I swear Stefan, you say in every video that you're too lazy to do things, yet you're probably the most thorough machinist that I've seen hahaha. Love the video awesome as always
The best engineers and machinists (and other craftspeople) are lazy. If you do it right the first time, there's no need to spend effort doing the same job over. Engineering as a profession is the embodiment of lazy efficiency: make machines & structures to do work instead of having people do it!
@@PeregrineBF In software we say "laziness is a virtue". Some argue that it is the _greatest_ virtue of a software engineer. A lazy software engineer doesn't want to come back and fix things. They want to write code once and never think about it again. They automate repetitive tasks so they never have to do them again. They write good documentation so nobody asks them any questions.
Stefan: I bought a Multifix because don't have time to make toolholders anymore. Stefan two minutes later: Wouldn't it be great if I had a rotary broach holder that went in the new Multifix. It's an addiction. We all share it.
I think the idea with the multifix index pins in the bottom is that you use just one into your fixed post. With the pin in the ‘zero’ position you have 40 positions (9 degrees apart). If you take off the holder and use the second hole in the tool post it is 165 degrees from the other. So you have another 40 positions, 6 or 3 degrees offset from the original 40 positions. I am currently making a fixed post to attach my Multifix E.
We really appreciate you taking the time to create your content for us. You have a wealth of knowledge that would be a shame not to share. Like the great minds in history, their job was to share and enlighten us so we may all go to higher levels. Cheers
tip from a toolmaker whos had to grind some fairly long and small diameter core pins with shut offs on the ends on compound angles: choke up on the pin as much as possible. if spinning in a fixture, support under the pin with a spring loaded copper pin in some kind of mild steel holder. a wire spring is almost strong enough to make everything run true again. Also, all my vee block screws have copper tips on them so parts dont get dented.
Hi Stefan. You are a treat to watch and listen to. So well organised and so very good at explaining things. Not just a practitioner or expert but a guru of the mechanical arts. And very self-effacing with it. Fascinating and uplifting as always. Thank you. BobUK.
I don't really have a comment other than thanks for bringing us along with you for a trip through your shop and workings. Hope you are doing well in this day and enjoy your summer. Thank you for the videos.
Excellent from start to finish, thank you. What I learned...... So basically, EVERYTHING is made of rubber. Snap gauges are like playing with hand grenades. Multi-fix tool posts deserve to be fondled.
What a great way to start my day; with a Shop Talk from Stefan Gotteswinter! I learn a lot with every video you make. Having your data for the telescoping gages helps with my own measurements. I was expecting way too much out of them and this was instructive. Thank you so much. Please do a video on the last operation that you were doing! That looked so cool.
I used to teach the use of 'T' gauges, your oversize measurements were due to plunger creep (you mentioned it at beginning ) Allowing plungers to 'snap' out without being in a hole to contain them will eventually result in a broken tool ( even happens with Mitutoyo) Far east mports last between 'first time out of the packet' to 10 days in 'trade school classroom' environment) The radiused ends of T plungers should find correct centre but averaging is a good idea for critical sizes One of the 'Final Test' tasks was to measure various motorcycle components to see if they were still within manufacturers spec. Easiest way to take measurement is put working end in hole at angle, pinch up lock, 'swipe' handle across hole and remove gauge (at the opposite angle) If your right handed, only eve have T gauge in left hand and micrometer in right hand, it's far quicker to develop a consistent feel (and obviously, if your left handed do the opposite) You can get a very accurate reading when you develop a feel for 'stick' It feels a bit like a cheap telescopic magnet grabbing something 'too heavy' and letting go again (there isn't an easy way to describe feel)
Very interesting update Stefan. I liked the V-block deflection measurement - had never considered that a V block could cause round part deflection.. Also the broad angle/form tool at the end was very impressive. That's a lot of engagement!
In lockdown, time flies when Stefan post a long video. I'm waiting for him to give up the day job so he spend more time in the bunker creating more exotic tooling.
I just made up a 5C stop with an interchangeable end. Good tip on facing in place. It never occurred to me how annoying the fine thread in the rod is until now.
Hi Stefan, really interesting gauge R & R on the telescoping gauge, and an insight into statistical analysis and differentiating common cause variation from special cause variation. For anyone who doesn't deal with this stuff in their day job, you broke it down really well and demonstrated it flawlessly. Up to your usual excellent standard, well done and thanks. Jon
Stefan..I made a live tooling holder that directly mounts to a BXA tool post and doubles as a rotary broach. I just rotate the tool post the 1* and indicate the angle in (would be problematic with your fixed setup). I used tapered roller bearings vs a thrust bearing and balls. This allows me to dial out any spindle end play and set the preload with the bearing nut. I have a slinger built into the spindle shaft to protect the front oil seal and the bearings runs in an oil bath...simple setup and it will take all the thrust you can throw at it. Only thing I would change is to make the spindle compatible with ER collets.
