I have worked in a NATA certified metrology for over 30 years and I have been taught the following and using logic would be correct.When you do it between centres you are actually checking concentricity to the centre point to which you are rotating it and is the best way to check concentricity of a number of diameters. Any run out could be attributed to either roundness or concentricity or a combination to both. The only 2 ways to check roundness is to use an indicator and a surface table or a toolmakers microscope that has a rotating head or a machine that has been checked for run out with the shaft in a vertical orientation. With the indicator and table you just roll the diameter while it is sitting on the table under the indicator and take a reading, then rotate the diameter so you are not checking the same part of the diameter and then repeat. The more different sections of the diameter you check gives you a more accurate reading. It is the same as using a mircometer but is way more accurate. 2 contact points is all that is needed. Any more, like using a vee block and you are introducing a variable that cannot be measured. This method can be used to check long shafts. Short shafts can be set up vertically in a vee block and by using an indicator set up in a vertical milling machine you set the zero in the x and y and any run out is out of round. We use a toolmakers microscope that has been calibrated so we know it is accurate but it can only check shafts that are less than 5 inches long due to height restrictions. Just my 2 cents worth.
The idea of testing the work-piece in two V-blocks is that when an elliptically shaped part is rotated on a V-block is that when an elliptically shaped part is rotated on a V-block of angle 60°, no change in reading is indicated, whereas if the same part is rotated on a 90° angle Vee-block, two maximum and two minimum readings are indicated on the indicator.
If you look at Moore's "Foundations of Mechanical Accuracy" page 258, there is a good explanation for the use of different angle v-blocks. A part may be out of round in a regular way - that is, it can have a number of even, symmetric lobes. If the number of lobes is even, the out of roundness can be discovered using a mike. But if there are an odd number of lobes, then you need to use other methods. You need to use different angle v blocks because for each angle of block, there is a number of lobes that will measure as round when it is not. For a 90 degree v block, a seven lobed part will measure as round, and a 60 degree block will hide a 5 lobed out of roundness - and magnify a 3 lobed condition. By using both angles of block, you can get a better idea of the true roundness of the part.
That was fun to watch! I love your personality Don. Thanks for taking the time to teach. I just love the machining trade, I got lucky choosing this field of work.
I see that height gauge again that you made used on other video's you have done. I, am more interested in that than your video on this occasion . How does it move up & down? What is the movement? I worked with a toolmaker years ago. He in his apprenticeship, made a vernier height gauge, it was beautiful! You couldn't tell it was hand made, but it did not have that height adjustment that interests me. Pedro...
Thank you for publishing your videos on RUclips. They are very informative and entertaining to boot! If I could make one suggestion - I think you should edit the videos to make them more concise and shorter. Not to hide any mistakes but today people have much less patience and would appreciate them more. Jeff
Don, I am of the belief that in the first test, you are checking concentricity to the centers. This would also show an out of round condition, but it would be difficult to tell if the part is out of round or if its not concentric to the centers. Picture this, take a shaft perfectly round, drill a center in each end directly inline with one another, but .001" off of center then put it between centers. The resulting measurement would be .002" out of concentricity to the centers, but yet, the part is still perfectly round. - Thank you for the excellent videos. Scott
Thanks Don for nice videos. I think the 90 degree V-blocks keep the center of rod almost always at the same level. That's why they can show run out more correctly while the 60 degree V-blocks are hiding run out by moving up and down the rod center while rotating!
Watching this video several times, I believe that the reason for the glaring difference in FIM is mostly (if not entirely) due to the position of the Test Indicator probe - it's not nearly, not close enough even - to the same contacting position for each V-block measure taken. If I wanted to discover and share any difference between two differing V-Block angles, I'd fix the Test Indicator to the same relative position for each measure taken. That's a huge variable when taking measurements and expecting tenths to be the yardstick with such a casual set up. Hell, I coulda used a yellow, wooden yardstick with .0001" tick marks is what I'm saying (yeah, the one I got at the County Fair). Forgive me, that was a bit harsh - He shoulda used a Dial Indicator with a flat-surfaced T probe instead of a Test Indicator with a ball probe. Then the relative position of the probe centerline wouldn't have been so critical and the differences would have been just tenths - and the set up coulda remained casual. ruclips.net/video/YUeFNGslzZo/видео.html
Don, Years ago research was done by the TAFT-PIERCE company that showed centerless grinders produced a cam shape that would appear to measure as round. The company produced a variety of samples that measured the same diameter in every angular position but were exaggerated in size to clearly reveal that they were cam shaped and not round. A true round shape must be created by grinding on centers and not on a centerless grinder. When the Moore company designed their jig bores and later jig grinders, they took this fact into consideration when they manufactured their spindles. That is one of many reasons their machines are so precise. The purpose of a 60 degree V-Block and a 60 degree anvil micrometer is to accurately measure parts coming off of a centerless grinder. A standard micrometer will not give an accurate measurement for spindles ground with a centerless grinder. For a shaft ground ON centers standard micrometers and 90 degree V-Blocks will produce an accurate measurement. david
Ah but SOME "round" parts are INTENTIONALLY AND SPECIFICALLY "CAM-GROUND" so they're NOT "round" but are "oval" instead. Many PISTONS are ground that way so when the expand from heat they expand INTO a "round" overall shape and have to be ground that way because they expand MORE "horizontally" i.e. along the "axis" of the engine cylinder block/crankshaft where the most material and wrist pin bosses/bores are and expand LESS "vertically" or perpendicular to the axis of the cylinder block/crankshaft. Because of their very high "density" and significant expansion when heated FORGED PISTONS are often "cam-ground" so they start "egg-shaped" and "loose" in a "cold" engine and then expand into "perfect circles" that are "tight" in their bores once warmed up. They also are typically fit "looser" overall than cast/hypereutectic pistons for the same reason which is why its pretty common for even a "fresh" forged-piston high-performance engine to have a little "piston slap" when cold. Other than THAT I agree totally that its completely asinine to think/claim that you can GRIND anything to a "perfect" flatness/concentricity when the grinding stone itself is CONTINUALLY "shrinking" and any "unflatness" or "out of round" is only going to get WORSE the more the part is GROUND. In particular SURFACE GRINDERS are only used by wannabe machinist HACKS to cover up their crappy "mill work" that results from going out and getting the "best deal" possible on some worn-out but REPAINTED Bridgeport or other "high-quality" mill based upon APPEARANCE ONLY without the FIRST CLUE that there's so much "slop" in the spindle bearings and "bends" in the "frame" and "gullies" in the "bed" that there's no chance in hell they'll ever put a part in it that will come OUT "flatter" than it WENT IN. There's no such thing as GOOD CHEAP TOOL PERIOD and DAMNED SURE NOT A MACHINE TOOL. But they go out and buy a bunch of Dupont-overhauled junk for their "machine shop" and then when they start turning out crappy work from their "good deal" mill they go buy a SURFACE GRINDER and ALL THEIR PROBLEMS ARE SOLVED!
Fat man DEERMEYER1, And it should be noted that you have posted NOT ONE machining video! Not one! Because I doubt you even have a Bridgeport on that run-down farm of yours. Seems as all you have are guns.
awesome video thanks so much for doing a test that was very nice of you to show that. goes to show that you cant always go off of what others say until you have tried it or seen it like this first hand. i love knowing new and better ways for checking parts considering i have only been machining since i started college after high school back in 2006 and missed a few years from dang lay off back in 09. iv been very fascinated with all your videos and always in search to further my machining knowledge. unfortunately iv been stuck in a cnc operator position for the last few years and got life flighted to harborview a year ago with a badly broken leg and scull fracture that has been just none stop pain trying to continue as a cnc machinist making ingot and billet casting molds and equipment around nasty slippery floors and big cnc machines. I finally decided to put in for a QA position that just came up because i feel more intrigued by that type of stuff and hopefully its less strain and pain. if i can get the position i will for sure have to quiz them on this lol. already missed out on a actual cnc machinist/programming position because apparently im too valuable of and asset in my department they said. clearly thats why iv been forced into all the higher end parts and more physically demanding machines but im still in recovery and man its painful and sometimes sucks to be the only one who has yet to scrap a part and injured being treated like the injures are no longer a big deal :(
wow, this is a tricky subject because of all the variables! there are different out of rounds too! if your out of round has 2 high points, the flatter the angle you spin it in, the higher number you will get. like 60 to 90. but what if your out of round is a triangle shape??
