Knowing if your parts are round is sometimes extremely critical. If on two mating parts you feel a clunk when you rotate them, you know one thing for sure. Watch and see.
Every "would-be" mechanical engineer should be required to watch videos like these! I have witnessed countless hours of debate along these lines in factories and meeting rooms with suppliers across this country. Unfortunately common sense did not prevail in half of the situations. Your approach of "exaggeration for illustration" is spot on. The only way that I find that I can trigger their minds into action is with leading, and pointed, questions with illustration. The notion of understanding has to come from within their own minds- not yours. Summary statistics can be debated and "method-error" quoted to death in a crowd. A simple illustration and/or tactile hands on example, however, has the power to influence thought in a primordial level. The mind triggers on questions, soooo ask the questions :-). As a designer, and lifelong quality professional, I really appreciate your content and mastery of conveyance. Thanks again for putting all this amazing content out into the world!
Don thank you for the video. It shows a great practal way of checking for roundness. Obviously between centres is the most accurate but if you use a V block you need to be aware that if the part is out of round then as you rotate it the centre height of the part will rise and fall. The narrower the included angle of the V block the less it will rise and fall, as you demonstrated. Between centres you had 0.002", 45 degree 0.0018' and 60 degree 0.0010". Basically the narrower the included angle the further apart the tangent points and the less the rise and fall of the overall height of the part being tested.
+John Linquist Checking between centers is only possible if your centers are actually in the center. Many mold components are manufactured with center holes in them (to support in a lathe), but the actual grinding is done with a centerless grinder, thus the centers are not concentric with the outside diameter. This is usually told in the catalogs.
I did too George. This was just for demonstration purposes. I actually ground on a Landis Center type grinder, the eight roller bearings that went into our first satellite launched by the United States Air Force. It was called the Early Bird Satellite. I believe it was in 1963.
Very well explained. Thanks....so many people explain things in a way that doesn't make sense (either because they are just poor at teaching the concept, or ....(more often then we would like to admit) because of the fact that they don't have a solid understanding themselves....thanks again!
thanks for the video, learned something new today. I'm a machinist but like beginner. I work at tilite and I'm doing alot of fixturing block. so I keep an look out on ur videos to learn more. thanks.
True as far as it goes. 1971 I went to work for a shop that made hydraulic valve lifters for Chrysler, Ford, and some others. We used centerless grinders to finish grind and if they were used incorrectly we could get multi lobe out of round. The check was to use V blocks with multiple angles. Some angles would show out of round and others would show perfect. We also had to pass a leak down test for each lifter which meant the inner part could not be forced down too fast or too slow with oil inside the body. Talk about fine tolerance. If you saw how these worked, you would change your oil per schedule as dirt would damage the function.
I was waiting for you to check the i.d. of the mating part. Those are a bit more difficult measurements to make! I was once asked what it means when a rotating cam tightens up and then loosens. I replied that the cam was bent, but of course it also means the cam journals are also not in line.
I've found some TG&P shaft material that was triangularly out of round.. check it with a regular micrometer and it seems perfectly round, but chuck it in a lathe or a 3 point micrometer and you see three highs and three lows by (typically) about .0015 inch,. putting it on a v-block or straightening rollers (because of 3 point contact of this setup) it shows (typically) more like .003 out of round. Such material, we have returned to the vender... it kind of pisses them off a few times because we cut a piece out of their 20' stick of material before finding the defect. They had probably wanted to re sell it to someone else.
+tracyguilbeau That's an indication of poor quality control in centerless grinding if the end result is 3-lobed pieces sold to customers. I would change vendors if it happens consistently.
So slightly out of round parts, especially those that rotate inside something, can contribute to vibration when rotating, even if only a little. This could affect lifetime and efficiency of the machine, even if it's not a large vibration. Reminds me of a ball bearing demonstration elsewhere on TH-cam.
The only true way to check for out of roundness is with micrometers. I've seen many people test shafts the way you did and claim the shaft was out of round when in reality it was bent. I've also seen folks claim a shaft was bent when it was actually out of round. The test you performed, along with OD mic readings taken in at least three locations around a circumference will provide you with accurate information.