Thank you, always enjoy your thorough explanation of what ever you are doing; which ensures that even a novice such as myself can follow and understand. I was surprised, but pleased, to see you move to the multi-fix as I bought one last year to use on my little Myford 7 - I now know how to set it up properly. Must add that I very much enjoy listening to you on the Machinists Podcast too. Stay well, blessings from Lancashire, England.
Stefan, thank you for the multi project video. You are my favorite u tube video producer You are an amazing machinist and a very talented designer. You always come up with the perfect solution to every part or tool that you need. You also make them to perfection and explain why you chose a certain set of steps to achieve the best end product. Stay safe my friend!!!
It’s worth looking for Create Tools - a Chinese company that makes a very high quality version. I bought an E set from them and they are super quality, not to be confused with cheap Chinese toolholders. I am pretty sure they supply PeWe. After buying a set I went back and bought more toolholders - pretty quick delivery, good prices and they do their best to minimise import duty.
I always learn so many things from your videos! I don't have any of these things yet (except a set of 5 telescoping gauges), but will try and remember them for later. lol
Hi Stefan , Firstly thank you for a very interesting and stimulating video ! I was taught to use a telescoping or snap gauge and an inside mic in the vertical position so gravity doesn't pull the gauge down as you wobble it - never really tested whether it makes a difference but if some one could it would be you ! With telescoping or snap gauges i feel getting the torque on the locking screw as consistent as possible can be difficult especially for the newcomer - too little and it can move - too much and you risk twisting the gauge as you tighten it . The next problem is transferring that measurement to a micrometer and the "bump "in the bore has to be the same "bump"when you measure it with a micrometer . One tip is if you haven't used a snap gauge , micrometer ( inside or outside ) is to get your hands on a decent quality roller bearing and use this to practice your measuring skills as a good quality bearing will be very close to the stated specifications and fairly cheap to purchase compared to something like a setting ring or standard . I wonder if some one one day will come up with a ratchet system for telescoping or snap gauges ? I have a set of Moore and Wright small bore gauges that use a taper pin to drive out two hardened balls. You use it a bit like an inside mic. It doesn't tell you any direct reading as you need to measure it with a micrometer and you can sort of gauge if the bore is parallel or out of round with them .
If anyone is curious, Robin talks about V block deflection (briefly, as Stefan mentioned), in his recent video ruclips.net/video/OjQ5pOkCeW0/видео.html about 11m in... (I had to go look, because I rewatched it recently). "Making Grooving Boring Bars"
The first time I saw your Trippan QCT I was so impressed that I made myself one. It was then I realized you can have only the ID kind of tool holders if you slot the short face as well as the long one. You have to orient the body of the QCT perpendicular to the lathe's axis and the tool holder goes upside down for one position and right side up the other. The slots are 90 deg to each others.
Thanks for the bore measuring stuff. I try to measure woodwind instruments and often get frustrated about my lack of repeatability. Now I have some pointers to how I can improve.
Great tips, especially since they're coming from the cloak and dagger side of machining. Stefan could, much more clearly, reveal many more secrets but then, sadly, he'd have to kill us :) Spies, intrigue, and machining. Who knew you could mix those things together? Gun? What gun? I have calipers and gauge blocks. I don't need guns! Torture? I don't believe in torture but I do have an overhead grinder that can persuade answers to within a half a micron or better. Generally when the grinder gets switched on, they start talking right away...... So, anyway, some machinists do live a dangerous and exciting life. Remember Clint Eastwood in the Eiger Sanction? Mild mannered college professor by day and deadly spy by night....
You are right when you said around @23:30 that the mean should be off, the reason is that you have some outliers, like the 26.673mm, but if you use the geometric mean (which reduces the effect of the outliers) you get 26.5300mm. The variance of this set of measurements is around 0.00361mm^2, maybe a good result or maybe not given your obsession with microns. No accuracy test is accurate if you don't report both the central location of the measurements and the variance/standard deviation (0.0601mm in this case). The SD is telling you that, based on this set of measurements, 68.2% of the time you will get a measurement in the interval (26.470, 26.590). More measurements can mitigate the effect of outliers even more, but they are measuring errors you cannot discard. Moreover, much more important is to find the confidence level of your measurements. That is to say what is the probability that your measurements do not fall within the tolerance. It is not hard these days to make such analysis, a spreadsheet software is capable of setting a hypothesis testing setup. I decided to add this part: A nice trick to get better results is to take an odd set of measurements, order them, and consider the central one as the best measurement (you use the median instead of the mean). You can also discard those that are outliers, but in that case you must discard the same number of samples from both ends. You can also do this for even sample size, but then the median would be the center point between the to central measurements. In this case the bore should be 26.501mm, much more closer to the mean you computed after dropping the outliers.
