Waaaaay back in engineering school, I had a professor who started a discussion of geometric dimensioning and tolerances with the story of Plato's Machine shop, where all parts are built exactly to nominal dimension. The only problem is that everything they produce is infinately costly and takes an infinite amount of time to deliver. I also had a riflery instructor make the point that until you can out shoot your gun, that is reliably make groups at or below the rated capability of your equipment, spending a lot of money on fancier gear isn't going to improve your scores.
When I first started my apprenticeship with Rolls-Royce I was using a tolerance and fit chart to ISO 286 to help me design a simple location fixture. When one of the ‘older’ engineers saw me doing this he promptly directed me to the official RR chart that indicated tolerances and fits as used by Royce’s As I was only seventeen at the time and young and innocent I asked him why I could not use the ISO data. His answer still resonates with me today, he replied to me in a very strong tone of voice “because we are Rolls-Royce son, that’s why”. Even after all these years I still keep a copy of this data in my desk drawer!
I saw a sign one time that said Rolls Royce's Never Break Down , they only fail to precede forward . ( the photo of the sign was in a dealership in India ) Pete M
Quinn did a great job connecting the practical implications of each type of fit to the specification. This is super important if you are designing your own parts and mechanisms. Bravo!
Quinn, as a woodworker, hobby steam engine enthusiast, wannabe machinist, I find your videos just my cup of tea. And you sense on humour is the icing on the cake. Your delivery is spot on. I find that I watch through your vids even though they are longer than my typical TLDR. Thank you for your effort. I have subscribed and will follow you with interest
@@Blondihacks Quinn, you have earned the kind words. You make a great teacher and I do love that you are willing to show the reasons behind the Imperial and Metric crying spells. Still love that one shot of you just planting your head on the face of the mill spindle after a boo boo.
Odd duck, you are the best teacher for anything machinist related! you cover all the problems and have a very respectable way of making mistakes and how to correct them, thanks i follow your channel, looking for the new videos all the time, thank you L.C.
These are things as a designer I have to keep in mind when designing things. Some of the work I do is incredibly expensive and requires an incredible amount of precision. Unlike my last employer which was in the ag equipment industry, to where the guys would tell me "If I can't read it on a tape measure I don't want to see your print." So designing mildy complex machines to have plus minus 1/16" fits and still working was quite challenging.
A nice thing about ISO is that you usually design with the same tolerance for all holes (H7 is typical), and varies the shaft tolerance according to your needs. This allows you to use only one set of tools for drilling. For most of general purpose machining, you only need a set of H7 drills and reamers. (And the undersized drills for use before reamers). Since shafts are usually machined in a lathe or purchased as needed, you don't need any special tools to hit whatever tolerance.
@@GeoffTV2 Even so, as a machinist (on my late father’s Imperial lathe) I can work to a thou., as a hand tool woodworker I am happy if my dovetails are within a quarter of a mm.
@@Canalcoholic That's nice going on the woodwork. I can't imagine what my dovetails would be like. I think an eighth of an inch would be optimistic :-)
This is a superb video, clear, concise although complete, i never watched anything clearer about tolerances ! Wonderful 18 minute learning session, thanks a bunch !!
Quinn, Thank you for a very informative video. I am in my 80s and an experienced machinist. Even so I learned alot from this video. You might want to mention in a future video that reaming can be unpredictable. The diameter will vary with rpm, feed rate, coolant (or cutting oi), and probably what you had for breakfast.
As to reaming being unpredictable, don't forget the phase of the moon, which way the wind is blowing, the temperature of the shop and so much more. You even have to factor in the number of ducks flying by at the time.
Quinn nice job covering a complex subject. Another point that enters in is the circular form of the ID to OD components being roundness / runout or degree of eccentricity variation from the center axis. Just clamping a part in a three jaw chuck can produce a trilobed geometry or shape. Drill press can also create this condition . This aspect becomes important as fits get closer and is often controlled by Geometric tolerancing on modern part prints. . .
A fabulous mini-treatise! But I will admit to crying a time or two when machining. Maybe not crying. Feeling glum. But then brightening when the second one turns out so much better than the first. 😁
Queen, your love for sharing your knowledge puts You in a class of your own, thanks for making aspiring machinists understand what lots of master machinist, before You , did not teach to their apprentices, and toke their knowledge in the grave,!!
For me was a superb video and learning. You are far from offensive, in my tolerance scale, and your sense of humor is what makes me like your content more than others. But your level of respect to everyone is exemplary. Wish you well, thank you.
Really captivating stuff ! Before even attempting to climb that 'hill of precision', I might start off by wandering a bit into the Dunning-Kruger 'Valley of Despair' first...
