrectly, called "oval pistons" in the
trade.
--
Ed Huntress
Ed Huntress
You'll get in the habit of wearing a swing-away loupe on your glasses when
you're reading gages, after you pass a certain age. <g>
You original reasons for buying a vernier caliper were good ones. Dial
calipers are no more accurate; they're just easier to read. The same is true
for the digitals.
I use my TESA vernier caliper when I'm not in a hurry and I want to keep up
the skill of reading them. But if I'm doing a lot of measuring, I'll pull
out one of my digitals. I've never owned a dial caliper.
--
Ed Huntress
Larry Jaques
<cdt22...@cantabgold.net> scrawled the following:
>Larry Jaques wrote:
We all do.
--
"Politics is the art of looking for trouble, finding it whether it
exists or not, diagnosing it incorrectly, and applying the wrong
remedy." -- Ernest Benn
DoN. Nichols
> DoN. Nichols wrote:
>
>> Well ... the ones which I have are quite old. These days, most
>> people buy either dial or digital calipers. Dial is nice if you can
>> keep chips out of the rack, but Digital (if good) is far more
>> convenient, including the ability to zero at any point (such as on what
>> you are fitting to) and then measure how much more needs to be removed
>> to reach the proper dimensions. Also, measuring the distance between
>> the centers of two holes of identical size by zeroing it while measuring
>> the ID of one hole, then shift to measuring the distance between two
>> opposite sides of the two holes in question. This subtracts the
>> diameter of one hole from the final measurement which is the sum of the
>> center distance and half of each hole's diameter, giving the actual
>> center-to-center distance.
>
> I chose the vernier calipers because I wanted both imperial and metric
> on the same tool, and I wanted them to last. I hate replacing tools.
O.K. There *are* dial calipers with dual scales, and two
separately geared pointers -- but you still have the problem of the
pinion skipping a tooth shifting the zero point either from a shock or
from chips getting into the rack gear in the bar.
Digital are nice for switching between metric and imperial, but
have the problem of keeping fresh cells around to power them, as some
will kill their cells in six months, and I think a year and a half is
pretty long for one to last.
And -- they are more easily damaged -- things like the liquid
crystal display are rather fragile -- behind the clear plastic window.
>> O.K. But it misses three-lobed bores, which can be in a shape
>> which actually measures the same between any two opposite points, but
>> still is potentially seriously out of round. (This is more likely to be
>> formed by certain machining operations -- including centerless grinding
>> for OD shapes, and in a used automobile engine, the ovality is more
>> likely, and for that your bore gauge is probably quite adequate.
>
> I've not seen a three-point bore gauge intended for measuring engine
> cylinders.
For checking *wear* in the cylinders (as you want) what you got
is better -- quicker to use and all. And it lets you measure parallel
to and at right angles to the crankshaft to check for oval wear.
For checking a reboring job, I would prefer the three-point,
since the boring had *better* be cylindrical. :-) But since this is
going to be in an automobile, and you have dissimilar metals between the
cylinder and the piston, you have differing thermal coeifficents of
expansion, and the temperature range between sitting outside on a really
cold day (maybe -40 in some areas, where special lubricants and coolants
are also needed, and a bit over 212 F (100 C) would really require a
good starting clearance, or it will seize at one extreme or the other.
> All the gauges I found on the market were two-point with a
> centring device. I picked the one I mentioned because it's complete with
> all the anvils, locking nuts and the spanner, and is in good condition.
> There were more accurate gauges available (measured down to 0.0001"
> instead of 0.0005"), but they were much more worn and had missing parts.
Any idea what the required starting clearance is between the
pistons and the bores? *That* would settle how accurate you need the
measurement to be for checking a rebore.
> Also, I figured that for investigating wear as opposed to reboring,
> having a 40 thousandths range on the dial might be more useful than a 10
> thousandths range. If people are interested, I can take a picture of the
> bore gauge and calipers.
If so -- post it to the dropbox, or on a private web site and
post the URL to find it.
Enjoy,
DoN. Nichols
>
>> "Wringing" two blocks together involves cleaning the surfaces
>> about to mate, crossing them at right angles, and rotating them until
>> the two blocks are parallel. A good block will cling to its neighbor,
>> and you can progress to a fairly long stack of blocks which can be
>> lifted by the top-most block and they will all stay together. (The
>> cheap Chinese sets, supposedly accurate to 0.000050", tend to not have a
>> good enough finish to bond together properly -- at least my set does
>> not. But I have a good B&S/"Jo"hansen block set which does this well,
>> and which comes with a certificate of calibration listing the actual
>> size of each block.