Very cool analysis. I definitely need to learn more about stats. Taking the median came to mind for me too. One plus point of that, is it's so easy to do on paper, just sort and take the middle entry. (Or if there's an even number of measurements, take the middle two and find the midpoint)
Yeah I'd go for the median too and use the robust version of SD to go with it (Median Absolute Deviation) . I'm always wary of using geometric mean to make a smaller value than arithmetic mean, unless there's reason to suspect a log-normal distribution.
@@2lefThumbsThese measurements will probably follow a log normal distribution. I was thinking of it having a long right tail, which makes sense if we think that the measuring arms can slip out once the gauge is removed from the bore. Measuring with these gauges requires training and adjusting to the set one uses because. There has to be a delicate balance between the amount of drag to prevent slip during gauge removal. However slip can occur because those springs seem to be quite strong. Then looking into the log-normal distribution I noticed a mention of its use in engineering and metrology, so it seems that many instruments will produce readings that follow this distributions. Although in practice I would prefer to take five measurements, order and keep the central value as the best. If all measurements are close to each other it means little probability of error. I wouldn't trust a single measurement unless for roughing (like when we use a caliper to check where we are and take the final measurements with micrometers).
Thanks for making a video just for us, it has very useful and interesting. The work i do is under a tightly controlled contract also. Thanks for sharing.
For the first part with parting off the aluminium bar I can also recommend an end stop in the tail stock. used this myriad times. Clamp e. g. 10mm bar stock (faced) in your drill chuck. Place a 5mm thick piece of rectangular stock between the part and the end stop. Wiggle it out so your bar stock can rotate and won't jam, pick up your zero point and start parting away. But I have to admit your / Robins method is beautiful and I just thought "why haven't I had that idea" 🤦♂️
Stefan, ich bin etwas entsetzt, dass du die Anwendung der Teleskoplehren falsch zeigst. Das Teleskop muss beim Messvorgang vertikal stehen, sonst rutscht es durch sein Eigengewicht unter die Mittellinie.
Stefan, thanks for a great shoptalk again. These may seem like being an number of unrelated random subjects but I love them. Just finished watching and realized I am still holding my cup of coffee, now cold, at the end 😁 Love the information about the Multifix holder. We have a A size Chinese one which I do not like at all It is very hard to adjust and seems to be not as accurate as I wanted. On my smaller lathe I just bought a DIxon style holder (made in China) - cheaper than the Multifix and it seems to be as good as the Multifix on the larger lathe ... Since I will be spending this years holiday in a cheap place (Hintergarten am Rhein) I am now almost tempted to spend my holiday allowance on an Axa Multifix
Clever idea on the depth stop. Given the 20° included angle of the 5C every .001"/.025 mm difference in diameter gives you .003"/.075 difference in length. A far better solution is of course dead length collets. Expensive (it must be the H, Hardinge expensive, Harley Davidson, expensive). An easier solution folks is a stop length collar. The Hardinge catalogue on workholding shows these but they are very easy to make. The collar slips over the work and bears against the shoulder (the draw back is you need a shoulder on the part) on the work as the collet draws in it pulls the workpiece against the collar it places the part in tension. Rigidity is markedly increased. The 5C Collet is not ideal due to a number of factors. But when one takes in the entire 5C Collet System one realizes just how useful the are. An ER collet might have more range but just where does one find an ER internal collet. Or Step Collet. Or Face Plate Collet. I'd post a link to pictures but my Photobucket account is being nasty of late. Need to find a new image service.
I so enjoy your thorough explanations as you fixture-cut and certify dimensions. Size is not of concern, but quality always tells. The only question I have is to wonder what other languages you machine in? Oh yes, I really like the shop and how you use your space.
Even though I don't have a multifix, the info your videos hash out is always something important to getting things right, and getting it right enough for you. THAT pushes everyone to ask themselves the same question before taking the cut... What would Stefan do in this situation? The answer is always, Don't settle on microns of error. If you can measure them, you can cut them proper!!
Let's not be ridiculous, Robin is a nice guy but no one works in sub tenths, not even Robin. Do you know what it takes to work in sub tenths? AA Jansson metrology lab works in sub tenths. Their sub tenths lab is in a small temperature humidity controlled room, they never shut off the light because it affects tolerances. The guy's making flux capacitors, they're imaginary!
I'm happy to watch your videos. I suggest you evaluate lamping collars. I propose ot splitting them as they are off the shelf but with a slit tangent to the bore. Them when tightening the clampingscrew tends to grip by wrapping not just squeezing. I've been designing parts this way with more reliability
Obviously too late to say that but my lathe had a 4 tool toolpost with a 8 position ratcheting mechanism that consisted of a toothed plunger on the compound slide and 8 serrations on the bottom of the toolpost. I made an adapter plate with 60 serrations on the bottom (for 6 degree indexing). The newer Tripan toolposts have a locating pin to lock them in position so after locking it at a full index position (as all positions in between are possible with the one way ratchet, and a circlip lifts it off the ratchet when the bolt is further unscrewed so you don't have to do a 360 in case you accidentally go to the next tooth) you can just indicated the toolpost directly and drill the hole but since mine had a parallel key going through the center of the locating bore I put a boring tool on the lathe chuck to mill a reference flat vertical to the lathe axis. I then set that parallel to mills Y, found the center of the bore and mill the key slots across it without touching the round boss of the adapter. In literally 30 minutes I have an appointment to have some someone blast through a few adapters...I'm actually really proud of that design because it can combine the speed and versatility of the multifix with the easier to clean, easier to make adapters and higher tailstock clearance and visibility of the Tripan.