Don, Thanks for all the great videos. I try not to post comments unless I have something valuable to share, but... I have an issue with how you're checking the smaller diameter of the shaft towards the end of the video. Checking it as you did in the video, lets say you did measure a deviation of a .0002 or something, would that indicate a roundness or a concentricity error? By supporting the shaft on the large diameter, and checking the smaller diameter, you really wouldn't be able to tell would you? I would support the diameter that I wanted to check for roundness in the V-Block I see this happen sometimes when people put a shaft between centers on a set of Bench centers. Then they start checking different diameters and making statements about concentricity, roundness, etc without realizing that they're actually measuring both at the same time and you can't say with certainty which condition is causing the error readings. Really, what needs to be done is to check each diameter independently for roundness in a V-Block, and define the roundness of each diameter first. Then put the shaft between centers and check for concentricity. I always say inspection is really a process of elimination. In this case, first define errors due to roundness. Then any additional errors must be due to concentricity. Thanks for doing these videos, lots of people, including me, are learning things here (which is more than I can say for the rest of the internet!). John
+Offyguy Yup, you are spot on. By checking it like in this video you can say it is round and concentric (special case, no errors in either), but if there is an error you can't differentiate which it is. Unless of course you attach an angle measurement device to the shaft and plot the measurements to see if it is round.
@@JaakkoF Strictly speaking there can be cases were the roundness and concentricity errors cancel each other out, and you won't notice that the part is out of spec.
He did check the large diameter first. It was perfect. Therefore going on to check small diameter, if there's a roundness /concentric error it will show up. This was a quick check. If there was an error, then yes you would have to determine if it was one of concentricity or roundness. But they will NOT cancel each other out.
PS. If there was a roundness error when checking the large diameter. Then yes it could cancel out a roundness error on small Dia. But still won't cancel out a concentric error
Out of roundness is usually and commonly caused by bad centers where the guy on the lathe just jams the center drill in as fast as he can and dose,nt let it finish the center... We used to regrind the centers to get perfect roundness when grinding.
try putting it in a 60deg included angle v-block and see then that is what helps show out of round a 90deg one really wont show you anything as far as out of round goes iv been told. i could be wrong iv never tried but as far as i know thats the way to check.
+turboed350 Why repeat the same old story if you even don't know if it holds any truth? It is like the story of a mother cutting the ends off of a pot roast.
to test out of round you'll want to have minimal degrees to the linear axis, technically speaking a 1 degree v or a 179 degree v will show you the best result, where a 90 degree v will be the most inaccurate, 0 / 180 degrees will be the most accurate, however you can't hold the part stationary while rotating in a v with no angle ^^
You check for out of round by measuring the OD of the shaft across the shaft in two locations 180 degrees apart. All this demonstration proves is that the shaft has no "runout" and typically "out of round" results from wear and that "shaft" could have out of round at some point within the length resting on or in the V-block and still have zero runout. and zero out of round at the end.
A part could be round ,but checking off the first part could shoe that the second part is not concentric withe the first dia, thereby making the second dia look like its out of round.
+George Lee I saw the same thing. When he checked the smaller diameter, he was really looking at concentricity relative to the major DIA in the v block and not actual roundness.
Where I worked for Vickers in the UK they used a Tallyrond machine to check roundness, highly accurate machine.I have never seen one in the 30 yrs I worked in the USA.
I was the gage lab supervisor for a company that made hydraulic valve lifters for Chrysler, Cat, Cummings and others. Each lifter had to pass a leakdown test for function. Lots of bad things happen including out of round. To test for out of round , a V block was used as shown but it is necessary to test using several angles of V because it might test ok with one, but off with another. Most of the OD `s were finish ground on centers grinders. Details are forgotten as this was 50 years ago. I did love that job. In retirement I do woodworking and try to work to the same standards, impossible but something to aspire.