>
> Thanks for the explanation. Won't a stack be somewhat less accurate than
> a single block manufactured to the right size? Or is the tolerance on
> gauge blocks so small that it doesn't matter?
As someone else mentioned, the joint between a properly wrung
pair of blocks is on the order of two microinches (0.000002"), so only
with the highest grade blocks do you need to take account of the
interface thickness.
>> The combination of multiple blocks will allow you to build up
>> almost any dimension as long as it is shorter than the multiplicity of
>> ones which are just above '0.10000"' (they don't often make thinner ones
>> because they can warp more easily). As a result, some sine plates have
>> an extra section ground 0.20000" or 0.30000" below the main surface, so
>> you can build a stack of blocks which will achieve the small angle you
>> want to make.
>
> Just briefly, what's a sine plate used for?
It is a plate (or a bar -- the more common ones are sine bars)
which has a pair of cylindrical surfaces at each end, separated by a
precise center distance. The most common is 5.000", though I have one
at 2.500", and have seen some offered at 10.000").
For a sine bar, there is no captive base plate, so you set it on
a surface plate, and with both cylinders in contact, its top is parallel
to the surface plate.
Now -- let's say you want a precise fifteen degrees. O.K. Look
up the sine of 15 degrees (0.258819045) and multiply by the length of
the bar (5.000"), so you get 1.294095226". Wring a stack of gauge
blocks to get 1.2941" and you will get very close to 15 degrees.
Calculating back from that, I get 15.000056 degrees +/- a bit given
the accuracy of the gauge block set. I stopped the blocks at 1.2941"
assuming a cheap Chinese set with +/- 0.000050 accuracy. You can get
greater precision with the more expensive and accurate sets.
But -- to get that 1.2941" -- we need to build a stack. Let's
see -- start with a 0.1001" block, add a 0.1040" block, then a 0.1900"
block, so we are up to 0.3941" and need only 0.9000" to make up our
total size. When calculating/building a stack, always start with the
last decimal place and work backwards. Here, for example, if you had
started at the big end, you would have picked up a 1.0000" block, and
when you got the lesser digits you would have discovered that the total
was too long.
Now -- a sine plate is like a sine bar, except that it is wider
(the sine bar may be 1" wide) and is captive to a base, so you can build
the angle and lock it in and then carry it to the magnetic chuck of a
surface grinder to grind the desired angle on something mounted to the
top (angled) plate.
To see a sine bar -- here is one on MSC's site:
<http://www1.mscdirect.com/CGI/NNSRIT?PMPXNO=17520867&PMT4NO=80436460>
Or MSC part number 85005502 in case the url above turns out to
be a temporary one built by my search.
The toe on one end is to keep the workpiece from sliding off the
end while measuring.
An example sine plate (much larger and *much* more expensive
than the one which I have) is MSC item 08020216
<http://www1.mscdirect.com/CGI/NNSRIT?PMPXNO=1778994&PMT4NO=80436812>
You can see the near roll under the top plate in the image, and a raised
block for the zero point, which is a precise height above the base
plate. So you can either build blocks above the raised block, or above
the base plate - whichever lets you reach your desired height more
easily.
Note that there are double sine plates -- a second one hinged at
right angles to the first to allow compound angles.
I've used mine to make Acme threading tools to fit a boring bar.
I first used the smallest sine bar that I have (2.5" long between
centers) to machine a 14 degree angle plate (half of an Acme) guide to
hold a HSS tool bit at that angle in a small vise, then mounted the vise
on the sine plate, and set the gauge blocks under it for the desired
side relief on one side of the bit. This went onto the surface grinder,
and was used to make the desired angle on that side. Then I reversed
the bit and angle plate, and built a different stack of gauge blocks to
provide the different relief angle on the other side of the bit. (This
was calculated based on the helix angle of the thread. After this was
complete, I put the bit upright in the vise and set a nose relief angle
to grind while I ground the nose back for the proper nose width for that
particular Acme thread pitch.
>> Understood -- and also to discover whether you need to bore
>> 0.100" oversized to fit the next size piston up.
>
> I've not seen +0.100" oversize pistons. For this engine the options are
> +0.020" or +0.040".