In statistical studies, particularly for QC and 6 Sigma type studies, we consider quartiles and median figures, rather than mean, to help account for outliers.
I found the last video segment interesting. A tapered cut, looked like low speed and a wide chip. Could you provide details on the type of material being machined, doc, and more on the tool doing the cutting?
Interesting . I am not sure if you have tried the Moore & Wright style fixed end telescopic gauges , i find they are a step above the style that both plungers move when feeling for the size . Cheers .
I had some brand-X gauges - didn't like them, bought some Mitutoyo - not better! I could see lots of burrs under the microscope - no better than the Chinese. I think the problem is some of the players are just putting their names on things. I used some old ones once - nice an smooth - worked fine - but he wouldn't sell them. I took the Mitutoyos apart - smoothed things with some pencil stones - still not right. Not sure you can fix the cheap ones - and the expensive ones may be cheaply made today. I think the tolerance of the pistons vs tube is sloppy - not something fixable. TOT didn't seem to fix his either.. In production shops they just don't use these anymore, so it could be no one cares.
Thanks Stefan, you've given me some ideas! I wanted one of those Multifix toolposts for my lathe but in the end they were just too expensive for me. Maybe when I get a larger lathe...
Most older german lathes come equipped with the Multifix holder. My Weiler, for exampåle, had the Hahn&Kolb version of the post. Now, some far east toll holders are NOT compatible with the "original" Multifix tool post. So, do not assume that the tool holder is compatible if the size designation (A, Aa, B, ...) is the same. The two brands Stefan mentioned are OK. The AXA is made in Germany, The PwTools is made in far east under very rigid german control. There is a german phrase; "Trust is good, control is better". I bought my holders from PwTools as I was actually able to call the owner and ask whether the holders are compatible.
My last job we had one lathe set-up with a Multi-Flex. The only thing I did not like about them was the screws in the holders are very hard to find in the US. Not because they are Metric (or ISO if one prefers) but because they were 7mm or 9mm screws. I know they weren't 8mm. Of course the company was expanding their machining department and the new boss decided to restrict access to people he hired. I guess 45 years wasn't enough experience 🙄. But then the company started doing inhouse production work and the machining I did there was related to repair work
I quite like the British Dickson holders, but also Aloris because of its ability to do super quick tool changes. This is something cnc lathe manufacturers need to work on.
Hallo Stefan Zwei Ergänzungen könnte man noch machen: Es gibt noch einen kleineren Multifix als den Aa, nämlich den Aaa. Und man hätte das mit den zwei Stiften vielleicht noch erklären sollen. Dass der 2. Stift um 4,5 Grad versetzt ist, was einem zu den 40 möglichen Positionen des Multifix noch weitere 40 bringt.
Ja, bei den regulären Mfix gibts die nicht, deswegen hab ich die ausgelassen . Gibts nicht sogar mitlerweile einen noch kleineren als AAA? Das mit dem zweiten Stift war mir nicht klar, man lernt nicht aus :)
Why do telescoping gauges have flat faces vs pointed, or even rounded to the radius of the minimum dimeter they are capable of measuring? It seems you have to wiggle them around to feel when they catch to take a measurement.
Gday Stefan, thank you the tips on telescoping gauges, I always enjoy you videos, I’m always learning, I think it was on Instagram you mentioned about the ikea lamps you use, mine arrived in the post today, I’m looking forward to seeing a bout better on the lathe in the future, thank you, take care, Matty
I guess that the telescopic gauges , where off at the end , creeping under 25.500 mm . Due to the heating up of the telescopic gauge .. Indeed the mitutoyo micrometer is well insulated , but the telescopic gauge wasn't at all. reason its getting smaller (counter intuitif) , was because you measured a ID so when tips get hotter they touch earlier . Is my throw at trhe repebilaty numbers . I like such testing alot , weel done Stefan grtzz from the Netherlands Johny Geerts
With 9deg incremental rotation, assuming you set it up like Stephan did adjusting the boring bar to be aligned with the ways as a test, then you Always have 90 and 0/180 deg options for whatever op including external and internal threading. What Stephan has done in line with his rigid, no compound slide technique is to carefully set that 0/90 orientation and dowel pin the 40 pos core block. Even going as far as showing his first failed attempt.. I have worked 40 pos equiped lathes, with compounds, none of which have been dowel fixed, and you do find yourself having to adjust the position of the core block tooo often.
These are nice. I like my Dorian though. Very solid. Very accurate. Far easier to mount. Lots of block styles. Easier to keep dirt out of the tool and blocks.