@@oliphab7468 Glad you still like it :) I have been planning to machine/CNC a digital sundial for a while now. But the more I learn about machining, the more I feel I still need to learn some more...
G'day Quinn. Excellent, excellent, excellent! The hill of precision - what an excellent way to explain the topic. By an excellent teacher! You have a special skill so keep it up.
A talented machinist, electrician or other craft person can get away with less-than-great equipment. This is why I need the best tools I can get my hands on!
There is an old story about this young man, the only gun that he could afford to get the game that he needed to feed the kids was a shotgun that would fall apart after every shot. Because the gun was so bad, he learned to be a very good shot. This, as Quinn points out, makes you better quicker.
Found that a very important thing is an adjustable reamer, especially for bushings in shafts, and especially for those times you are replacing a worn bushing, and press the new one in. you need to have the new bush slightly oversize for the hole, to get a good press fit, and I made them 0.5mm undersize on the inside, to allow me to ream them out to fit the existing shaft, as that was slightly worn, but making or getting a new one was not possible time wise. Used some Vesconite hilube, as that would self lubricate, instead of the white nylon bushings the OEM used, and ran dry. Then used a piece of the rod to make an insert tool for them as well, 1mm undersize all dimensions, and gently tapped the bushes into place. Took around an hour to get out to the clearance fit, as you really only get a single chance before it is the same as before. 2 sets of both bushes on that shaft, one still as spare, and they will outlast the original. Could do little about shaft wear or it being slightly bent, other than to lightly polish it and apply a little correction to the bend, plus remove the original burrs on all edges. It is true about Chinese equipment, it arrives as an assembled kit, so you do not need a manual to assemble it, but you still need to take it apart and fix it up to workable from new. However, as the price was one hundredth the cost of the equivalent US manufacturers ( which is ironically made all in India and Vietnam now, nothing aside from the price of parts is US any more) machine, and it is a faster and more capable machine as well, it was well worth the hassle. Most spares are also easy to fabricate, or at least use the part to make oversize fit parts for, and only a few specialised parts you need from the OEM. Just only gripe is why the hell did they have to use 36VAC for all internal electrical works, because that is a pain to get spares, so I was doing a lot of repairs and converting parts to 24VAC, which was thankfully also available inside, but unused.
Great video I might of randomly stubble myself to your channel but I'm very glad I did. As a wood working I've been really getting interested in more metal work as of lately. Thanks for the information!
When making aircraft parts, sometimes it was necessary to put the part under your arm when taking it to the checking desk, it made the part swell up by two tenths, you also had to watch it if the machine shop was a bit on the cold side. Hopefully, that part ended up in someplace hot.
That track day analogy reminded me of when Top Gear got that one lady to drive a van around the Nurburgring and she smoked any time the hosts put up in fancy cars. I miss classic Top Gear...
You know, I'm a seasoned machinist and I enjoy your channel. I hope you continue the honest path you're on. You're more like TubalCain, with an honest intent to help others! WTG and KUDOS. DELETED uncalled for remarks. My apologies.
Great practical explanation of fits. In engineering school they teach us how to calculate for stresses and stains but never a practical breakdown like this.
can i get a amen, the explanation about skill and machines is amazing and so true. nobody thinks its amazing to hold .0001 or even .00005 on an brand new million dollar machine, it cost a million bucks it better hold that tolerance. holding .0001 on a 800 dollar hobby lathe or mill is impressive. its more about the person behind the machine then the machine is the truth.
True, when you breath on a block of metal it deforms slightly. Turn it sideways and the effects of gravity changes it's shape. I myself think of everything, but metal in especial, as rubber.
Odd ducks are always fun but you've never seen a Tassie turbo chook, running in 3 directions at the same time is a rare skill. A very deep rabbit hole...🤔 duck nest?... indeed but I really want to know if the train makes it thru the tunnel
A loose fit is a touch and go, a running fit is a decent landing and an interfence fit is one of those landings after which you need full power to taxi. 😂
Wow--what a great video! I know you've said a rotary table video (or videos) will be coming sometime, and I look forward to something as clear and comprehensive as this--something which--as far as I can tell--doesn't exist anywhere on RU-vid. Joe Pi's are great but they're way over my head, and no one gets at the basics of tools and procedures as clearly as you do.