That is really surprising. If that was on a test I would have failed :) I thought 60° would show more out of roundness because the shallower angle would cause the part to rise and fall more as the tangential "width" changes. Although the CL of the part does rise and fall more, i didn't expect it to cancel out movement at the the indicator. This is why I love your videos Don.
Yes., Vee block method will show more accurate results than any two- point instrument. The primary concern when using the vee-block method is that the vee block must be a specific angle in order to effectively measure circularity error. The FIM results require 'conversion' to obtain the actual radial circularity error of the part. The angle of the vee-block is calculated using the equation , vee-block angle = 180° - (360°/n) , where n = Number of equally spaced lobes. But, actual number of lobes on the part is unknown unless an instrument like the precision spindle is used. For most application, a 60° vee-block will suffice. A secondary limitation of the vee-block method is that the method, with the correct-angle vee-block can exaggerate the actual circularity error of the part For this reason, the observer must compensate for this error in the result. So, out of roundness = (( FIM / (1+COSECEANT A) ) , where A = 1/2 * vee block included angle . Finally, Try various vee blocks of different angles to find the one block angle that shows the highest (worst) indications, then use that vee-block & the proper correction factor to obtain the result needed. Thanks
The above assumes the shaft is out of round, but not BENT. If the shaft is not straight plus out of round, each test will indicate a different result. They cannot just be substituted for each other without properly considering what you're measuring.
out of round is the deviation of the radius, concentricity is or what people usually mean to call straightness of the axis. the technical use of concentricity is to specify axial congruency between different features on a shaft usually.
With the larger axis vertical, I would have expected the centre height on the 90 deg right-hand example to have been slightly lower due to the part sitting lower in the V as it is in the bottom right-hand example.
I prefer using two V blocks with the indicator perpendicular to the shaft. However since I only have 60 deg blocks I can't try your comparison. Now this will be a lingering question until I get some 90's. Thanks. Love watching.
The biggest difference between the 2 V-blocks is that the 60 degree block contact points are working from the sine of the angle and therefore effects the way the part rises and falls in the block. Remember you got .001 tir. The sign of the angle is 30 degrees is .5. Taking that into consideration it's exactly right. Also the sign of the 45 degree angle block is .707.
A cylinder that is round AND concentric will have a CONSISTENT DIAMETER from end to end regardless of where its measured relative to its axis and radius and out of roundness when checked by people who know what the hell they're doing and do it for a living instead of selling the tools to do it don't use a dial indicator to do so. Out of roundness is generally checked in engine cylinders and main/cam bearing bores and on crankshaft/camshaft bearing journals and other IDs and ODs that are supposed to be "perfectly round". And its done with an INSIDE MICROMETER or telescoping gauge and outside micrometer when an ID is being measured and at TWO "points" 180 degrees apart and with an outside micrometer when an OD is being checked. Again, two points 180 degrees apart, Typically "vertically" and "horizontally" when possible to do so in a bearing bore with the component the bearing bores are located in sitting in its "normal" position or 180 degrees from that position i.e. a cylinder block "upright" or "upside down" on an engine stand. Same goes for "centerline" bearing journals on crankshafts/camshafts i.e. the main journals in the case of the camshaft. Connecting rod journals are typically "worn" or actually "pounded" out of round by excessive engine lugging and/or detonation/pinging etc so connecting rod journals should be measured from "top to bottom" with the rod journal at "TDC" and 180 degrees from that or "side to side". The "top" of the journal is where out of roundness will be found. TAPER of IDs and ODs is ALSO measured by making a set of measurements at each "end" of the ID or OD. For instance at the top and bottom of a cylinder bore (but above the bottom of "ring travel" because bore wear stops where the rings "bottom out" with the piston at BDC and also below the top of ring travel below the "cylinder ridge" that's formed because the pistons stop before the rings get to the "top edge" of the cylinder. A bearing journal or bore is treated the same way. Out of round located at the "front and back" of the bore/journal and if its a "shared" connecting rod journal like on a V8 crankshaft then the shared journal is "split" and is treated as TWO journals for "blueprinting" purposes and EIGHT total OD measurements will be taken to determine out of round and taper of each. Concentricity is having the OD of a shaft be "equidistant" from the "centerline" so there is no "wobble" as the shaft rotates. Which can still occur even if there is "zero out of round" and even if there is "zero runout" of the shaft itself. For example there are some camshafts that have fuel-pump drive "eccentrics" that simply "offset" circular and perfectly round "lobes" that are slightly "offset" from the centerline of the camshaft. The "lobe" is both perfectly circular and if the shaft isn't bent it will have zero runout but that "lobe" will NOT be "concentric" on the camshaft. The BEARING JOURNALS on the other hand WILL BE. When the camshaft is placed in a PAIR of V-blocks and a PROPER DIAL INDICATOR with the good old VERTICAL PLUNGER instead of that stupid "lever-action" linkage setup that's ONLY used by SMART machinists/mechanics when a standard indicator simply won't fit/work is located ABOVE each bearing journal with the plunger "perfectly" vertical so runout is being measured/displayed DIRECTLY a "perfect" camshaft will have ZERO RUNOUT at the indicator (up/down "motion") and at another "smooth" which will also prove bearing journal "concentricity' IF the bearing journals have also been measured with a MICROMETER and they are not "out of round". If the bearing journals are NOT "out of round" and are "concentric" on the camshaft centerline then the shaft is "bent" and its "runout" you're seeing on the dial indicator and you'll find that the runout is decreased/increased if the indicator is moved to journals further from/closer to the "bend" itself. And THAT is how out of round/concentricity/runout are measured in the "real world" by people who actually know what the hell they're doing, are doing it for a living, have to do it right and are in the business of USING PRECISION MEASURING TOOLS PROPERLY instead of the business of SELLING "precision" measuring tools to "machinists" who if they believe anything this guy says/demonstrates is "factual" and his processes are "correct" may be "making a living" as "machinists" but are highly unlikely to be making a GOOD LIVING or at least unlikely to be doing so on a "long-term basis". Because whether you believe it or not, this guy is just about as full of shit as they come on pretty much everything he demonstrates/claims and he may have "worked in" a machine shop once upon a time and may still have a "machine shop" he hawks his "precision tools" out of but he's sure as hell not in the "machine shop" business as his bread and butter or he and his employees wouldn't have the time and inclination to make all these "commercials" because they'd have work stacked up to last them months and months like every GOOD machinist/machine shop I've ever known/been in. Anybody who continually reaches for a DIAL INDICATOR to "measure" things like "flatness" and "out of round" when anybody with half a clue knows that's straight-edge and micrometer territory and even uses the "wrong" dial indicator set-up at that (those lever-action indicator linkage contraptions "lose accuracy" as the "tip" moves through its arc because it MOVES in an arc while the "geometry" remains constant and they're only "perfectly accurate" at the "center" of their travel which is why they're only used by smart/good machinists/mechanics when nothing else will work and THEN only to measure MOVEMENT since we use a "sled gauge" with a conventional indicator in it to measure things like "liner protrusion" from diesel engine cylinder blocks and other "flat work" where other measuring tools just won't work. As for him running for the SURFACE GRINDER for every "flatness emergency" I MAY have been in