Sorry -- yes, 0.100" is a bit large for automotive pistons at
least. :-)
>> Note that the skirts of some pistons are designed to move with
>> temperature to fit well at operating temperature. The solid part near
>> the top remains more constant in dimensions than the bottom of the
>> skirts.
>
> I haven't removed the piston yet. The wobble I can see is adjacent to
> the solid part, at the top.
O.K.
Jim Wilkins
> On 2010-02-23, Christopher Tidy <cdt22NOS...@cantabgold.net> wrote:
> > DoN. Nichols wrote:
> ...
>
> > Just briefly, what's a sine plate used for?
> Now -- let's say you want a precise fifteen degrees. O.K. Look
> up the sine of 15 degrees (0.258819045) and multiply by the length of
> DoN.
Nice description.
A sine bar can also be used to measure an unknown angle, such as the
taper on a shaft. Lay the taper on a flat surface and clamp the sine
bar on top of it, rolls up. A ground toolmakers vise works well for
this. Measure the difference in the heights of the rolls.
If you have only a short-range 0.0001" test indicator and can't
measure that much distance, tweak an adjustable parallel to make up
the height difference, so the indicator reads the same on both, then
measure the parallel with a micrometer.
jsw
DoN. Nichols
> On Feb 23, 5:26�pm, "DoN. Nichols" <dnich...@d-and-d.com> wrote:
>> On 2010-02-23, Christopher Tidy <cdt22NOS...@cantabgold.net> wrote:
>> > DoN. Nichols wrote:
>> ...
>>
>> > Just briefly, what's a sine plate used for?
>> ...
>> � � � � Now -- let's say you want a precise fifteen degrees. �O.K. �Look
>> up the sine of 15 degrees (0.258819045) and multiply by the length of
>> the bar (5.000"), so you get 1.294095226". �....>
> Nice description.
Thanks!
> A sine bar can also be used to measure an unknown angle, such as the
> taper on a shaft. Lay the taper on a flat surface and clamp the sine
> bar on top of it, rolls up. A ground toolmakers vise works well for
> this. Measure the difference in the heights of the rolls.
>
> If you have only a short-range 0.0001" test indicator and can't
> measure that much distance, tweak an adjustable parallel to make up
> the height difference, so the indicator reads the same on both, then
> measure the parallel with a micrometer.
Aha! A nice trick which I had not thought of.
I've got one much smaller sine bar -- 1.000" between rolls --
which is mounted on a micrometer, with a bar of the same length as the
anvil, so it can be used to measure the sine of existing angles within a
limited range of diameters.
Of course -- you lose the multiplier effect of a 5" or 10" sine
bar, so you don't get the accuracy that you can with one of those, but
it is still quite good and a lot more convenient for quick readings.
Enjoy,
Christopher Tidy
Hi Don,
Thanks for the description of the sine plate. Sorry I've taken a few
days to reply. Bad week!
Not what I was expecting! I had imagined a steel plate with a surface
shaped like a sine wave, sitting on a table (though what you'd use that
for, I don't know). Thanks for the explanation. Some day I'll probably
need one!
Best wishes,
Chris
Christopher Tidy
> For checking *wear* in the cylinders (as you want) what you got
> is better -- quicker to use and all. And it lets you measure parallel
> to and at right angles to the crankshaft to check for oval wear.
>
> For checking a reboring job, I would prefer the three-point,
> since the boring had *better* be cylindrical. :-) But since this is
> going to be in an automobile, and you have dissimilar metals between the
> cylinder and the piston, you have differing thermal coeifficents of
> expansion, and the temperature range between sitting outside on a really
> cold day (maybe -40 in some areas, where special lubricants and coolants
> are also needed, and a bit over 212 F (100 C) would really require a
> good starting clearance, or it will seize at one extreme or the other.
Actually, it's a big motor lawnmower (80 mm bore).
>> All the gauges I found on the market were two-point with a
>>centring device. I picked the one I mentioned because it's complete with
>>all the anvils, locking nuts and the spanner, and is in good condition.
>>There were more accurate gauges available (measured down to 0.0001"
>>instead of 0.0005"), but they were much more worn and had missing parts.
>
>
> Any idea what the required starting clearance is between the
> pistons and the bores? *That* would settle how accurate you need the
> measurement to be for checking a rebore.
I've found that figure in the manual. I'll post it along with my
measurements in a new thread shortly.
>>Also, I figured that for investigating wear as opposed to reboring,
>>having a 40 thousandths range on the dial might be more useful than a 10
>>thousandths range. If people are interested, I can take a picture of the
>>bore gauge and calipers.