That's all well and good if you can rotate one part inside another. But that doesn't really prove nothing is "out of round" simply because there is no runout on a roating shaft nor is it common to check a plain round shaft or bore where two parts are a "slip fit" for out-of-round. Where out of round is checked is camshaft and crankshaft bearing bores, pistons and cylinders, etc where because parts expand and contract as the heat up and cool and loads are variable and are "radial" as well as "axial" and one part is stationary while the other is in motion out-of-round parts or "holes" cause multiple problems. In those situations TAPER is checked for the same reason and a dial indicator and V-block(s) are useless. You have to use a micrometer and measure ODs and IDs at two points "180 degrees apart" and if there is variation then you know you have an actual "out of round" condition. The same process would work in your demonstration case. And since most "out of round" conditions result from WEAR and material that isn't where it's supposed to be instead of material where it shouldn't be, checking for "out of round" by rotating a shaft in a single V-block" with no "slapping" of the shaft against the block doesn't mean that shaft ISN'T "out of round". A crankshaft connecting rod journal can be badly out of round to the point the out of roundness can be felt with a bare hand and when "measuring" with an index finger and thumb but with a connecting rod and bearing on that journal rotating it around the journal by hand trying to feel that "slap" it can be impossible to "feel" and the "slap" will only be heard/felt with the engine assembled/running under load or even when lugged heavily in a non-typical operating condition. In fact that's what typically CAUSES an out of round situation like that. And without the crankshaft in a crank grinder and set up so the "centerline" of rotation is that connecting rod journal so an indicator can be placed on the journal surface you'll never see "out of round" with a dial indicator on an offset journal or other bearing surface like that. You'll find that out of round in no time as you go from journal to journal with a micrometer checking each journal for out of round and taper by making 4 or 8 simple, quick measurements at each journal (shared rod journals such as those on V-8 crankshafts are treated as separate journals for "blueprinting" purposes when measuring them for bearing clearance, taper and out of round) and you'll find out of round cylinder bores and pistons and bearing bores the same way. Even camshaft lobe base circle diameter, out of round and taper as well as "lift" are more easily, accurately and precisely measured with a micrometer than a dial indicator. Particularly since flat-tappet camshaft lobes have taper machined into them to cause lifter rotation so the "foot" of the flat-tappet lifter and lifter body itself are "spinning" along the camshaft lobe surface instead of sliding and scraping along it. Those "flat" lifter bases are slightly convex and not flat at all for the same reason. What can you RELIABLY measure with a dial indicator and V-block(s) when it comes to shafts across the board but that is also easily visible to the naked eye either by observable "wobble" or wear patterns/damage to bearings and other parts, etc? Runout. And a shaft can still have runout and be bent while not being "out of round" and be out of round and have no runout. As for that shaft being so perfectly machined the indicator doesn't show even 50 millionths of "out of round" (aka runout) I'd love to see the lathe that produced such perfect concentricity while leaving such a "rough" surface finish.
In this case, the second test is no proof of roundness, it could also be showing whether the two parts are concentric to each other or not. if both diameters were round but not concentric the gage would deflect by that amount. The best way to test for roundness is to put the part being tested in contact with the vee block.
As always Don a very interesting video. But I'd sure like to see who did the part and how they got zero movement on a 10ths reading indicator. That's more than impressive to me at least. I can't think of anything I own right now including some pretty expensive tooling that doesn't show at least a 10th or 2 of needle deflection with a good indicator. Can that type of accuracy even be obtained with roller bearings, or are solid bronze type bearings still needed to produce those accuracy's?
+Turning Point That can be achieved with either bearing type if your bearings are in good working condition. Usually the "dead indicator" situation is only achieved with ground or lapped surfaces, as many machining methods such as lathe work produce a very fine pattern on the workpiece that the indicators picks up.
+Turning Point A sensitive indicator will pick up the surface texture of a part as little needle flicks, it's the longer cyclical indications that you have to watch out for, they will tell you what you need to know.
Hi Don, love your videos but I do have a question on this particular one. When you had the shaft on the V-block and measured the smaller O.D on the shaft using the larger O.D, isn't that measuring concentricity, not roundness??
+Paulo Esquenazi Thanks for watching Paulo. In regards to your question, we already knew that the major diameter was round. So, being that it was round and using it as a surface to check with; we would then be able to determine if the minor diameter was either out of round OR not concentric.
Thanks for the quick response and informative videos. While I still see it as concentricity, I agree that given the right tolerance call out, the larger O.D can be used used as the datum to check roundness on the smaller O.D
Sorry but you cannot check for OOR on the O.D. with a 90 degree v block .! won't tell you if you have 3 point lobing! It can be greatly out of round with 3 point lobing and you may never see it. Check it out.
My brother was engineer at a motor mfg. They were having problems with noise in motors and couldn't figure it out. He found exactly what your saying. 3pt lobes and they had indicator on top like in video. When he put the indicator on the side at 90 degrees you could see it. Never heard why the shafts were like that. Must have had something to do with how they were centerless grinding them.
I read this interesting info from the link. I asked brother about it and now the story is they had to use a 60 degree v-block. I think he told me wrong the first time but it was probably 15 years ago.
...In high school...machine shop...a slip-fit like this was maybe..3-5 'tenths'...which would be 1/10th of 1/1000....that 50 millionths he mentioned here is a 1/2 of a 'tenth'....slightly more than 1 micron.... I was in high school in early 1960's....I don't think they teach this stuff in HS anymore...maybe college?