I watched a video about the mutifix toolpost that showed it made a difference in the direction you tighten the cam. Just happened to be the opposite way that you were tightening yours. It seems that the inconsistency of the way the cam was aligned. Wish I could give credit where credit is due , but I don’t recall the channel name.
Thank you for the multifix review, I have one in my basket ready to buy but out of interest most of us Myford boys use a Dixon/ Dickson quick change tool post. What would you say is there main disadvantage?
I had a Dickson for my Myfords but I gave it away because it was so bulky, the cutting force is too far overhung for my liking, the toolholders or bits have to be so shallow, and there is so little room to spare as regards clearing the tailstock and the chuck. I would opt for Tripan, ideally, but now I have bigger lathes, I only use the Myfords for instrument work and the like, so I've actually gone back to the Dutch Clog OEM clamp on my Super 7 & ML7 because it is so versatile and compact. With 16mm shank insert tooling the height is bang on, for most HSS I use Ifanger tangential tooling, which don't require packing, and for other tools, I ground a bunch of graduated packers from 3.0 to 4.5 in 0.1mm increments, and also a few thicker ones, and I write the height of the packer required on the side of each tool in indelible pen. I worked out a couple of nifty techniques to quickly get the clamp surface dead level and I threw away the annoying spring. I also modified a deep socket to fit it with a sliding crossbar for the Super 7, and I cut down a cranked ring spanner for the ML7 On my bigger lathes I use Dickson, because that's what they came with, and second hand holders are readily available. I would like either Multifix or the American one with the clamping dovetail, but the Dickson (one genuine, one a good Italian Rapidue clone) are pretty solid and very trouble free for big lathe work, I find. I just bought a T-slot style faceplate for my biggest lathe, and it would make a good stand for one of my Myfords. It weighs more than both of them put together. Come to think of it, it weighs more than I do! Anyone reading this who has just upsized their lathe: make sure you have a long, strong "cheater bar" to do up the toolpost clamp nut. on a 400mm swing lathe it wants to be about 800mm long. On a bigger lathe, it pays to either use a commercial cheater bar, or if you make your own, make it with a heavy wall extension pipe to fit over the end to provide the necessary lever arm when you expect to take big cuts.
Lindsay Drabsch Stefan I have retired from an engineering business after 50 years and this is not the way to use telescopic gauges .I have been using these gauges together with Swiss Tesa micrometers to measure bores .The way I was taught was to place the gauge in the bore at an angle ,lock it and pull it so the handle is parallel with the bore .The gauge will side to the exact size of the bore .The measurement will be repeatable to a micron on the Swiss micrometer provided the cylinder has sufficient wall thickness .This method of using telescopic gauges does wear the the gauges so I do have a number that are worn out .In business where mounting ball races or press fits it is necessary to get precision in the shortest time
You are correct in that my method wears the gauges in fact i have a box of worn out gauges The precision that is obtained in measuring is worth treating the gauges as consumable I use the gauges in all materials both aluminum and brass It is important that the bore to be measured has a high grade finish and the wall is thick enough that it does not distort The method I use is to complete the bore first while there is sufficient material that does not distort
I swear Stefan, you say in every video that you're too lazy to do things, yet you're probably the most thorough machinist that I've seen hahaha. Love the video awesome as always
The best engineers and machinists (and other craftspeople) are lazy. If you do it right the first time, there's no need to spend effort doing the same job over.
Engineering as a profession is the embodiment of lazy efficiency: make machines & structures to do work instead of having people do it!
@@PeregrineBF In software we say "laziness is a virtue". Some argue that it is the _greatest_ virtue of a software engineer.
A lazy software engineer doesn't want to come back and fix things. They want to write code once and never think about it again. They automate repetitive tasks so they never have to do them again. They write good documentation so nobody asks them any questions.
Yeah I wish i was that lazy.
@@outputcoupler7819 Spending 20h to automate a 5 min task so you don't have to do it ever again. Totally normal software development.
Stefan: I bought a Multifix because don't have time to make toolholders anymore.
Stefan two minutes later: Wouldn't it be great if I had a rotary broach holder that went in the new Multifix.
It's an addiction. We all share it.
It could be his sense of humor, you never know😂
So impressed with your calm, organised and precise work. It makes me want to clean my workshop.😀
It is so good to listen to a person who really understands the basics, works extremely carefully and is a genius!
I think the idea with the multifix index pins in the bottom is that you use just one into your fixed post. With the pin in the ‘zero’ position you have 40 positions (9 degrees apart). If you take off the holder and use the second hole in the tool post it is 165 degrees from the other. So you have another 40 positions, 6 or 3 degrees offset from the original 40 positions. I am currently making a fixed post to attach my Multifix E.
I also remember seeing this somewhere. Probably it was a PeWeTools video. It enables to reach angles like 15°, 30°, 60°, 75°.