>
>
> If so -- post it to the dropbox, or on a private web site and
> post the URL to find it.
Here's a picture of my actual calipers:
http://www.mythic-beasts.com/~cdt22/etalon_vernier_calipers.jpg
Best wishes,
Chris
Christopher Tidy
>>Interesting. Why are pistons intentionally made elliptical?
>
>
> Because the thicker sections, which are the boss areas for the wrist pins
> (piston pins) expand with much more force than the thin sections. So the
> pistons have a smaller diameter across the boss area.
Ah, I think I misunderstoood you. I thought you meant a piston shaped
like a beer barrel. Instead you mean a piston which is slightly
elliptical when viewed from the top or bottom?
>>I've heard of pistons where the top land is of a smaller diameter
>>(apparently it reduces wear because the lubrication is poorer at the top),
>>but I've not heard of an elliptical piston.
>
>
> Almost all production automobile pistons made today are elliptical. In fact,
> they're often elliptical with the major axis in one direction at the top of
> the piston, and in the other direction at the bottom. The bottom ellipse is
> for better sealing, to meet emission requirements. It has to do with
> differential friction and heating between the neutral axis, which is
> parallel to the crankshaft, versus the other axis.
>
> When I was at Wasino we had some drawings from Ford that actually had three
> different ellipses along their length, from top to bottom, and they had to
> blend into each other.
>
>
>>I'm surprised it doesn't increase the wear on the sides due to the reduced
>>surface area, and I'd have thought it could allow the piston to vibrate in
>>an angular sense about the gudgeon pin. Any more information, Ed?
>
>
> If you don't get an expert to chime in here, I'll see what I can dig up for
> you. There is one guy who stops in here from time to time who is an engineer
> for one of the world's top piston manufacturers; you won't get any better
> info than that from him.
The thought I had about vibration only applies to a piston shaped like a
beer barrel. And if the piston expands to become a near-perfect cylinder
when it's heated, I can see why wear wouldn't be an issue either. Thanks!
Best wishes,
Chris
Ed Huntress
> Ed Huntress wrote:
>
>>>Interesting. Why are pistons intentionally made elliptical?
>>
>>
>> Because the thicker sections, which are the boss areas for the wrist pins
>> (piston pins) expand with much more force than the thin sections. So the
>> pistons have a smaller diameter across the boss area.
>
> Ah, I think I misunderstoood you. I thought you meant a piston shaped like
> a beer barrel. Instead you mean a piston which is slightly elliptical when
> viewed from the top or bottom?
Yes. I wish I still had the 3D versions of those Ford programs that I
produced in Rhino, for our promotional material on cutting elliptical
pistons. I applied multipliers in Excel to the values in the CAD drawing
until you could actually see the shapes with the naked eye.
I don't want to confuse things, but they were shaped a lot lie beer barrels.
<g> As I mentioned, there were three different ellipses from top to bottom,
with major axes arranged differently, and the appearance from some angles
was that they were widest in the middle.
>
>>>I've heard of pistons where the top land is of a smaller diameter
>>>(apparently it reduces wear because the lubrication is poorer at the
>>>top), but I've not heard of an elliptical piston.
>>
>>
>> Almost all production automobile pistons made today are elliptical. In
>> fact, they're often elliptical with the major axis in one direction at
>> the top of the piston, and in the other direction at the bottom. The
>> bottom ellipse is for better sealing, to meet emission requirements. It
>> has to do with differential friction and heating between the neutral
>> axis, which is parallel to the crankshaft, versus the other axis.
>>
>> When I was at Wasino we had some drawings from Ford that actually had
>> three different ellipses along their length, from top to bottom, and they
>> had to blend into each other.
>>
>>
>>>I'm surprised it doesn't increase the wear on the sides due to the
>>>reduced surface area, and I'd have thought it could allow the piston to
>>>vibrate in an angular sense about the gudgeon pin. Any more information,
>>>Ed?
>>
>>
>> If you don't get an expert to chime in here, I'll see what I can dig up
>> for you. There is one guy who stops in here from time to time who is an
>> engineer for one of the world's top piston manufacturers; you won't get
>> any better info than that from him.
>
> The thought I had about vibration only applies to a piston shaped like a
> beer barrel. And if the piston expands to become a near-perfect cylinder
> when it's heated, I can see why wear wouldn't be an issue either. Thanks!
Yes, that's more or less what happens.