ONE machine shop EVER that had a "surface grinder" and that would be Gessford Machine in Hastings, NE. They can and do make/machine damned near everything and anything there and PROBABLY have one but I'll GUARANTEE IT AIN'T USED TO "MACHINE" SURFACES because SURFACE GRINDERS NEVER HAVE BEEN "MACHINE TOOLS" and were for SURFACE FINISHING ONLY to make a "rough" surface "smooth and shiny" OR a "smooth and shiny" surface "rough" AFTER it had been machined "flat" in a MILL or other actual MACHINE TOOL until the CURRENT CROP of online "expert machinists" and "wannabe machinists" who apparently "learn" from them came along and stared selling "surface grinding" as a way to make an "unflat" surface "flat" or "unparallel" surfaces "parallel" all the way down to MILLIONTHS OF AN INCH OF "FLATNESS". Too bad there are "surface grinding professionals/shops that on their own websites speak the TRUTH about surface grinding and that's its to ALTER THE POST-MACHINING SURFACE FINISH AS CREATED BY THE MACHINING PROCESS/TOOL USED TO MACHINE IT and that AT BEST they can "hold" about .0015" of "flatness" across/along a 12" surface even they they "target" .001" and occasionally "luck out" and can "achieve" .0007" or so. You gotta be just pretty damned dense to not realize that since a GRINDING WHEEL is "shrinking" as it GRINDS its NOT a "precision cutter" and the MORE PASSES YOU MAKE REGARDLESS OF DIRECTION THE MORE WORN DOWN IT WILL GET AND THE MORE "unflat" the surface being GROUND will become. But guys that seek to "learn" how to be "machinists" from RUclips VIDEOS and the "experts" who seek to SELL THEM TOOLS apparently don't/won't make that "mental leap" and would much rather use a "surface grinder" to make their "machining" look PRETTY even if we never actually SEE them doing the REAL "machining" and there's not a MILL TO BE SEEN or at least NOT VERY OFTEN in their "how-to videos". Which is PROBABLY because MILLS that are still "precise" and capable of making ANYTHING "flat" decades and decades after being built/rebuilt are as RARE AS HEN'S TEETH and those who HAVE THEM don't SELL THEM or at least not for what a "beginning machinist" who is just opening his own "machine shop" THINKS is a "fair price". Instead they go out and by some cheap POS they think they got a "good deal" on and when they "mill" anything it ends up being "less flat" and "rougher" than when they started. So THEN they RUN TO THE SURFACE GRINDER so there "wavy" so-called "flat" surfaces that they just can't find ANY "unflatness" in with their "dial indicator" that must be so flat its in the "millionths" because to the "beginning machinist" and the guy who wants to SELL HIM PRECISION TOOLS that's APPARENTLY the next "increment" after "ten thousandths" IF they even have an indicator that "precise" and are using/reading it "properly? (which is never the case if its on a magnetic base "stem" instead of in a SLED GAUGE). Measuring "flatness" and "out of round" with a dial indicator. I'm going to have to show these "how-to" videos to a few of my REAL machinist buddies. They're so damned busy/stressed they could use a laugh. There IS one type of "dial indicator" that is commonly used to check bores for out of round and taper ahnd they're called "dial bore gauges" but they're precision instruments that basically combine an "inside micrometer" or telescoping gauge with a spring-loaded "plunger" and properly designed/constructed linkage so the "horizontal" measurements in the bores are displayed on a dial indicator "display" and the gauge is "set" to read "0" when installed in the "standard" factory-new bore diameter for that particular engine so that when the dial bore gauge is placed in a worn bore it can be "slid" up and down and around the cylinder bore quickly and easily to the same four points that would otherwise be measured with an inside micrometer or telescoping gauge/outside micrometer and the bore's out-of-round and taper measured/calculated in a few seconds instead of a few minutes.
DEEREMEYER1 jesus, youre whoring up the whole thread with bullshit. Concetricity has nothing to do with the radius error of a part, its used to specify axial congruence of features.
out of round = roundness of 1 cylinderical surface (egg shaped, 3 point out of round from 3 jaw chuck, etc.) not concentric = the various cylindrical surfaces on a crankshaft are intentionally not concentric. Those individual surfaces are hopefully not out of round. in the real would its often a combination of both, depending on fabrication methods. to the best of my knowledge the 'optimal' v angle for a true reading depends on the type of out of round you got. A 3 point out of round from the lathe chuck has the lowest points offset by 120 degrees, the points of contact will be different between 45/60/90 degree v-blocks.
I think checking a shaft between centers, like what you did first, shows the concentricity between the center holes and the diameter where the indicator needle is. Measuring with a micrometer, 90° apart should show circularity (roundness).
Measurement with a micrometer is necessary but not sufficient to measure roundness. For example if you measure the diameter of the rotor in a Wankel rotary engine you will get a peculiar result. The rotor is obviously not round but it has the same diameter no matter where you measure it with a micrometer. This condition is called lobing and can happen especially with centerless grinding. The work can have 3,5 or more odd number of lobes, though 3 is the most common. This lobing is probably in the millionths of an inch but can be critical for (as others have mentioned) things like automotive hydraulic valve lifters and fuel injector pintles.
@@aceroadholder2185 Whether or not the "lobes" are equally-spaced makes a big difference, too. Refer to tri-lobular screws, such as Taptite®, a completely different kettle of lobular fish - you gotta use both a regular mic and a tri-mic with a 60° V to take all three of its necessary diametral measurements. E-Z beans since it's invention in 1965, at least a couple billion of this family of screws have been reliably measured on the shop floor - one.pan.at.a.time - with oily, dirty hands and lots of cigarette smoke. To quote Johnny Cash, "Huh - 1965. That's older even than this video."
Roundness/Circularity would ideally be inspected at diametrically opposed points to so that we can determine the center. Unfortunately, nobody makes 180 degree vee blocks, so the 90 degree vee block represents the next best thing. The problem with the 60 degree vee block is that it creates a larger space for eccentricity to 'hide' during an inspection for circularity. The 45 degree vee block closes that gap significantly whilst also making for easy and steady rotation.
to measure the roundness you actually would use two v-blocks, the one having a v as wide as possible the other as narrow as possible an elliptical out of round will give a complete different measurement than a tri-lobed one in either of the blocks by comparing the measurements one can determine the form of out of roundness and the amount of out of roundness
It is possible to misadjust a centerless grinder so that it grinds a trilobular part. The only to find out is with a 60 degree v. They make micrometers with a 60 degree anvil for checking this.
+Richard Haisley You can see the error of a trilobular part in a 90 degree block too. Best method is to use two indicators measuring from different spots, so it really doesn't matter what the angle of the block is.
Compelling. What kinds of "Out of round" occur most often? All I've ever seen are ovals, D-shaped (Worn on one side) , egg shaped (worn over a wide angle), and the possibility of a rounded square. But that's just from crankshafts, camshafts, and reciprocating tools.
Hi Don! Thanks alot for your very informative videos. I have one question though: Very often I see people using finger indicators, while they are not paying attention to have the finger in a 90° angle to the measurement direction. I was under the impression, that this can cause significant error.. having a 40° angle causes 25% error... see "cosine error".. What do you think about this?
Cosine error isn't much of a problem when using a test indicator to make a comparative measurement. For instance: indicating you vise parallel to the table, indicating in a hole to check concentricity, or checking a shaft for run-out. However, if an absolute linear measurement is needed, cosine error can be a problem and may require a corrective multiplier. Using a "pear" shaped contact will eliminate the cosine error up to approx. 36 degrees. The other factor that can come in to play is replacing the contact point with a new point that is longer or shorter than the original, which will also affect your measurements.