Nice treatise on the subject. I use the iso system but now I'm curious about the ansi system! Oh, by the way, nothing wrong about liking trains. I love em too.😎
A friend of mine got his first job (fresh from school) in a machine shop milling out pieces from a drawing, they had to produce a certain ammount in a certain time. (this was before the CNC era) As all newbies, he got to work on the oldest, most worn out machine, but still did pretty good, he had to work some overtime if he fell behind but it was pretty ok. Until one day (about halfway trough the run) when he had some time to go see where the pieces were to be mounted, it turned out that none of the measurements needed a tolerance of less then 0.5mm, but the drawing was with tolerance of 0.01mm. The enginerd who made the drawing had that as a default in the program, and only changed it if he needed tighter tolerances. He went out and got all the parts he scrapped and said, I'm done with this run, I'll be home the rest of the week... That was a wakeup call both for him and for the enginerd, that they had costed the company enormous money in wasting time (and material) to make pieces better then needed. "Good enough" is a thing....
hello Quinn, I am an amateur old fart machinist, but at one point in my life I had the great opportunity to work with a man who became a very estimated colleague! He said that I invented the "sloppy fit", that I worked with "rock and roll tolerances" (I am also a musician) and tat everybody should wear safety goggles and helmets whenever I came in the shop to tinker on my projects! Malcolm was a cool Guy! :)
ok going to have to watch that one again as I never knew this stuff before so its really good too get my head around it.. Thanks Quinn for another really good video :)
The way standard dowel pins are made make transitional fits easy. For example a 1/4" dowel pin is made +0.0002" . So if you use your nominal sized 1/4" reamer, 0.2500" to finish the hole the dowel pin will be a nice drive fit that is easy to remove if needed. If you are new to making mechanisms or an assembly of some sort do not locate the parts with screws or bolts. It is invariably a headache to put it together or re-assemble. Follow the tables for hole clearances and clearances for interchangeable parts using threaded fasteners. Here is where dowel pins or spring roll pins are your friend. On your drawings or sketches make a note that the holes for the locating pins are to be match drilled after assembly, i.e. get the parts located and clamped in the correct position then go to the drill press and drill and ream for the pins and you are done. If blind locating holes for the pins are unavoidable there are dowel pins made with a threaded hole in the end so you can extract them with a slide hammer. If the parts have to come apart regularly, consider tapered dowel pins as you only have to tap them to loosen an remove avoiding having to drive them all the way out. Alternatively after match drilling and reaming the parts take a +0.0005" reamer and run it through one of the parts. Drive the pin into one part and then slide the parts together. Cheers from NC/USA
Like she says, always design things with the greatest possible tolerances that will still allow the mechanism to perform satisfactory. One has to consider though, that play in consecutively dependant parts add up. So I began designwork for a coin mechanism in a vending machine, with the aim that it would actually work with 0,5mm play for moving parts. Despite struggling to design for this very sloppy play, the tolerance quickly decended to 0,35-0,25mm, but had to end up at 0,2mm. (The amount of gunk and grit that ends up in a vending machine is unbelievable. So making the mechanical workings tolerate a lot is a good idea.) Then post the prototype series, stainless blind / poprivets were delivered, instead of aluminium rivets. That meant that one of the moving clearances of 0,2mm disappeared when the (also stainless steel) parts were riveted together with the much greater force that results from stainless pop rivets, so the parts could not move at all! Also by mistake, instead of pins that were supposed to be a press fit into somewhat undersize laser cut holes into thick stainless steel sheet, that I intended to drill and / or ream to a press fit for the pins. I fiddled around with several sizes of reamers and drills, but the pins would not stick in the hole. I got quite frustrated, but in the end concluded the diameter of the pins had to be the cause. It turned out the supplier had just assumed (for whatever reason), I wanted a sliding/rotating fit H8. They then exchanged for a press fit M7 pins, and it worked like a charm! It turned out the laser cut holes were actually so exact and consistent in size, no drilling or reaming was necessary for a perfect press fit! That saved me drilling and reaming some 1'600 holes!😅 (Still had to tap another 600 holes M3 in the 300 mechanisms i built, but with a tapping machine and a good cutting lubricant, that was manageable. No broken taps, in spite of it beeing stainless steel! Though I always also make tapping holes somewhat over size. You loose very little mechanical strength by doing so.😉) P.S. In the end the 300 machines were operated +600'000 times over a period of 10 years. Three machines had screws loosen because I applied the loctite wrong, and another four machines had pins bent. Probably due to attempted theft. Only one machine got stuck. I believe someone had poured a sugary soft drink down the mechanism, so a rinse cleared that. I consider the design to have been a success, much thanks to designing for maximum play and tolerances.
Very nice wrap up and overview of small shop tolerance. Well done. Actually probably 95% of all tolerance needs. ( I am leaving out those 25,000 rpm jet engine seals you might want for your next turbo steam project !)
I forget where I read it, but remember an article where a father and his son needed crazy high precision in measuring the error in the axel alignment. So what they ended up doing was using a laser to bound off the axel and show it on a wall. By moving the axel closer or farther away from the wall they could make the error show more or less distance on the wall. I've never had to use such a technique, but I have always thought it cool that you could measure something so with such accuracy if needed.