--
Ed Huntress
DoN. Nichols
> DoN. Nichols wrote:
>
> Hi Don,
>
> Thanks for the description of the sine plate.
You're welcome.
> Sorry I've taken a few
> days to reply. Bad week!
A lot of that going around -- just from the weather. I hope
that is all that yours was.
>>>Just briefly, what's a sine plate used for?
[ ... description of appearance and use snipped ... ]
> Not what I was expecting! I had imagined a steel plate with a surface
> shaped like a sine wave, sitting on a table (though what you'd use that
> for, I don't know).
Actually -- there is something like that -- but a pair of them
go in a milling vise to support the workpiece. They are less prone to
topple over when the vise is loosened than some standard parallels, and
they can also be positioned to support workpieces with gaps near the
edges in some places and nearer the center in others.
I think that they are called "wave parallels". I've never owned
any, or had a chance to handle some belonging to someone else, but they
strike me as useful for production runs.
> Thanks for the explanation. Some day I'll probably
> need one!
If you see one used cheap -- go for it even before you need it.
I've gotten a lot of my tools that way against future needs.
Enjoy,
Christopher Tidy
>>Ah, I think I misunderstoood you. I thought you meant a piston shaped like
>>a beer barrel. Instead you mean a piston which is slightly elliptical when
>>viewed from the top or bottom?
>
>
> Yes. I wish I still had the 3D versions of those Ford programs that I
> produced in Rhino, for our promotional material on cutting elliptical
> pistons. I applied multipliers in Excel to the values in the CAD drawing
> until you could actually see the shapes with the naked eye.
>
> I don't want to confuse things, but they were shaped a lot lie beer barrels.
> <g> As I mentioned, there were three different ellipses from top to bottom,
> with major axes arranged differently, and the appearance from some angles
> was that they were widest in the middle.
Got it. I can see the shape now. That's complicated!
>>The thought I had about vibration only applies to a piston shaped like a
>>beer barrel. And if the piston expands to become a near-perfect cylinder
>>when it's heated, I can see why wear wouldn't be an issue either. Thanks!
>
>
> Yes, that's more or less what happens.
At least the concept sounds simpler :-)
Chris
Christopher Tidy
>>Interesting. Why are pistons intentionally made elliptical?
>
>
> Because the thicker sections, which are the boss areas for the wrist pins
> (piston pins) expand with much more force than the thin sections. So the
> pistons have a smaller diameter across the boss area.
By the way, Ed, how are elliptical pistons machined? Are they made using
a CNC mill, or a special lathe on which the tool can move in phase with
the rotation of the workpiece? Just curious...
Chris
Ed Huntress
A darned good question. I wish the piston man would step in here (Anthony?).
He's a world-class expert on this subject.
In high-volume production, they've traditionally been turned on special
lathes with cam-operated cross slides. That requires a cam for each piston
profile. Takisawa makes, or made, a programmable machine that, IIRC, uses
solenoid-actuated slides. It's slow. There also have been some piezoelectric
actuators. And that was all ten years ago. There may be something new.
I was involved with these things due to a joint project we had going at
Wasino lathes, where I worked in those days. We were using a
magnetostrictive actuator, made of the material used in submarine sonar. It
worked great but programming the hysteresis out of it was a nightmare. The
project was dropped about the time I left.
Anyway, the challenge has been to come up with a programmable system that
provides adequate thrust at high turning speeds. They're doing it, I think;
I just haven't kept up.
--
Ed Huntress
Christopher Tidy
> A darned good question. I wish the piston man would step in here (Anthony?).
> He's a world-class expert on this subject.
>
> In high-volume production, they've traditionally been turned on special
> lathes with cam-operated cross slides. That requires a cam for each piston
> profile. Takisawa makes, or made, a programmable machine that, IIRC, uses
> solenoid-actuated slides. It's slow. There also have been some piezoelectric
> actuators. And that was all ten years ago. There may be something new.
>
> I was involved with these things due to a joint project we had going at
> Wasino lathes, where I worked in those days. We were using a
> magnetostrictive actuator, made of the material used in submarine sonar. It
> worked great but programming the hysteresis out of it was a nightmare. The
> project was dropped about the time I left.
>
> Anyway, the challenge has been to come up with a programmable system that
> provides adequate thrust at high turning speeds. They're doing it, I think;
> I just haven't kept up.