'Outlier' is the word you were looking for rather than offshot. Another great video Stefan.
You and robin has started a solid post/compound delete movement all over the internet. Love it👍👌🇦🇺
We really appreciate you taking the time to create your content for us. You have a wealth of knowledge that would be a shame not to share. Like the great minds in history, their job was to share and enlighten us so we may all go to higher levels.
Cheers
tip from a toolmaker whos had to grind some fairly long and small diameter core pins with shut offs on the ends on compound angles: choke up on the pin as much as possible. if spinning in a fixture, support under the pin with a spring loaded copper pin in some kind of mild steel holder. a wire spring is almost strong enough to make everything run true again. Also, all my vee block screws have copper tips on them so parts dont get dented.
Your experiments are the reason I enjoy your videos so much.
Hi Stefan. You are a treat to watch and listen to. So well organised and so very good at explaining things. Not just a practitioner or expert but a guru of the mechanical arts. And very self-effacing with it. Fascinating and uplifting as always. Thank you. BobUK.
Thanks Stefan. Your telescope gauge experiment was very instructive.
Working in .001s use a telescoping gage, working in tenths a dail bore gage.
I don't really have a comment other than thanks for bringing us along with you for a trip through your shop and workings. Hope you are doing well in this day and enjoy your summer.
Thank you for the videos.
Well presented and informative video Stefan (as usual)
I like the spreader idea in the v block. It also stops the screw from marking softer materials
Excellent from start to finish, thank you. What I learned...... So basically, EVERYTHING is made of rubber. Snap gauges are like playing with hand grenades. Multi-fix tool posts deserve to be fondled.
These are Telescoping gauges, subtle differences from 'Snap' gauges which are generally an OD, Qa micrometer like gauge.
What a great way to start my day; with a Shop Talk from Stefan Gotteswinter! I learn a lot with every video you make. Having your data for the telescoping gages helps with my own measurements. I was expecting way too much out of them and this was instructive. Thank you so much. Please do a video on the last operation that you were doing! That looked so cool.
I used to teach the use of 'T' gauges, your oversize measurements were due to plunger creep (you mentioned it at beginning )
Allowing plungers to 'snap' out without being in a hole to contain them will eventually result in a broken tool ( even happens with Mitutoyo)
Far east mports last between 'first time out of the packet' to 10 days in 'trade school classroom' environment)
The radiused ends of T plungers should find correct centre but averaging is a good idea for critical sizes
One of the 'Final Test' tasks was to measure various motorcycle components to see if they were still within manufacturers spec.
Easiest way to take measurement is put working end in hole at angle, pinch up lock, 'swipe' handle across hole and remove gauge (at the opposite angle)
If your right handed, only eve have T gauge in left hand and micrometer in right hand, it's far quicker to develop a consistent feel (and obviously, if your left handed do the opposite) You can get a very accurate reading when you develop a feel for 'stick'
It feels a bit like a cheap telescopic magnet grabbing something 'too heavy' and letting go again (there isn't an easy way to describe feel)
Very interesting update Stefan. I liked the V-block deflection measurement - had never considered that a V block could cause round part deflection.. Also the broad angle/form tool at the end was very impressive. That's a lot of engagement!
Thanks for another good shoptalk Stefan. Stay safe stay well. regards from the UK
In lockdown, time flies when Stefan post a long video. I'm waiting for him to give up the day job so he spend more time in the bunker creating more exotic tooling.
I just made up a 5C stop with an interchangeable end. Good tip on facing in place. It never occurred to me how annoying the fine thread in the rod is until now.
Hi Stefan, really interesting gauge R & R on the telescoping gauge, and an insight into statistical analysis and differentiating common cause variation from special cause variation. For anyone who doesn't deal with this stuff in their day job, you broke it down really well and demonstrated it flawlessly. Up to your usual excellent standard, well done and thanks. Jon
Stefan..I made a live tooling holder that directly mounts to a BXA tool post and doubles as a rotary broach. I just rotate the tool post the 1* and indicate the angle in (would be problematic with your fixed setup). I used tapered roller bearings vs a thrust bearing and balls. This allows me to dial out any spindle end play and set the preload with the bearing nut. I have a slinger built into the spindle shaft to protect the front oil seal and the bearings runs in an oil bath...simple setup and it will take all the thrust you can throw at it. Only thing I would change is to make the spindle compatible with ER collets.
Thank you, always enjoy your thorough explanation of what ever you are doing; which ensures that even a novice such as myself can follow and understand. I was surprised, but pleased, to see you move to the multi-fix as I bought one last year to use on my little Myford 7 - I now know how to set it up properly. Must add that I very much enjoy listening to you on the Machinists Podcast too. Stay well, blessings from Lancashire, England.
Stefan, thank you for the multi project
video. You are my favorite u tube video producer
You are an amazing machinist and a
very talented designer. You always
come up with the perfect solution to
every part or tool that you need.
You also make them to perfection
and explain why you chose a certain
set of steps to achieve the best end
product.