Maybe this is a silly question but wouldn't you need to check the inner diameter of the part used in this slip fit to check for roundness? Perhaps by placing the indicator on the inner diameter and spinning the part?
Whoa whoa, you are checking total run out with that v block right? You may be getting roundness for free but, There could be a Bow in the cylinder to move that indicator but it could still technically be round huh? Best way (and easy way) to check roundness by hand is with a micrometer for an o.d, and a bore gauge for an i.d.
+john Doe A micrometer won't catch lobe shaped out of roundness such as you see when drilling thin sheet metal with an ordinary twist drill. Google "lobe shape constant diameter" to see more examples.
+john Doe If there is a bow, it can be measured to exist and thn use other methods to check the roundness. Measuring with a micrometer doesn't confirm roundness.
intresting to concider however not very helpfull in the way that if you have a shaft fitting in a hole the hole or the shaft could be out of round and without messuring you would not know by hand. if both were yes you would feel it
+bcbloc02 Consistent diameter is one thing, roundness is different. You can have a part that measures the same in multiple locations with a mic, but not round. And as a general rule, you can't test for roundness with any two-point measuring system, such as micrometers, calipers and so on.
voron11000 While theoretically you could measure a hexagon 3 times and come up with the same diameter I think in practice odds are you would not be in the exact right spots for that to happen and you would indicate a different diameter and thus know part is out of round.
+bcbloc02 Yes, hexagon is one of the grossly exaggerated examples the author of the channel likes so much. But practically you don't check a hexagon for roundness, right? Practically you'd have a little bump on the surface. Or a little pit. Or a combination of a bump on one side and a pit on the other which would give the reading you expect on a mic. But still not round. Let's just talk theory. A circle is a shape when the distance from all the points to the center is the same. But you can't check for that directly, you're not measuring from the center, right? Moreover, in theory, you can have such a shape so all the distances between diametrically opposite points are the same, but the distance to the center varies, so it's not a circle. Of course I should use the terms "diametrically" and "center" in quotes. :-) So, like you said, you indicate it in a Vee-block, or spin it in some kind of accurate spindle, and it makes out-of-roundness obvious.
+bcbloc02 If you ever do centerless grinding or 3-jaw chuck work and start checking parts with a micrometer around the part and say that is round, you will start wondering why your customer send the parts back as they are not round. If your workpiece is lobed (which a centerless grinder can do), it measures the same all around but it is not round. The 3-lobed shape is called Realeux triangle, but there canbe 5-lobed also, it depends on the manufacturing method. So, consistent diameter is not an indication of roundness.
+Chris Anderson Check the OD in the V, if OD is round, then use the same indicator placed inside for the ID in the same setup.... Hope I made it clear... ;)
Every "would-be" mechanical engineer should be required to watch videos like these! I have witnessed countless hours of debate along these lines in factories and meeting rooms with suppliers across this country. Unfortunately common sense did not prevail in half of the situations. Your approach of "exaggeration for illustration" is spot on. The only way that I find that I can trigger their minds into action is with leading, and pointed, questions with illustration. The notion of understanding has to come from within their own minds- not yours. Summary statistics can be debated and "method-error" quoted to death in a crowd. A simple illustration and/or tactile hands on example, however, has the power to influence thought in a primordial level. The mind triggers on questions, soooo ask the questions :-). As a designer, and lifelong quality professional, I really appreciate your content and mastery of conveyance. Thanks again for putting all this amazing content out into the world!
+Doug P Doug you keep up your good work too.
Oh, another designer who thinks he knows better than an engineer. You're a rare breed. *eye roll*
Don thank you for the video. It shows a great practal way of checking for roundness. Obviously between centres is the most accurate but if you use a V block you need to be aware that if the part is out of round then as you rotate it the centre height of the part will rise and fall. The narrower the included angle of the V block the less it will rise and fall, as you demonstrated. Between centres you had 0.002", 45 degree 0.0018' and 60 degree 0.0010". Basically the narrower the included angle the further apart the tangent points and the less the rise and fall of the overall height of the part being tested.
+John Linquist Checking between centers is only possible if your centers are actually in the center. Many mold components are manufactured with center holes in them (to support in a lathe), but the actual grinding is done with a centerless grinder, thus the centers are not concentric with the outside diameter. This is usually told in the catalogs.