It absolutely does and even when used "correctly" the additional linkage and "slop" still results in "inaccuracy" compare to a "standard" dial indicator. Which is why PROFESSIONALS only use "finger indicators" where it's impossible to USE a "standard" indicator setup. Of course we ALSO don't use a DIAL INDICATOR to measure ANYTHING BUT MOTION when its installed on "stem" and put an indicator in a SLED GAUGE if we want to do any "measuring" on a "flat surface" and THEN we're only using them to measure how far some other surface/part PROTRUDES ABOVE that surface such as when checking valve protrusion from a diesel engine cylinder head "fire deck" or liner protrusion above the cylinder block deck surface. For measuring FLATNESS we use a STRAIGHT EDGE AND FEELER GAUGE and if the "unflatness" is too "small" for a .001" feeler gauge to fit underneath an 18" straight-edge, that's a "perfect" block anyway. As for OUT OF ROUND that's MICROMETER TERRITORY whether bore ID or shaft, journal etc OD or dial bore gauge territory for most REAL MACHINISTS checking a block for bore size and to see if it will "clean up" in a hurry since GOOD MACHINISTS rarely have the kind of spare time all the supposed/aspiring "machinists" that make/haunt these RUclips "how-to" videos seem to have. I do not currently know and in 25 years of diesel-mechanicking have never known a GOOD machinist/machine shop that didn't have at LEAST several WEEKS of work piled up waiting and could work for more than maybe 20-30 minutes AT MOST and even THEN only rarely without having the phone ring or the door open with more work for him/it. Never known a good machinist/shop to have a SURFACE GRINDER to "machine" parts "flat" either and I THINK the only machine shop I've been in that HAS ONE AT ALL would be Gessford's in Hastings, NE. And if they DO I'll damned well guarantee it AIN'T THERE TO "MACHINE" anything. It's there for SURFACE FINISHING like ALL surface grinders owned/operated by good machinists/shops that know what the hell they're doing, know what surface grinders are actually for and have a use/need for them do. GRINDING A PART "FLAT" WITH A CONSTANTLY WEARING/SHRINKING WHEEL. Hilarious.
Very interesting & thanks for the stimulating video. Haven't totally wrapped my mind around the difference just yet, but it has to revolve around the tangent angles of contact. 60° block has ??; 90° v-block has 120° of separation; wouldn't any more obtuse angle more apparently show an out-of-round condition? To the maximum extent, which would be measuring from two tangent points which are diametrically opposed (I.e. - [180° of separation] - micrometer).
Consider two extreme examples: a 180 degree "V block" -- a flat surface -- and a 20 degree V block. When the piece is rotated on the flat surface the indicator will show 0.1000 change. When it's rotated in the 20 degree V block the top surface will show about 0.208 change. The geometry of a 60 degree V block is such that the top surface of a perfect ellipse will show no vertical movement, irrespective of the rotation angle. So it's the worst possible choice. The geometry of a 90 degree V block is such that the top surface of a perfect ellipse will show exactly half the difference of the two axes. In both the 60 and 90 degree blocks the highest point on the top surface will move left and right (in the view on the drawing) as it's rotated. If the indicator is not moved so it stays on the highest point, it will show a deflection, but it will not be the same as with the 90 degree block, exactly as in your demonstration.
I really think it depends on how it's out of round. There are many ways to make shapes of constant width. If there where 3 high points would read very differently in a 60 degree v block than a 90 when compared to one with 4 high spots. I might do some drawings in fusion360 later to see if what I just said actually works like I think it would.
i think the 90 degree is better to use because it gives the biggest readout. as you guys as toolmakers always have tight tolerances you know you screwed it up when you got that far off your maximum drawing tolerance.
Since the egg is always trying to center itself in the Vee (left to right) A better way to inspect the part is from the side then, the angle doesn't pose a factor. In other words tip the V-block on its side and measure perpendicular to the side surface which is now down on the plate.
As a Cold Form Tooling Design Engineer, roundness in tooling isn't extremely critical especially with modern technology. 90% of our tooling designed is toleranced between .001 and .002 TIR. Very easily achieved considering almost all the tooling is CNC made, hardened, and then finish ground to spec. When it comes down to it, we are more concerned with concentricity. A die is made, pressed in a sleeve, and then the sleeve and die are pressed into a die case. With the die, sleeve, and die case all pressed together, it can be put into a lathe, indicated in on the od of the die case, which has been centerless ground to spec, and then observed for concentricity. The case can easily be indicated in to less then .0005. Now we can check the i.d. of the die pressed in the case. and that too will be within .0005 roundness with the whole assembly. As long as everything is concentric, it can be thrown in the machine and be just fine. In my opinion the act of putting the part, die, workpiece etc. on a rotating axis is the best way to measure for out of round and concentricity if required. You can easily find highs and lows. Im sure there are applications where roundness is super critical but not so much in the tool and die field I would say. But thats all based off of our applications. Very good video! Always enjoy watching them! :)
Uh, try measuring the diameter of the part at several points along the length and at two positions 180 degrees apart at each of those points. That's how we check "round" parts for "roundness" here in the real world when it comes to things like cylinder bores, crankshaft journals, bearing bores, etc. Runout and "out of round" are two entirely different things and out of round is determined by measuring IDs of bores, ODs of shafts, etc. And if you want to check RUNOUT you don't use a V-block long enough to drop the entire shaft in. You place a V-block at EACH END so you can rotate it and check the RUNOUT in the center where it will be MOST EVIDENT. You also use a STANDARD DIAL INDICATOR with a "vertical plunger" placed "on top of" that shaft AT the center so runout is directly "measured" and indicated instead of through a "lever-action" mechanism that's typically used ONLY where a "standard" dial indicator can't be located DIRECTLY IN LINE WITH THE "TRAVEL" OF THE MOTION BEING MEASURED/INDICATED. Any And EVERY time you introduce "angles" into the "linkage" of a dial indicator plunger setup you're introducing error and inaccuracy. The tip of those "lever-action" indicator setups moves in an ARC and not "vertically" and its "geometry" relative to the plunger/rack-and-pinion of the indicator itself is constantly changing AS it swings through that arc.
DEEREMEYER1 internet genius, whats the cosine error at 15° for .002? Ill fill you in, its less than a 10th. On a test indicator which has virtually no travel, a .030 sweep has a total sector angle of 2.6°, the cosine error from one end to the other is +-15 millionths, there arent many drop indicators that can do that. As far as "run out" there are 2 types, circular runout or out of round, and total. Circular only is concerned with circularity at any single cross section. Total is the compound of circularity, straightness, angularity and cylindricity.
So these lever type indicators are incorrect? You should explain that to Mitutoyo, Starret, Tesa and all other high end test indicator manufacturers. These are the same style indicators that machine manufactures and certified repair companies use to set up machines down to the half micron.
DEERMYERER1 its quite obvious you're not a "real" machinist. If you paid attention to what Don said at the beginning you would know the difference, moron
The most important fact in which measure is "true" have shape of cirualrity - how many lobbing we have around the part. In my industrial practic on part with diameter above 250 up to 700 [mm] ( in mm because im from Poland) most comon are 3 point out of round - reason is technologu turning in 3 jaw chucks, In your case you have elliptical shape in my opinion and it's no matter which v-block you will use 60 or 90deg both result are correct but you have to correct read it including geometrical issues. here you have little bit more about it what-when-how.com/metrology/measurement-of-circularity-metrology/ Realy intresting video, sorry for writting mistakes and all the best from Poland
I think a 170* would be even better. The reason is pretty simply, you are making the out of round condition visible along the axis it is measuring on. Oh wait... It doesnt make a difference. The math is the same. ;-) Great video'
I have worked in a NATA certified metrology for over 30 years and I have been taught the following and using logic would be correct.When you do it between centres you are actually checking concentricity to the centre point to which you are rotating it and is the best way to check concentricity of a number of diameters. Any run out could be attributed to either roundness or concentricity or a combination to both. The only 2 ways to check roundness is to use an indicator and a surface table or a toolmakers microscope that has a rotating head or a machine that has been checked for run out with the shaft in a vertical orientation. With the indicator and table you just roll the diameter while it is sitting on the table under the indicator and take a reading, then rotate the diameter so you are not checking the same part of the diameter and then repeat. The more different sections of the diameter you check gives you a more accurate reading. It is the same as using a mircometer but is way more accurate. 2 contact points is all that is needed. Any more, like using a vee block and you are introducing a variable that cannot be measured. This method can be used to check long shafts. Short shafts can be set up vertically in a vee block and by using an indicator set up in a vertical milling machine you set the zero in the x and y and any run out is out of round. We use a toolmakers microscope that has been calibrated so we know it is accurate but it can only check shafts that are less than 5 inches long due to height restrictions. Just my 2 cents worth.
+dikkybee Thank you for taking the time.
"Is that time for lunch or to go home?"
That's when you know you've found something you like. And you get a paycheck out of it.
Thank you. Your videos helped alleviate the tedium of my recent, month long, hospital stay.