Stay safe my friend!!!
That solid base on the tool holder makes tremendous difference in overall ridgidity. What I like about it is parting becomes easier and safer.
I have seen those tool holders... now they have the Stefan stamp of approval I will add one to my shopping list.
It’s worth looking for Create Tools - a Chinese company that makes a very high quality version. I bought an E set from them and they are super quality, not to be confused with cheap Chinese toolholders. I am pretty sure they supply PeWe. After buying a set I went back and bought more toolholders - pretty quick delivery, good prices and they do their best to minimise import duty.
I always learn so many things from your videos! I don't have any of these things yet (except a set of 5 telescoping gauges), but will try and remember them for later. lol
Hi Stefan ,
Firstly thank you for a very interesting and stimulating video !
I was taught to use a telescoping or snap gauge and an inside mic in the vertical position so gravity doesn't pull the gauge down as you wobble it - never really tested whether it makes a difference but if some one could it would be you !
With telescoping or snap gauges i feel getting the torque on the locking screw as consistent as possible can be difficult especially for the newcomer - too little and it can move - too much and you risk twisting the gauge as you tighten it . The next problem is transferring that measurement to a micrometer and the "bump "in the bore has to be the same "bump"when you measure it with a micrometer .
One tip is if you haven't used a snap gauge , micrometer ( inside or outside ) is to get your hands on a decent quality roller bearing and use this to practice your measuring skills as a good quality bearing will be very close to the stated specifications and fairly cheap to purchase compared to something like a setting ring or standard .
I wonder if some one one day will come up with a ratchet system for telescoping or snap gauges ?
I have a set of Moore and Wright small bore gauges that use a taper pin to drive out two hardened balls.
You use it a bit like an inside mic. It doesn't tell you any direct reading as you need to measure it with a micrometer and you can sort of gauge if the bore is parallel or out of round with them .
If anyone is curious, Robin talks about V block deflection (briefly, as Stefan mentioned), in his recent video ruclips.net/video/OjQ5pOkCeW0/видео.html about 11m in... (I had to go look, because I rewatched it recently).
"Making Grooving Boring Bars"
The first time I saw your Trippan QCT I was so impressed that I made myself one. It was then I realized you can have only the ID kind of tool holders if you slot the short face as well as the long one. You have to orient the body of the QCT perpendicular to the lathe's axis and the tool holder goes upside down for one position and right side up the other. The slots are 90 deg to each others.
Thanks for the bore measuring stuff. I try to measure woodwind instruments and often get frustrated about my lack of repeatability. Now I have some pointers to how I can improve.
I am in shock with the fact that you actually own an eraser AND you needed it. 😳😁
In fact he never made a mistake like this and just uses the eraser to appeal to regular people like us.
Great tips, especially since they're coming from the cloak and dagger side of machining. Stefan could, much more clearly, reveal many more secrets but then, sadly, he'd have to kill us :)
Spies, intrigue, and machining. Who knew you could mix those things together?
Gun? What gun? I have calipers and gauge blocks. I don't need guns! Torture? I don't believe in torture but I do have an overhead grinder that can persuade answers to within a half a micron or better. Generally when the grinder gets switched on, they start talking right away......
So, anyway, some machinists do live a dangerous and exciting life. Remember Clint Eastwood in the Eiger Sanction? Mild mannered college professor by day and deadly spy by night....
He faked it so that we don't feel so bad about ourselves
Good stuff Stefan, enjoyed!
ATB, Robin
Thanks Robin!
You are right when you said around @23:30 that the mean should be off, the reason is that you have some outliers, like the 26.673mm, but if you use the geometric mean (which reduces the effect of the outliers) you get 26.5300mm. The variance of this set of measurements is around 0.00361mm^2, maybe a good result or maybe not given your obsession with microns.
No accuracy test is accurate if you don't report both the central location of the measurements and the variance/standard deviation (0.0601mm in this case). The SD is telling you that, based on this set of measurements, 68.2% of the time you will get a measurement in the interval (26.470, 26.590). More measurements can mitigate the effect of outliers even more, but they are measuring errors you cannot discard.
Moreover, much more important is to find the confidence level of your measurements. That is to say what is the probability that your measurements do not fall within the tolerance. It is not hard these days to make such analysis, a spreadsheet software is capable of setting a hypothesis testing setup.
I decided to add this part: A nice trick to get better results is to take an odd set of measurements, order them, and consider the central one as the best measurement (you use the median instead of the mean). You can also discard those that are outliers, but in that case you must discard the same number of samples from both ends. You can also do this for even sample size, but then the median would be the center point between the to central measurements. In this case the bore should be 26.501mm, much more closer to the mean you computed after dropping the outliers.