Thanks for the information. Sorry I've been slow to reply; I've been
away. I worked myself as part of a team developing electrostatic
actuators a few years back. The actuators consisted of a sheet of a
rubbery material with conductive coatings on each side. When subjected
to a high DC voltage, the conductive coatings could be made to attract
or repel, squeezing or stretching the material (the actuation direction
was perpendicular to the electric field, as this gave a greater
movement). One idea was to use these actuators to move a car wing
mirror, although I'm unsure if a prototype was made as I'd left the
laboratory by then. But the actuators we had at that time were
definitely too flexible to be used for moving a lathe tool.
Best wishes,
Chris
Ed Huntress
It requires a lot of force. The material we used, Terfenol-D, is one of the
few that can deliver the force with good speed (several thousand cycles per
second with high force; up to 20 kHz at lower force levels, IIRC).
Piezoelectrics, which are the other option, are used in stacks to get the
required travel, but they're a little fragile for the application:
http://en.wikipedia.org/wiki/Magnetostriction
Terfenol is the material used in advanced sonar detectors. We were first
looking at it as a counter-vibratory attachment, but my boss got the idea
that it would work for turning elliptical pistons. We had a demo machine at
IMTS 2000, I think, and it attracted a lot of attention. We could get it to
work extremely well and with high accuracy and repeatability. But if you
make a small change in the program, magnetic hysteresis would complicate the
programming, making the whole thing problematic.
--
Ed Huntress
Bill McKee
"Ed Huntress" <hunt...@optonline.net> wrote in message
news:4b99a83e$0$22515$607e...@cv.net...
I worked on a product using the Polaroid Piso distance sensors. 1980's.
Problem was the ringing in the piso unit required a separate sender and a
receiver for up close sensing, which is what I was trying to do. Figured
out was not a big enough market to continue on the idea. As to fast, heavy
duty, accurate movers, Voice coils work great. Use a linear transformer as
the sensor. Was the way we build disk drives in the 1970's early 80's. The
magnet on a CDC 200 megabyte drive probably weighs 25-30 pounds total.
Ed Huntress
"Bill McKee" <bmckee...@ix.netcom.com> wrote in message
news:iuWdnXlu0JJoNgTW...@earthlink.com...
There were some reasons that solenoid-type actuators weren't ideal. I think
it was the response rate with the weight of the actuator required for the
job, but it's been too long for me to trust my memory.
I see that Takisawa is still making their line of oval-piston machines (TPS
Series). Maybe there's something on their website about the actuator:
http://www.takisawa.co.jp/-e/index-e02.htm
--
Ed Huntress
Don Foreman
A voice coil actuator is not a solenoid in the usual sense. Voice
coil actuators can exhibit fairly linear force vs current over a fair
range of motion, while solenoid force is a highly nonlinear function
of current and distance from closed. They tend to "snap shut" when
actuated, are best-suited for two-position situations like valve
actuators. A voice coil actuator requires a magnet while a solenoid
does not.
Ed Huntress
"Don Foreman" <dfor...@NOSPAMgoldengate.net> wrote in message
news:nlgjp5to7rlcc3tbs...@4ax.com...
Right, I forgot about how they work. But rest assured that the piston
manufacturers have tried about everything, and that combination of
programmability, adequate force, and speeds high enough for productive
turning of aluminum pistons has been a son of a gun to accomplish. All sorts
of electromagnetic devices have been tried.
--
Ed Huntress
Bill McKee
Voice coils and solenoids are totally different beasts.
Ed Huntress
Right. Don reminded me of that. But as I said to him, the biggest car
manufacturers and lathe builders have all tried everything they can think
of. We're talking about a manufacturing issue that's worth many tens of
millions, if not hundreds of millions, of dollars. There are an awful lot of
pistons made and, until the EPA relented, even Homelite was exploring our
machines for making pistons for string-trimmer engines. All of the car
manufacturers, all over the world, are caught up in it. They want
programmable machines.
If Takisawa is still selling their TPS machines, it seems likely that no
one has succeeded in building anything better. And they all have some
engineer or another who knows how a voice coil works. <g>
--
Ed Huntress
Joseph Gwinn
"Ed Huntress" <hunt...@optonline.net> wrote:
See US Patents 5,085,109 and 5,313,694. The first uses a hydraulic servo
system, the second uses a servo motor of some kind. Lots of other details are
given. Referenced and referencing patents may also be instructive.
Joe Gwinn
Tim Wescott
Voice coils tend to be limp, long travel things. They're great for
machines that need to completely isolate one part from another, but
that's the opposite of what you want in a lathe.