Thanks Don and Jim, a very interesting video (as always).
btw: I'm in love with that surface plate... :-)
The idea of testing the work-piece in two V-blocks is that when an
elliptically shaped part is rotated on a V-block is that when an elliptically shaped part is rotated
on a V-block of angle 60°, no change in reading is indicated, whereas if the same part is rotated
on a 90° angle Vee-block, two maximum and two minimum readings are indicated on the
indicator.
I have nothing to do with tools and machining and surface plates but I just love these videos.
Hopefully, you're learning something from these videos. If you're interested in machining give one of us a shout.
If you look at Moore's "Foundations of Mechanical Accuracy" page 258, there is a good explanation for the use of different angle v-blocks.
A part may be out of round in a regular way - that is, it can have a number of even, symmetric lobes. If the number of lobes is even, the out of roundness can be discovered using a mike. But if there are an odd number of lobes, then you need to use other methods.
You need to use different angle v blocks because for each angle of block, there is a number of lobes that will measure as round when it is not. For a 90 degree v block, a seven lobed part will measure as round, and a 60 degree block will hide a 5 lobed out of roundness - and magnify a 3 lobed condition.
By using both angles of block, you can get a better idea of the true roundness of the part.
+David Atkinson I was thinking something like that but could not put it into words. Thank you and thanks to ST too of course
That was fun to watch! I love your personality Don. Thanks for taking the time to teach. I just love the machining trade, I got lucky choosing this field of work.
I see that height gauge again that you made used on other video's you have done. I, am more interested in that than your video on this occasion . How does it move up & down? What is the movement?
I worked with a toolmaker years ago. He in his apprenticeship, made a vernier height gauge, it was beautiful! You couldn't tell it was hand made, but it did not have that height adjustment that interests me.
Pedro...
+Peter Spence We will do some close ups in an upcoming video so you can see it better. Thanks for watching.
Thank you for publishing your videos on RUclips. They are very informative and entertaining to boot! If I could make one suggestion - I think you should edit the videos to make them more concise and shorter. Not to hide any mistakes but today people have much less patience and would appreciate them more. Jeff
Don, I am of the belief that in the first test, you are checking concentricity to the centers. This would also show an out of round condition, but it would be difficult to tell if the part is out of round or if its not concentric to the centers. Picture this, take a shaft perfectly round, drill a center in each end directly inline with one another, but .001" off of center then put it between centers.
The resulting measurement would be .002" out of concentricity to the centers, but yet, the part is still perfectly round.
-
Thank you for the excellent videos.
Scott
I see your point, thank you for sharing. And thank you for watching, we appreciate it.
Thanks Don for nice videos.
I think the 90 degree V-blocks keep the center of rod almost always at the same level. That's why they can show run out more correctly while the 60 degree V-blocks are hiding run out by moving up and down the rod center while rotating!
You're welcome, thanks for watching!
Cool!!!
I was un-aware of this fray, entirely.
I guess I won't be needing a 60 degree block..
Thanks yet again!
Mark
Watching this video several times, I believe that the reason for the glaring difference in FIM is mostly (if not entirely) due to the position of the Test Indicator probe - it's not nearly, not close enough even - to the same contacting position for each V-block measure taken. If I wanted to discover and share any difference between two differing V-Block angles, I'd fix the Test Indicator to the same relative position for each measure taken. That's a huge variable when taking measurements and expecting tenths to be the yardstick with such a casual set up. Hell, I coulda used a yellow, wooden yardstick with .0001" tick marks is what I'm saying (yeah, the one I got at the County Fair). Forgive me, that was a bit harsh - He shoulda used a Dial Indicator with a flat-surfaced T probe instead of a Test Indicator with a ball probe. Then the relative position of the probe centerline wouldn't have been so critical and the differences would have been just tenths - and the set up coulda remained casual.
ruclips.net/video/YUeFNGslzZo/видео.html
Don,
Years ago research was done by the TAFT-PIERCE company that showed centerless grinders produced a cam shape that would appear to measure as round. The company produced a variety of samples that measured the same diameter in every angular position but were exaggerated in size to clearly reveal that they were cam shaped and not round. A true round shape must be created by grinding on centers and not on a centerless grinder.
When the Moore company designed their jig bores and later jig grinders, they took this fact into consideration when they manufactured their spindles. That is one of many reasons their machines are so precise.
The purpose of a 60 degree V-Block and a 60 degree anvil micrometer is to accurately measure parts coming off of a centerless grinder. A standard micrometer will not give an accurate measurement for spindles ground with a centerless grinder. For a shaft ground ON centers standard micrometers and 90 degree V-Blocks will produce an accurate measurement.
david
Thank you for you input
Yes, the centerless grinder specifically tends to produce parts with triangular-ish cross-section, and the 60deg angle accentuates that.
Ah but SOME "round" parts are INTENTIONALLY AND SPECIFICALLY "CAM-GROUND" so they're NOT "round" but are "oval" instead. Many PISTONS are ground that way so when the expand from heat they expand INTO a "round" overall shape and have to be ground that way because they expand MORE "horizontally" i.e. along the "axis" of the engine cylinder block/crankshaft where the most material and wrist pin bosses/bores are and expand LESS "vertically" or perpendicular to the axis of the cylinder block/crankshaft.
Because of their very high "density" and significant expansion when heated FORGED PISTONS are often "cam-ground" so they start "egg-shaped" and "loose" in a "cold" engine and then expand into "perfect circles" that are "tight" in their bores once warmed up. They also are typically fit "looser" overall than cast/hypereutectic pistons for the same reason which is why its pretty common for even a "fresh" forged-piston high-performance engine to have a little "piston slap" when cold.
Other than THAT I agree totally that its completely asinine to think/claim that you can GRIND anything to a "perfect" flatness/concentricity when the grinding stone itself is CONTINUALLY "shrinking" and any "unflatness" or "out of round" is only going to get WORSE the more the part is GROUND. In particular SURFACE GRINDERS are only used by wannabe machinist HACKS to cover up their crappy "mill work" that results from going out and getting the "best deal" possible on some worn-out but REPAINTED Bridgeport or other "high-quality" mill based upon APPEARANCE ONLY without the FIRST CLUE that there's so much "slop" in the spindle bearings and "bends" in the "frame" and "gullies" in the "bed" that there's no chance in hell they'll ever put a part in it that will come OUT "flatter" than it WENT IN. There's no such thing as GOOD CHEAP TOOL PERIOD and DAMNED SURE NOT A MACHINE TOOL.
But they go out and buy a bunch of Dupont-overhauled junk for their "machine shop" and then when they start turning out crappy work from their "good deal" mill they go buy a SURFACE GRINDER and ALL THEIR PROBLEMS ARE SOLVED!
Fat man DEERMEYER1,
And it should be noted that you have posted NOT ONE machining video! Not one! Because I doubt you even have a Bridgeport on that run-down farm of yours. Seems as all you have are guns.
awesome video thanks so much for doing a test that was very nice of you to show that. goes to show that you cant always go off of what others say until you have tried it or seen it like this first hand. i love knowing new and better ways for checking parts considering i have only been machining since i started college after high school back in 2006 and missed a few years from dang lay off back in 09. iv been very fascinated with all your videos and always in search to further my machining knowledge. unfortunately iv been stuck in a cnc operator position for the last few years and got life flighted to harborview a year ago with a badly broken leg and scull fracture that has been just none stop pain trying to continue as a cnc machinist making ingot and billet casting molds and equipment around nasty slippery floors and big cnc machines. I finally decided to put in for a QA position that just came up because i feel more intrigued by that type of stuff and hopefully its less strain and pain. if i can get the position i will for sure have to quiz them on this lol. already missed out on a actual cnc machinist/programming position because apparently im too valuable of and asset in my department they said. clearly thats why iv been forced into all the higher end parts and more physically demanding machines but im still in recovery and man its painful and sometimes sucks to be the only one who has yet to scrap a part and injured being treated like the injures are no longer a big deal :(
wow, this is a tricky subject because of all the variables! there are different out of rounds too! if your out of round has 2 high points, the flatter the angle you spin it in, the higher number you will get. like 60 to 90. but what if your out of round is a triangle shape??