I did too George. This was just for demonstration purposes. I actually
ground on a Landis Center type grinder, the eight roller bearings that
went into our first satellite launched by the United States Air Force.
It was called the Early Bird Satellite. I believe it was in 1963.
I like Don Bailey's videos. He is good at explaining what is going on and why. Thanks Don!
Very well explained. Thanks....so many people explain things in a way that doesn't make sense (either because they are just poor at teaching the concept, or ....(more often then we would like to admit) because of the fact that they don't have a solid understanding themselves....thanks again!
You're welcome! Glad I could help.
Great video and tx for sharing. I am Qc/Qa inspector, this is the old way to check roundness on the shafts tubes pipe and cylinder
thanks for the video, learned something new today. I'm a machinist but like beginner. I work at tilite and I'm doing alot of fixturing block. so I keep an look out on ur videos to learn more. thanks.
You're welcome, glad we are able to help.
Thank you. I didn't know that one part could be out of round and not be feeling it.
True as far as it goes. 1971 I went to work for a shop that made hydraulic valve lifters for Chrysler, Ford, and some others. We used centerless grinders to finish grind and if they were used incorrectly we could get multi lobe out of round. The check was to use V blocks with multiple angles. Some angles would show out of round and others would show perfect.
We also had to pass a leak down test for each lifter which meant the inner part could not be forced down too fast or too slow with oil inside the body. Talk about fine tolerance.
If you saw how these worked, you would change your oil per schedule as dirt would damage the function.
I was waiting for you to check the i.d. of the mating part. Those are a bit more difficult measurements to make! I was once asked what it means when a rotating cam tightens up and then loosens. I replied that the cam was bent, but of course it also means the cam journals are also not in line.
I've found some TG&P shaft material that was triangularly out of round.. check it with a regular micrometer and it seems perfectly round, but chuck it in a lathe or a 3 point micrometer and you see three highs and three lows by (typically) about .0015 inch,. putting it on a v-block or straightening rollers (because of 3 point contact of this setup) it shows (typically) more like .003 out of round. Such material, we have returned to the vender... it kind of pisses them off a few times because we cut a piece out of their 20' stick of material before finding the defect. They had probably wanted to re sell it to someone else.
+tracyguilbeau That's an indication of poor quality control in centerless grinding if the end result is 3-lobed pieces sold to customers. I would change vendors if it happens consistently.
So slightly out of round parts, especially those that rotate inside something, can contribute to vibration when rotating, even if only a little. This could affect lifetime and efficiency of the machine, even if it's not a large vibration. Reminds me of a ball bearing demonstration elsewhere on TH-cam.
Thank you Mr. Baily!
If you really want to check for roundness and concentricity get a Talyrond m/c
Excellent info old school always best
Love all the videos. You're never too old to learn anything. What surprised me is the haters that comment
The only true way to check for out of roundness is with micrometers. I've seen many people test shafts the way you did and claim the shaft was out of round when in reality it was bent. I've also seen folks claim a shaft was bent when it was actually out of round. The test you performed, along with OD mic readings taken in at least three locations around a circumference will provide you with accurate information.
Don, Thanks for all the great videos. I try not to post comments unless I have something valuable to share, but... I have an issue with how you're checking the smaller diameter of the shaft towards the end of the video. Checking it as you did in the video, lets say you did measure a deviation of a .0002 or something, would that indicate a roundness or a concentricity error? By supporting the shaft on the large diameter, and checking the smaller diameter, you really wouldn't be able to tell would you? I would support the diameter that I wanted to check for roundness in the V-Block
I see this happen sometimes when people put a shaft between centers on a set of Bench centers. Then they start checking different diameters and making statements about concentricity, roundness, etc without realizing that they're actually measuring both at the same time and you can't say with certainty which condition is causing the error readings. Really, what needs to be done is to check each diameter independently for roundness in a V-Block, and define the roundness of each diameter first. Then put the shaft between centers and check for concentricity.
I always say inspection is really a process of elimination. In this case, first define errors due to roundness. Then any additional errors must be due to concentricity.
Thanks for doing these videos, lots of people, including me, are learning things here (which is more than I can say for the rest of the internet!).
John
+Offyguy Yup, you are spot on. By checking it like in this video you can say it is round and concentric (special case, no errors in either), but if there is an error you can't differentiate which it is. Unless of course you attach an angle measurement device to the shaft and plot the measurements to see if it is round.