You may be able to do it with a voice coil driving a lever, with a
flexure at the other end and the tool holder a lot closer to the flexure
than the coil. It'd be one involved set of work to get it working, and
to make it fast enough to keep up with the spindle.
It'd be a fun development project, for sure.
--
Tim Wescott
Control system and signal processing consulting
www.wescottdesign.com
Tim Wescott
>
> It would really help to have a good engine man chime in here. There are
> several around; try a new thread with a title like "Measuring engine bores"
>
> Many of us can help you with handling gages but engine cylinders have other
> issues, like taper and ovality, etc. If it's an old engine, the pistons may
> be cylindrical. If it's a newer one, they're probably elliptical ("oval").
> And they're tapered along their lengths, too.
That sounds like what I can reasonably expect to come off of _my_ lathe!
And not because that's what I wanted, either!
Don Foreman
<hunt...@optonline.net> wrote:
Right. Voice coil actuators are quick but they're not stiff or
high-force devices. That said, the shakers used to vibration-test
aerospace hardware to 20 G's and beyond were voice coil actuated.
Several feet in diameter, water cooled. Helluva ride!
Ed Huntress
"Joseph Gwinn" <joeg...@comcast.net> wrote in message
news:joegwinn-986820...@news.giganews.com...
There are many ideas and many patents for doing this. What's in short supply
is solutions that actually work.
--
Ed Huntress
Ed Huntress
"Tim Wescott" <t...@seemywebsite.now> wrote in message
news:16GdnQMS_sgy4QfW...@web-ster.com...
Maybe that's how they came up with the first elliptical piston. <g>
--
Ed Huntress
Joseph Gwinn
"Ed Huntress" <hunt...@optonline.net> wrote:
These are Takisawa's patents.
Joe Gwinn
Ed Huntress
Oh. <g>
--
Ed Huntress
Bill McKee
"Tim Wescott" <t...@seemywebsite.now> wrote in message
Voice coils are can also be short travel and very precise. I designed disk
drives and head actuators buy and large are voice coils. The positioning
track to track can be short or far and keeping the head on track is very
precise.
DoN. Nichols
>
> "Tim Wescott" <t...@seemywebsite.now> wrote in message
> news:16GdnQAS_sii4QfW...@web-ster.com...
[ ... ]
>> Voice coils tend to be limp, long travel things. They're great for
>> machines that need to completely isolate one part from another, but that's
>> the opposite of what you want in a lathe.
>>
>> You may be able to do it with a voice coil driving a lever, with a flexure
>> at the other end and the tool holder a lot closer to the flexure than the
>> coil. It'd be one involved set of work to get it working, and to make it
>> fast enough to keep up with the spindle.
[ ... ]
> Voice coils are can also be short travel and very precise. I designed disk
> drives and head actuators buy and large are voice coils. The positioning
> track to track can be short or far and keeping the head on track is very
> precise.
They are precise only because they get feedback from the heads
and "servo tracks" written on one surface (at least the ones in sealed
hard drives, though I think that the old 5MB removable platter ones used
an an optical sensor to determine position, because in those days every
surface was precious.
Bill McKee
"DoN. Nichols" <dnic...@d-and-d.com> wrote in message
news:slrnhpm5vv....@Katana.d-and-d.com...
They also used linear transformers. The CDC 2 mb cartrige disk was this
way.
Tim Wescott
just supply it with a constant current -- and it displays zero spring
rate, zero damping rate, just a constant force.
It's just what you want for a disk drive -- speedy, no hysteresis,
rigidity doesn't matter much because the whole assembly is mechanically
quiet, etc.
(take apart a hard drive today and you'll find a voice coil, with some
Really Strong rare earth magnets providing a field)
But this is quite the opposite of what you need on a lathe. To get the
rigidity from the voice coil itself you need an exceedingly high
bandwidth on your control loop, and to develop lots of force you need a
physically large coil with lots of inductance -- which is going to cause
all sorts of electrical difficulties.
Not having tried this I couldn't say for sure, but _having_ worked on
voice coil actuated control loops that shove a 30 pound gimbal around, I
can say that you're probably going to be at least two orders of
magnitude short of nirvana with just a voice coil, sensor, and 'the
usual' electronics.
Tim Wescott
> On 2010-03-12, Bill McKee <bmckee...@ix.netcom.com> wrote:
>> "Tim Wescott" <t...@seemywebsite.now> wrote in message
>> news:16GdnQAS_sii4QfW...@web-ster.com...