@ 9:16 the image 90 degree on the 90 degree block - the center dimension cannot stay the same... But the image rather makes sense.
That is really surprising. If that was on a test I would have failed :) I thought 60° would show more out of roundness because the shallower angle would cause the part to rise and fall more as the tangential "width" changes. Although the CL of the part does rise and fall more, i didn't expect it to cancel out movement at the the indicator. This is why I love your videos Don.
Yes., Vee block method will show more accurate results than any two- point instrument.
The primary concern when using the vee-block method is that the vee block must be a specific angle in order to effectively measure circularity error.
The FIM results require 'conversion' to obtain the actual radial circularity error of the part.
The angle of the vee-block is calculated using the equation , vee-block angle = 180° - (360°/n) , where n = Number of equally spaced lobes.
But, actual number of lobes on the part is unknown unless an instrument like the precision spindle is used. For most application, a 60° vee-block will suffice.
A secondary limitation of the vee-block method is that the method, with the correct-angle vee-block can exaggerate the actual circularity error of the part
For this reason, the observer must compensate for this error in the result.
So, out of roundness = (( FIM / (1+COSECEANT A) ) , where A = 1/2 * vee block included angle .
Finally, Try various vee blocks of different angles to find the one block angle that shows the highest (worst) indications, then use that vee-block & the proper correction factor to obtain the result needed.
Thanks
The above assumes the shaft is out of round, but not BENT. If the shaft is not straight plus out of round, each test will indicate a different result. They cannot just be substituted for each other without properly considering what you're measuring.
Nice followup...
please do a video on the differents between out if round and out of centricity I would like to see that.
Noted, I will consider it. Thanks for the comment.
out of round is the deviation of the radius, concentricity is or what people usually mean to call straightness of the axis. the technical use of concentricity is to specify axial congruency between different features on a shaft usually.
With the larger axis vertical, I would have expected the centre height on the 90 deg right-hand example to have been slightly lower due to the part sitting lower in the V as it is in the bottom right-hand example.
I prefer using two V blocks with the indicator perpendicular to the shaft. However since I only have 60 deg blocks I can't try your comparison. Now this will be a lingering question until I get some 90's. Thanks. Love watching.
The biggest difference between the 2 V-blocks is that the 60 degree block contact points are working from the sine of the angle and therefore effects the way the part rises and falls in the block. Remember you got .001 tir. The sign of the angle is 30 degrees is .5. Taking that into consideration it's exactly right. Also the sign of the 45 degree angle block is .707.
Thanks Don. Please explain the difference between out of round and concentric or "concentricity.
A cylinder that is round AND concentric will have a CONSISTENT DIAMETER from end to end regardless of where its measured relative to its axis and radius and out of roundness when checked by people who know what the hell they're doing and do it for a living instead of selling the tools to do it don't use a dial indicator to do so. Out of roundness is generally checked in engine cylinders and main/cam bearing bores and on crankshaft/camshaft bearing journals and other IDs and ODs that are supposed to be "perfectly round". And its done with an INSIDE MICROMETER or telescoping gauge and outside micrometer when an ID is being measured and at TWO "points" 180 degrees apart and with an outside micrometer when an OD is being checked. Again, two points 180 degrees apart, Typically "vertically" and "horizontally" when possible to do so in a bearing bore with the component the bearing bores are located in sitting in its "normal" position or 180 degrees from that position i.e. a cylinder block "upright" or "upside down" on an engine stand. Same goes for "centerline" bearing journals on crankshafts/camshafts i.e. the main journals in the case of the camshaft. Connecting rod journals are typically "worn" or actually "pounded" out of round by excessive engine lugging and/or detonation/pinging etc so connecting rod journals should be measured from "top to bottom" with the rod journal at "TDC" and 180 degrees from that or "side to side". The "top" of the journal is where out of roundness will be found. TAPER of IDs and ODs is ALSO measured by making a set of measurements at each "end" of the ID or OD. For instance at the top and bottom of a cylinder bore (but above the bottom of "ring travel" because bore wear stops where the rings "bottom out" with the piston at BDC and also below the top of ring travel below the "cylinder ridge" that's formed because the pistons stop before the rings get to the "top edge" of the cylinder. A bearing journal or bore is treated the same way. Out of round located at the "front and back" of the bore/journal and if its a "shared" connecting rod journal like on a V8 crankshaft then the shared journal is "split" and is treated as TWO journals for "blueprinting" purposes and EIGHT total OD measurements will be taken to determine out of round and taper of each.
Concentricity is having the OD of a shaft be "equidistant" from the "centerline" so there is no "wobble" as the shaft rotates. Which can still occur even if there is "zero out of round" and even if there is "zero runout" of the shaft itself. For example there are some camshafts that have fuel-pump drive "eccentrics" that simply "offset" circular and perfectly round "lobes" that are slightly "offset" from the centerline of the camshaft. The "lobe" is both perfectly circular and if the shaft isn't bent it will have zero runout but that "lobe" will NOT be "concentric" on the camshaft. The BEARING JOURNALS on the other hand WILL BE. When the camshaft is placed in a PAIR of V-blocks and a PROPER DIAL INDICATOR with the good old VERTICAL PLUNGER instead of that stupid "lever-action" linkage setup that's ONLY used by SMART machinists/mechanics when a standard indicator simply won't fit/work is located ABOVE each bearing journal with the plunger "perfectly" vertical so runout is being measured/displayed DIRECTLY a "perfect" camshaft will have ZERO RUNOUT at the indicator (up/down "motion") and at another "smooth" which will also prove bearing journal "concentricity' IF the bearing journals have also been measured with a MICROMETER and they are not "out of round". If the bearing journals are NOT "out of round" and are "concentric" on the camshaft centerline then the shaft is "bent" and its "runout" you're seeing on the dial indicator and you'll find that the runout is decreased/increased if the indicator is moved to journals further from/closer to the "bend" itself.
And THAT is how out of round/concentricity/runout are measured in the "real world" by people who actually know what the hell they're doing, are doing it for a living, have to do it right and are in the business of USING PRECISION MEASURING TOOLS PROPERLY instead of the business of SELLING "precision" measuring tools to "machinists" who if they believe anything this guy says/demonstrates is "factual" and his processes are "correct" may be "making a living" as "machinists" but are highly unlikely to be making a GOOD LIVING or at least unlikely to be doing so on a "long-term basis". Because whether you believe it or not, this guy is just about as full of shit as they come on pretty much everything he demonstrates/claims and he may have "worked in" a machine shop once upon a time and may still have a "machine shop" he hawks his "precision tools" out of but he's sure as hell not in the "machine shop" business as his bread and butter or he and his employees wouldn't have the time and inclination to make all these "commercials" because they'd have work stacked up to last them months and months like every GOOD machinist/machine shop I've ever known/been in.
Anybody who continually reaches for a DIAL INDICATOR to "measure" things like "flatness" and "out of round" when anybody with half a clue knows that's straight-edge and micrometer territory and even uses the "wrong" dial indicator set-up at that (those lever-action indicator linkage contraptions "lose accuracy" as the "tip" moves through its arc because it MOVES in an arc while the "geometry" remains constant and they're only "perfectly accurate" at the "center" of their travel which is why they're only used by smart/good machinists/mechanics when nothing else will work and THEN only to measure MOVEMENT since we use a "sled gauge" with a conventional indicator in it to measure things like "liner protrusion" from diesel engine cylinder blocks and other "flat work" where other measuring tools just won't work. As for him running for the SURFACE GRINDER for every "flatness emergency" I MAY have been in ONE machine shop EVER that had a "surface grinder" and that would be Gessford Machine in Hastings, NE.
They can and do make/machine damned near everything and anything there and PROBABLY have one but I'll GUARANTEE IT AIN'T USED TO "MACHINE" SURFACES because SURFACE GRINDERS NEVER HAVE BEEN "MACHINE TOOLS" and were for SURFACE FINISHING ONLY to make a "rough" surface "smooth and shiny" OR a "smooth and shiny" surface "rough" AFTER it had been machined "flat" in a MILL or other actual MACHINE TOOL until the CURRENT CROP of online "expert machinists" and "wannabe machinists" who apparently "learn" from them came along and stared selling "surface grinding" as a way to make an "unflat" surface "flat" or "unparallel" surfaces "parallel" all the way down to MILLIONTHS OF AN INCH OF "FLATNESS".