@@JaakkoF Strictly speaking there can be cases were the roundness and concentricity errors cancel each other out, and you won't notice that the part is out of spec.
He did check the large diameter first. It was perfect. Therefore going on to check small diameter, if there's a roundness /concentric error it will show up. This was a quick check. If there was an error, then yes you would have to determine if it was one of concentricity or roundness. But they will NOT cancel each other out.
PS. If there was a roundness error when checking the large diameter. Then yes it could cancel out a roundness error on small Dia. But still won't cancel out a concentric error
Good Video!! That really makes sense. I never really thought about that but it really is good to know.
Out of roundness is usually and commonly caused by bad centers where the guy on the lathe just jams the center drill in as fast as he can and dose,nt let it finish the center... We used to regrind the centers to get perfect roundness when grinding.
+George Lee ...Couple other guys have videos on You tube showing how to re-grind a center OD and ID...MrPete and Abom79....
Thank you very much for the info.
Simply awesome! Thanks for sharing don :)
Thank you don Bailey
I love these videos
try putting it in a 60deg included angle v-block and see then that is what helps show out of round a 90deg one really wont show you anything as far as out of round goes iv been told. i could be wrong iv never tried but as far as i know thats the way to check.
+turboed350 Watch next weeks video. We did a test.
+turboed350 Why repeat the same old story if you even don't know if it holds any truth? It is like the story of a mother cutting the ends off of a pot roast.
to test out of round you'll want to have minimal degrees to the linear axis, technically speaking a 1 degree v or a 179 degree v will show you the best result, where a 90 degree v will be the most inaccurate, 0 / 180 degrees will be the most accurate, however you can't hold the part stationary while rotating in a v with no angle ^^
You check for out of round by measuring the OD of the shaft across the shaft in two locations 180 degrees apart. All this demonstration proves is that the shaft has no "runout" and typically "out of round" results from wear and that "shaft" could have out of round at some point within the length resting on or in the V-block and still have zero runout. and zero out of round at the end.
A part could be round ,but checking off the first part could shoe that the second part is not concentric withe the first dia, thereby making the second dia look like its out of round.
+George Lee I saw the same thing. When he checked the smaller diameter, he was really looking at concentricity relative to the major DIA in the v block and not actual roundness.
Where I worked for Vickers in the UK they used a Tallyrond machine to check roundness, highly accurate machine.I have never seen one in the 30 yrs I worked in the USA.
Correct. Also the vee is a full contact and if the shaft had issue with straightness,, say no more.
I was the gage lab supervisor for a company that made hydraulic valve lifters for Chrysler, Cat, Cummings and others.
Each lifter had to pass a leakdown test for function. Lots of bad things happen including out of round. To test for out of round , a V block was used as shown but it is necessary to test using several angles of V because it might test ok with one, but off with another. Most of the OD `s were finish ground on centers grinders. Details are forgotten as this was 50 years ago. I did love that job.
In retirement I do woodworking and try to work to the same standards, impossible but something to aspire.
That's all well and good if you can rotate one part inside another. But that doesn't really prove nothing is "out of round" simply because there is no runout on a roating shaft nor is it common to check a plain round shaft or bore where two parts are a "slip fit" for out-of-round. Where out of round is checked is camshaft and crankshaft bearing bores, pistons and cylinders, etc where because parts expand and contract as the heat up and cool and loads are variable and are "radial" as well as "axial" and one part is stationary while the other is in motion out-of-round parts or "holes" cause multiple problems. In those situations TAPER is checked for the same reason and a dial indicator and V-block(s) are useless. You have to use a micrometer and measure ODs and IDs at two points "180 degrees apart" and if there is variation then you know you have an actual "out of round" condition. The same process would work in your demonstration case. And since most "out of round" conditions result from WEAR and material that isn't where it's supposed to be instead of material where it shouldn't be, checking for "out of round" by rotating a shaft in a single V-block" with no "slapping" of the shaft against the block doesn't mean that shaft ISN'T "out of round". A crankshaft connecting rod journal can be badly out of round to the point the out of roundness can be felt with a bare hand and when "measuring" with an index finger and thumb but with a connecting rod and bearing on that journal rotating it around the journal by hand trying to feel that "slap" it can be impossible to "feel" and the "slap" will only be heard/felt with the engine assembled/running under load