>
> [ ... ]
>
>>> Voice coils tend to be limp, long travel things. They're great for
>>> machines that need to completely isolate one part from another, but that's
>>> the opposite of what you want in a lathe.
>>>
>>> You may be able to do it with a voice coil driving a lever, with a flexure
>>> at the other end and the tool holder a lot closer to the flexure than the
>>> coil. It'd be one involved set of work to get it working, and to make it
>>> fast enough to keep up with the spindle.
>
> [ ... ]
>
>> Voice coils are can also be short travel and very precise. I designed disk
>> drives and head actuators buy and large are voice coils. The positioning
>> track to track can be short or far and keeping the head on track is very
>> precise.
>
> They are precise only because they get feedback from the heads
> and "servo tracks" written on one surface (at least the ones in sealed
> hard drives, though I think that the old 5MB removable platter ones used
> an an optical sensor to determine position, because in those days every
> surface was precious.
The are also precise because they're not responsible for holding a
cutting tool in position when the darn thing just wants to chatter!
Lathe == more rigidity is better.
Voice coil == infinitely compliant.
Christopher Tidy
>> If Takisawa is still selling their TPS machines, it seems likely that no
>>one has succeeded in building anything better. And they all have some
>>engineer or another who knows how a voice coil works. <g>
>
>
> See US Patents 5,085,109 and 5,313,694. The first uses a hydraulic servo
> system, the second uses a servo motor of some kind. Lots of other details are
> given. Referenced and referencing patents may also be instructive.
Thanks for the links to the patents. Those systems are more like what I
was imagining. I was surprised that the magnetostrictive actuators Ed
mentioned can give a great enough distance of travel. To what degree are
pistons typically elliptical, relative to their diameter? Is it visible
with the naked eye?
Best wishes,
Chris
Joseph Gwinn
Christopher Tidy <cdt22...@cantabgold.net> wrote:
I don't really know how big a deviation from circular is needed, but I'd guess
that it isn't much, and isn't visible.
Magnetostrictive actuators generate very high force levels, so one can use
levers to trade force for stroke. The length of a Terfenol-D rod changes by
about one part per thousand as one goes from zero to 2,000 gauss field.
Joe Gwinn
Christopher Tidy
> I don't really know how big a deviation from circular is needed, but I'd guess
> that it isn't much, and isn't visible.
Ah, right. I was thinking of a lathe that can turn parts that are
noticeably elliptical to the naked eye. I imagine it's simpler to design
a machine if the workpiece only needs to be slightly elliptical.
Chris
Ed Huntress
Typical ovality runs around 0.010 to 0.040 inches, depending upon the piston
size and purpose. It can be more or less. It could be enough to see, if you
took two identical pistons and laid them side-by-side on a surface plate,
and oriented them 90 deg. to each other in rotation.
This is for modern cast pistons. Forged pistons are made with an alloy that
has a significantly higher coefficient of thermal expansion. You might want
to check on diesel engine pistons to see what their range is.
I haven't looked, but I expect that you'll find good information about them
at the SAE's website. If you're Googling, look for "cam-ground pistons."
--
Ed Huntress
Christopher Tidy
> Typical ovality runs around 0.010 to 0.040 inches, depending upon the piston
> size and purpose. It can be more or less. It could be enough to see, if you
> took two identical pistons and laid them side-by-side on a surface plate,
> and oriented them 90 deg. to each other in rotation.
Hard to see then. Not quite beer barrel proportions.
> This is for modern cast pistons. Forged pistons are made with an alloy that
> has a significantly higher coefficient of thermal expansion. You might want
> to check on diesel engine pistons to see what their range is.
>
> I haven't looked, but I expect that you'll find good information about them
> at the SAE's website. If you're Googling, look for "cam-ground pistons."
Found a good link here:
http://www.autospeednet.com/sites/3dauto.com/viewterm/3332/
Thanks for the knowledge you've shared in this thread!
Best wishes,
Chris
Ed Huntress
That's good. Remember, as I mentioned with the Ford pistons, the latest and
greatest can have several different profiles from crown to skirt.
>
> Thanks for the knowledge you've shared in this thread!
I only wish the *real* expert here was around. He's an engineer with a
piston company you would recognize, but which I'm sworn not to name. <g> He
knows all the subleties, including ones he can't tell *us* about.
Anyway, 'glad you found it useful.
--
Ed Huntress