Too bad there are "surface grinding professionals/shops that on their own websites speak the TRUTH about surface grinding and that's its to ALTER THE POST-MACHINING SURFACE FINISH AS CREATED BY THE MACHINING PROCESS/TOOL USED TO MACHINE IT and that AT BEST they can "hold" about .0015" of "flatness" across/along a 12" surface even they they "target" .001" and occasionally "luck out" and can "achieve" .0007" or so. You gotta be just pretty damned dense to not realize that since a GRINDING WHEEL is "shrinking" as it GRINDS its NOT a "precision cutter" and the MORE PASSES YOU MAKE REGARDLESS OF DIRECTION THE MORE WORN DOWN IT WILL GET AND THE MORE "unflat" the surface being GROUND will become.
But guys that seek to "learn" how to be "machinists" from RUclips VIDEOS and the "experts" who seek to SELL THEM TOOLS apparently don't/won't make that "mental leap" and would much rather use a "surface grinder" to make their "machining" look PRETTY even if we never actually SEE them doing the REAL "machining" and there's not a MILL TO BE SEEN or at least NOT VERY OFTEN in their "how-to videos". Which is PROBABLY because MILLS that are still "precise" and capable of making ANYTHING "flat" decades and decades after being built/rebuilt are as RARE AS HEN'S TEETH and those who HAVE THEM don't SELL THEM or at least not for what a "beginning machinist" who is just opening his own "machine shop" THINKS is a "fair price". Instead they go out and by some cheap POS they think they got a "good deal" on and when they "mill" anything it ends up being "less flat" and "rougher" than when they started. So THEN they RUN TO THE SURFACE GRINDER so there "wavy" so-called "flat" surfaces that they just can't find ANY "unflatness" in with their "dial indicator" that must be so flat its in the "millionths" because to the "beginning machinist" and the guy who wants to SELL HIM PRECISION TOOLS that's APPARENTLY the next "increment" after "ten thousandths" IF they even have an indicator that "precise" and are using/reading it "properly? (which is never the case if its on a magnetic base "stem" instead of in a SLED GAUGE). Measuring "flatness" and "out of round" with a dial indicator. I'm going to have to show these "how-to" videos to a few of my REAL machinist buddies. They're so damned busy/stressed they could use a laugh.
There IS one type of "dial indicator" that is commonly used to check bores for out of round and taper ahnd they're called "dial bore gauges" but they're precision instruments that basically combine an "inside micrometer" or telescoping gauge with a spring-loaded "plunger" and properly designed/constructed linkage so the "horizontal" measurements in the bores are displayed on a dial indicator "display" and the gauge is "set" to read "0" when installed in the "standard" factory-new bore diameter for that particular engine so that when the dial bore gauge is placed in a worn bore it can be "slid" up and down and around the cylinder bore quickly and easily to the same four points that would otherwise be measured with an inside micrometer or telescoping gauge/outside micrometer and the bore's out-of-round and taper measured/calculated in a few seconds instead of a few minutes.
DEEREMEYER1 jesus, youre whoring up the whole thread with bullshit. Concetricity has nothing to do with the radius error of a part, its used to specify axial congruence of features.
@@deeremeyer1749 Just make your own video and we can all have a good laugh.
out of round = roundness of 1 cylinderical surface (egg shaped, 3 point out of round from 3 jaw chuck, etc.)
not concentric = the various cylindrical surfaces on a crankshaft are intentionally not concentric. Those individual surfaces are hopefully not out of round.
in the real would its often a combination of both, depending on fabrication methods.
to the best of my knowledge the 'optimal' v angle for a true reading depends on the type of out of round you got. A 3 point out of round from the lathe chuck has the lowest points offset by 120 degrees, the points of contact will be different between 45/60/90 degree v-blocks.
as i asked :) MANY THANKS !!!!!!!!!!!!!!!!!
Well, id have to say the mystery remains unsolved in my eyes. I smell a part 2 in the future! ...at least I hope!
Thanks Don!
thanx don, again.Load on dial indicators is very important for precise reading from my point of view.plz correct me if i am wrong.
I think checking a shaft between centers, like what you did first, shows the concentricity between the center holes and the diameter where the indicator needle is. Measuring with a micrometer, 90° apart should show circularity (roundness).
It should but not always work that way
Measurement with a micrometer is necessary but not sufficient to measure roundness. For example if you measure the diameter of the rotor in a Wankel rotary engine you will get a peculiar result. The rotor is obviously not round but it has the same diameter no matter where you measure it with a micrometer. This condition is called lobing and can happen especially with centerless grinding. The work can have 3,5 or more odd number of lobes, though 3 is the most common. This lobing is probably in the millionths of an inch but can be critical for (as others have mentioned) things like automotive hydraulic valve lifters and fuel injector pintles.
@@aceroadholder2185 Whether or not the "lobes" are equally-spaced makes a big difference, too. Refer to tri-lobular screws, such as Taptite®, a completely different kettle of lobular fish - you gotta use both a regular mic and a tri-mic with a 60° V to take all three of its necessary diametral measurements. E-Z beans since it's invention in 1965, at least a couple billion of this family of screws have been reliably measured on the shop floor - one.pan.at.a.time - with oily, dirty hands and lots of cigarette smoke.
To quote Johnny Cash, "Huh - 1965. That's older even than this video."
There are different ways or geometries of out of round the cycloidal ones can pass cretin tests and still not be round. that is where the fun begins.
Roundness/Circularity would ideally be inspected at diametrically opposed points to so that we can determine the center. Unfortunately, nobody makes 180 degree vee blocks, so the 90 degree vee block represents the next best thing.
The problem with the 60 degree vee block is that it creates a larger space for eccentricity to 'hide' during an inspection for circularity. The 45 degree vee block closes that gap significantly whilst also making for easy and steady rotation.
to measure the roundness you actually would use two v-blocks, the one having a v as wide as possible the other as narrow as possible
an elliptical out of round will give a complete different measurement than a tri-lobed one in either of the blocks
by comparing the measurements one can determine the form of out of roundness and the amount of out of roundness
It's kinda like a man once told me, You're Trying To Pick the Fly Poop outa the Pepper. lol
Good point and nice video. Clamp some blocks next time so that the indicator doesn't jump that much. Peace.
It is possible to misadjust a centerless grinder so that it grinds a trilobular part. The only to find out is with a 60 degree v. They make micrometers with a 60 degree anvil for checking this.
+Richard Haisley You can see the error of a trilobular part in a 90 degree block too. Best method is to use two indicators measuring from different spots, so it really doesn't matter what the angle of the block is.
I think the 90 degree is more accurate, the reason being that it is closer to being orthogonal to the line the finger on the clock gauge moves.
Compelling. What kinds of "Out of round" occur most often?
All I've ever seen are ovals, D-shaped (Worn on one side) , egg shaped (worn over a wide angle), and the possibility of a rounded square. But that's just from crankshafts, camshafts, and reciprocating tools.
Hi Don! Thanks alot for your very informative videos.
I have one question though: Very often I see people using finger indicators, while they are not paying attention to have the finger in a 90° angle to the measurement direction.
I was under the impression, that this can cause significant error.. having a 40° angle causes 25% error... see "cosine error".. What do you think about this?
Cosine error isn't much of a problem when using a test indicator to make a comparative measurement. For instance: indicating you vise parallel to the table, indicating in a hole to check concentricity, or checking a shaft for run-out. However, if an absolute linear measurement is needed, cosine error can be a problem and may require a corrective multiplier. Using a "pear" shaped contact will eliminate the cosine error up to approx. 36 degrees. The other factor that can come in to play is replacing the contact point with a new point that is longer or shorter than the original, which will also affect your measurements.