or even when lugged heavily in a non-typical operating condition. In fact that's what typically CAUSES an out of round situation like that. And without the crankshaft in a crank grinder and set up so the "centerline" of rotation is that connecting rod journal so an indicator can be placed on the journal surface you'll never see "out of round" with a dial indicator on an offset journal or other bearing surface like that. You'll find that out of round in no time as you go from journal to journal with a micrometer checking each journal for out of round and taper by making 4 or 8 simple, quick measurements at each journal (shared rod journals such as those on V-8 crankshafts are treated as separate journals for "blueprinting" purposes when measuring them for bearing clearance, taper and out of round) and you'll find out of round cylinder bores and pistons and bearing bores the same way. Even camshaft lobe base circle diameter, out of round and taper as well as "lift" are more easily, accurately and precisely measured with a micrometer than a dial indicator. Particularly since flat-tappet camshaft lobes have taper machined into them to cause lifter rotation so the "foot" of the flat-tappet lifter and lifter body itself are "spinning" along the camshaft lobe surface instead of sliding and scraping along it. Those "flat" lifter bases are slightly convex and not flat at all for the same reason. What can you RELIABLY measure with a dial indicator and V-block(s) when it comes to shafts across the board but that is also easily visible to the naked eye either by observable "wobble" or wear patterns/damage to bearings and other parts, etc? Runout. And a shaft can still have runout and be bent while not being "out of round" and be out of round and have no runout. As for that shaft being so perfectly machined the indicator doesn't show even 50 millionths of "out of round" (aka runout) I'd love to see the lathe that produced such perfect concentricity while leaving such a "rough" surface finish.
Very interesting. Thank you.
+Ralph Dratman Thanks for watching.
+SuburbanTool Inc The way the v-block works is pretty impressive. Does your company make those?
+Ralph Dratman Yes, www.subtool.com
+Ralph Dratman if you need on or two email me.
In this case, the second test is no proof of roundness, it could also be showing whether the two parts are concentric to each other or not. if both diameters were round but not concentric the gage would deflect by that amount. The best way to test for roundness is to put the part being tested in contact with the vee block.
Good tip,..thanks!!
Another good video Don. Could you use the same setup to check the ID ( v block and height gauge)?
all good
Great Man! Saludos desde Argentina!
Hello from the USA! Thanks again for watching.
As always Don a very interesting video. But I'd sure like to see who did the part and how they got zero movement on a 10ths reading indicator. That's more than impressive to me at least. I can't think of anything I own right now including some pretty expensive tooling that doesn't show at least a 10th or 2 of needle deflection with a good indicator. Can that type of accuracy even be obtained with roller bearings, or are solid bronze type bearings still needed to produce those accuracy's?
+Turning Point That can be achieved with either bearing type if your bearings are in good working condition. Usually the "dead indicator" situation is only achieved with ground or lapped surfaces, as many machining methods such as lathe work produce a very fine pattern on the workpiece that the indicators picks up.
Many thanks Jaakko, your also another one on TH-cam who I very much respect for you rknowledge and experience.
+Turning Point A sensitive indicator will pick up the surface texture of a part as little needle flicks, it's the longer cyclical indications that you have to watch out for, they will tell you what you need to know.
You cannot check roundness with a 90 degree included angle v-block.
You need a 60 degree included angle v-block for that.
+Greg Kozak Watch next weeks video. We did a test.
+Greg Kozak Of course you can check it, it doesn't matter which angle you have in your block if you just use two indicators in different spots.
+Jaakko Fagerlund
Jaakko, You absolutely correct. I now see the error of my ways.
Thank you, Greg.
Nice Job Guys
Hi Don, love your videos but I do have a question on this particular one. When you had the shaft on the V-block and measured the smaller O.D on the shaft using the larger O.D, isn't that measuring concentricity, not roundness??
+Paulo Esquenazi Thanks for watching Paulo. In regards to your question, we already knew that the major diameter was round. So, being that it was round and using it as a surface to check with; we would then be able to determine if the minor diameter was either out of round OR not concentric.
Thanks for the quick response and informative videos. While I still see it as concentricity, I agree that given the right tolerance call out, the larger O.D can be used used as the datum to check roundness on the smaller O.D
Nice video thanks.
You're welcome, thanks for leaving a comment!
Sorry but you cannot check for OOR on the O.D. with a 90 degree v block .! won't tell you if you have 3 point lobing! It can be greatly out of round with 3 point lobing and you may never see it. Check it out.
My brother was engineer at a motor mfg. They were having problems with noise in motors and couldn't figure it out. He found exactly what your saying. 3pt lobes and they had indicator on top like in video. When he put the indicator on the side at 90 degrees you could see it. Never heard why the shafts were like that. Must have had something to do with how they were centerless grinding them.
pmpaspeakingofprecision.files.wordpress.com/2010/09/lobing-015-cropadj.jpg?w=675&h=648
I read this interesting info from the link. I asked brother about it and now the story is they had to use a 60 degree v-block. I think he told me wrong the first time but it was probably 15 years ago.
Big fan btw.
...In high school...machine shop...a slip-fit like this was maybe..3-5 'tenths'...which would be 1/10th of 1/1000....that 50 millionths he mentioned here is a 1/2 of a 'tenth'....slightly more than 1 micron....
I was in high school in early 1960's....I don't think they teach this stuff in HS anymore...maybe college?
what is that gauge that he's using to touch the cylinder while turning it in the v block ?
its a dial indicator, you can find it on www.subtool.com
Maybe this is a silly question but wouldn't you need to check the inner diameter of the part used in this slip fit to check for roundness? Perhaps by placing the indicator on the inner diameter and spinning the part?
+John Lane If you use an indicator, What you're doing is measuring concentriciy to the larger diameter. This is assuming you're using a V-block
Whoa whoa, you are checking total run out with that v block right? You may be getting roundness for free but, There could be a Bow in the cylinder to move that indicator but it could still technically be round huh? Best way (and easy way) to check roundness by hand is with a micrometer for an o.d, and a bore gauge for an i.d.
+john Doe A micrometer won't catch lobe shaped out of roundness such as you see when drilling thin sheet metal with an ordinary twist drill. Google "lobe shape constant diameter" to see more examples.
+john Doe If there is a bow, it can be measured to exist and thn use other methods to check the roundness. Measuring with a micrometer doesn't confirm roundness.
intresting to concider however not very helpfull in the way that if you have a shaft fitting in a hole the hole or the shaft could be out of round and without messuring you would not know by hand. if both were yes you would feel it
I know this video is 3 years old, but how do you check if a round part is "out of round" if you cant fit it in a V-block or spin it like you did?
the other way would be to put the part between centers, and i am assuming that the parts does have centers.
You forgot to mention the most common practice of using a mic in multiple locations to verify consistent diameter.
+bcbloc02 Consistent diameter is one thing, roundness is different. You can have a part that measures the same in multiple locations with a mic, but not round. And as a general rule, you can't test for roundness with any two-point measuring system, such as micrometers, calipers and so on.
voron11000
While theoretically you could measure a hexagon 3 times and come up with the same diameter I think in practice odds are you would not be in the exact right spots for that to happen and you would indicate a different diameter and thus know part is out of round.
+bcbloc02 Yes, hexagon is one of the grossly exaggerated examples the author of the channel likes so much. But practically you don't check a hexagon for roundness, right? Practically you'd have a little bump on the surface. Or a little pit. Or a combination of a bump on one side and a pit on the other which would give the reading you expect on a mic. But still not round. Let's just talk theory. A circle is a shape when the distance from all the points to the center is the same. But you can't check for that directly, you're not measuring from the center, right? Moreover, in theory, you can have such a shape so all the distances between diametrically opposite points are the same, but the distance to the center varies, so it's not a circle. Of course I should use the terms "diametrically" and "center" in quotes. :-) So, like you said, you indicate it in a Vee-block, or spin it in some kind of accurate spindle, and it makes out-of-roundness obvious.
+bcbloc02 If you ever do centerless grinding or 3-jaw chuck work and start checking parts with a micrometer around the part and say that is round, you will start wondering why your customer send the parts back as they are not round. If your workpiece is lobed (which a centerless grinder can do), it measures the same all around but it is not round. The 3-lobed shape is called Realeux triangle, but there canbe 5-lobed also, it depends on the manufacturing method.
So, consistent diameter is not an indication of roundness.
+bcbloc02 mics are nor accurate enough in REAL precision grinding.OK down to a few tenths. depending on how good the guy using the mic is.
is a 60 vee or 90 vee the best choice ??
+johnptc Watch next weeks video. We did a test.
how would you inspect the id. part for roundness
+Chris Anderson Check the OD in the V, if OD is round, then use the same indicator placed inside for the ID in the same setup.... Hope I made it clear... ;)
Also with a 60 deg, V block is better...
+pierre beaudry Watch next weeks video. We did a test.
I'll be there...!
pierre beaudry OK but what if the od is off