It absolutely does and even when used "correctly" the additional linkage and "slop" still results in "inaccuracy" compare to a "standard" dial indicator. Which is why PROFESSIONALS only use "finger indicators" where it's impossible to USE a "standard" indicator setup. Of course we ALSO don't use a DIAL INDICATOR to measure ANYTHING BUT MOTION when its installed on "stem" and put an indicator in a SLED GAUGE if we want to do any "measuring" on a "flat surface" and THEN we're only using them to measure how far some other surface/part PROTRUDES ABOVE that surface such as when checking valve protrusion from a diesel engine cylinder head "fire deck" or liner protrusion above the cylinder block deck surface. For measuring FLATNESS we use a STRAIGHT EDGE AND FEELER GAUGE and if the "unflatness" is too "small" for a .001" feeler gauge to fit underneath an 18" straight-edge, that's a "perfect" block anyway. As for OUT OF ROUND that's MICROMETER TERRITORY whether bore ID or shaft, journal etc OD or dial bore gauge territory for most REAL MACHINISTS checking a block for bore size and to see if it will "clean up" in a hurry since GOOD MACHINISTS rarely have the kind of spare time all the supposed/aspiring "machinists" that make/haunt these RUclips "how-to" videos seem to have. I do not currently know and in 25 years of diesel-mechanicking have never known a GOOD machinist/machine shop that didn't have at LEAST several WEEKS of work piled up waiting and could work for more than maybe 20-30 minutes AT MOST and even THEN only rarely without having the phone ring or the door open with more work for him/it. Never known a good machinist/shop to have a SURFACE GRINDER to "machine" parts "flat" either and I THINK the only machine shop I've been in that HAS ONE AT ALL would be Gessford's in Hastings, NE. And if they DO I'll damned well guarantee it AIN'T THERE TO "MACHINE" anything. It's there for SURFACE FINISHING like ALL surface grinders owned/operated by good machinists/shops that know what the hell they're doing, know what surface grinders are actually for and have a use/need for them do. GRINDING A PART "FLAT" WITH A CONSTANTLY WEARING/SHRINKING WHEEL. Hilarious.
Very interesting & thanks for the stimulating video. Haven't totally wrapped my mind around the difference just yet, but it has to revolve around the tangent angles of contact. 60° block has ??; 90° v-block has 120° of separation; wouldn't any more obtuse angle more apparently show an out-of-round condition? To the maximum extent, which would be measuring from two tangent points which are diametrically opposed (I.e. - [180° of separation] - micrometer).
Great video. So, then 120deg. V-block would be even better? You knew someone was going to say it. :)
Consider two extreme examples: a 180 degree "V block" -- a flat surface -- and a 20 degree V block. When the piece is rotated on the flat surface the indicator will show 0.1000 change. When it's rotated in the 20 degree V block the top surface will show about 0.208 change.
The geometry of a 60 degree V block is such that the top surface of a perfect ellipse will show no vertical movement, irrespective of the rotation angle. So it's the worst possible choice.
The geometry of a 90 degree V block is such that the top surface of a perfect ellipse will show exactly half the difference of the two axes.
In both the 60 and 90 degree blocks the highest point on the top surface will move left and right (in the view on the drawing) as it's rotated. If the indicator is not moved so it stays on the highest point, it will show a deflection, but it will not be the same as with the 90 degree block, exactly as in your demonstration.
I really think it depends on how it's out of round. There are many ways to make shapes of constant width. If there where 3 high points would read very differently in a 60 degree v block than a 90 when compared to one with 4 high spots. I might do some drawings in fusion360 later to see if what I just said actually works like I think it would.
i think the 90 degree is better to use because it gives the biggest readout. as you guys as toolmakers always have tight tolerances you know you screwed it up when you got that far off your maximum drawing tolerance.
Since the egg is always trying to center itself in the Vee (left to right) A better way to inspect the part is from the side then, the angle doesn't pose a factor. In other words tip the V-block on its side and measure perpendicular to the side surface which is now down on the plate.
+AJKandK That is worth a try. I will do that when I have time to see the difference.
As a Cold Form Tooling Design Engineer, roundness in tooling isn't extremely critical especially with modern technology. 90% of our tooling designed is toleranced between .001 and .002 TIR. Very easily achieved considering almost all the tooling is CNC made, hardened, and then finish ground to spec. When it comes down to it, we are more concerned with concentricity. A die is made, pressed in a sleeve, and then the sleeve and die are pressed into a die case. With the die, sleeve, and die case all pressed together, it can be put into a lathe, indicated in on the od of the die case, which has been centerless ground to spec, and then observed for concentricity. The case can easily be indicated in to less then .0005. Now we can check the i.d. of the die pressed in the case. and that too will be within .0005 roundness with the whole assembly. As long as everything is concentric, it can be thrown in the machine and be just fine. In my opinion the act of putting the part, die, workpiece etc. on a rotating axis is the best way to measure for out of round and concentricity if required. You can easily find highs and lows. Im sure there are applications where roundness is super critical but not so much in the tool and die field I would say. But thats all based off of our applications. Very good video! Always enjoy watching them! :)
+Austin Washburn .001-.002 Thats lathes work.try + - .00005 like I did.
Uh, try measuring the diameter of the part at several points along the length and at two positions 180 degrees apart at each of those points. That's how we check "round" parts for "roundness" here in the real world when it comes to things like cylinder bores, crankshaft journals, bearing bores, etc. Runout and "out of round" are two entirely different things and out of round is determined by measuring IDs of bores, ODs of shafts, etc. And if you want to check RUNOUT you don't use a V-block long enough to drop the entire shaft in. You place a V-block at EACH END so you can rotate it and check the RUNOUT in the center where it will be MOST EVIDENT.
You also use a STANDARD DIAL INDICATOR with a "vertical plunger" placed "on top of" that shaft AT the center so runout is directly "measured" and indicated instead of through a "lever-action" mechanism that's typically used ONLY where a "standard" dial indicator can't be located DIRECTLY IN LINE WITH THE "TRAVEL" OF THE MOTION BEING MEASURED/INDICATED. Any And EVERY time you introduce "angles" into the "linkage" of a dial indicator plunger setup you're introducing error and inaccuracy. The tip of those "lever-action" indicator setups moves in an ARC and not "vertically" and its "geometry" relative to the plunger/rack-and-pinion of the indicator itself is constantly changing AS it swings through that arc.
DEEREMEYER1 internet genius, whats the cosine error at 15° for .002? Ill fill you in, its less than a 10th. On a test indicator which has virtually no travel, a .030 sweep has a total sector angle of 2.6°, the cosine error from one end to the other is +-15 millionths, there arent many drop indicators that can do that. As far as "run out" there are 2 types, circular runout or out of round, and total. Circular only is concerned with circularity at any single cross section. Total is the compound of circularity, straightness, angularity and cylindricity.
So these lever type indicators are incorrect? You should explain that to Mitutoyo, Starret, Tesa and all other high end test indicator manufacturers. These are the same style indicators that machine manufactures and certified repair companies use to set up machines down to the half micron.
,,, I think a mic would have shown close to 2 thou, but I would have liked to see him mic it .
DEERMYERER1 its quite obvious you're not a "real" machinist. If you paid attention to what Don said at the beginning you would know the difference, moron
@@bernardshrewsbury
That's a good one lol
Interesting. Looks like a 120 degree block would be even better than the 90.
The most important fact in which measure is "true" have shape of cirualrity - how many lobbing we have around the part. In my industrial practic on part with diameter above 250 up to 700 [mm] ( in mm because im from Poland) most comon are 3 point out of round - reason is technologu turning in 3 jaw chucks,
In your case you have elliptical shape in my opinion and it's no matter which v-block you will use 60 or 90deg both result are correct but you have to correct read it including geometrical issues.
here you have little bit more about it what-when-how.com/metrology/measurement-of-circularity-metrology/
Realy intresting video, sorry for writting mistakes and all the best from Poland
I think a 170* would be even better. The reason is pretty simply, you are making the out of round condition visible along the axis it is measuring on. Oh wait... It doesnt make a difference. The math is the same. ;-)
Great video'
I would have used two of each kind of block. Using just one seems wrong.