Very cool analysis. I definitely need to learn more about stats. Taking the median came to mind for me too. One plus point of that, is it's so easy to do on paper, just sort and take the middle entry. (Or if there's an even number of measurements, take the middle two and find the midpoint)
Yeah I'd go for the median too and use the robust version of SD to go with it (Median Absolute Deviation) . I'm always wary of using geometric mean to make a smaller value than arithmetic mean, unless there's reason to suspect a log-normal distribution.
@@2lefThumbsThese measurements will probably follow a log normal distribution. I was thinking of it having a long right tail, which makes sense if we think that the measuring arms can slip out once the gauge is removed from the bore. Measuring with these gauges requires training and adjusting to the set one uses because. There has to be a delicate balance between the amount of drag to prevent slip during gauge removal. However slip can occur because those springs seem to be quite strong.
Then looking into the log-normal distribution I noticed a mention of its use in engineering and metrology, so it seems that many instruments will produce readings that follow this distributions. Although in practice I would prefer to take five measurements, order and keep the central value as the best. If all measurements are close to each other it means little probability of error. I wouldn't trust a single measurement unless for roughing (like when we use a caliper to check where we are and take the final measurements with micrometers).
Thanks for making a video just for us, it has very useful and interesting. The work i do is under a tightly controlled contract also. Thanks for sharing.
For the first part with parting off the aluminium bar I can also recommend an end stop in the tail stock. used this myriad times.
Clamp e. g. 10mm bar stock (faced) in your drill chuck. Place a 5mm thick piece of rectangular stock between the part and the end stop. Wiggle it out so your bar stock can rotate and won't jam, pick up your zero point and start parting away.
But I have to admit your / Robins method is beautiful and I just thought "why haven't I had that idea" 🤦♂️
Love those multifix holders. Had the pleasure of modeling an organizer for the larger B size. The finish and fit was superb.
Love the idea of the Hemingway upgrade!
schön, dass du einen neuen Shoptalk gemacht hast. Den Anfang muss ich mir wohl noch in 360p ansehen
Gerne! Sollte eigentlich 1080p HD sein - Kann manchmal etwas dauern bis jeder Nutzer volle Auflösung bekommt :\
Stefan, ich bin etwas entsetzt, dass du die Anwendung der Teleskoplehren falsch zeigst. Das Teleskop muss beim Messvorgang vertikal stehen, sonst rutscht es durch sein Eigengewicht unter die Mittellinie.
Nice to see how you fix your multi fix thanks for the video
Stefan, thanks for a great shoptalk again. These may seem like being an number of unrelated random subjects but I love them.
Just finished watching and realized I am still holding my cup of coffee, now cold, at the end 😁
Love the information about the Multifix holder. We have a A size Chinese one which I do not like at all It is very hard to adjust and seems to be not as accurate as I wanted. On my smaller lathe I just bought a DIxon style holder (made in China) - cheaper than the Multifix and it seems to be as good as the Multifix on the larger lathe ... Since I will be spending this years holiday in a cheap place (Hintergarten am Rhein) I am now almost tempted to spend my holiday allowance on an Axa Multifix
I could listen all day, so calm and collected in your explanations even if you do have a problem pronouncing words with V in them like wise 😆😉
In the German language V is pronounced as a W ...It's not a problem...just a different language, and for sure, his English is better than our German.
Clever idea on the depth stop. Given the 20° included angle of the 5C every .001"/.025 mm difference in diameter gives you .003"/.075 difference in length. A far better solution is of course dead length collets. Expensive (it must be the H, Hardinge expensive, Harley Davidson, expensive). An easier solution folks is a stop length collar. The Hardinge catalogue on workholding shows these but they are very easy to make. The collar slips over the work and bears against the shoulder (the draw back is you need a shoulder on the part) on the work as the collet draws in it pulls the workpiece against the collar it places the part in tension. Rigidity is markedly increased. The 5C Collet is not ideal due to a number of factors. But when one takes in the entire 5C Collet System one realizes just how useful the are. An ER collet might have more range but just where does one find an ER internal collet. Or Step Collet. Or Face Plate Collet. I'd post a link to pictures but my Photobucket account is being nasty of late. Need to find a new image service.
A pleasure to watch as always 👍👍🏴
You always post such interesting content. Thanks for sharing your expertise with us.
I so enjoy your thorough explanations as you fixture-cut and certify dimensions. Size is not of concern, but quality always tells. The only question I have is to wonder what other languages you machine in? Oh yes, I really like the shop and how you use your space.
Another great video. Thanks!
Even though I don't have a multifix, the info your videos hash out is always something important to getting things right, and getting it right enough for you. THAT pushes everyone to ask themselves the same question before taking the cut... What would Stefan do in this situation? The answer is always, Don't settle on microns of error. If you can measure them, you can cut them proper!!
Let's not be ridiculous, Robin is a nice guy but no one works in sub tenths, not even Robin. Do you know what it takes to work in sub tenths? AA Jansson metrology lab works in sub tenths. Their sub tenths lab is in a small temperature humidity controlled room, they never shut off the light because it affects tolerances.
The guy's making flux capacitors, they're imaginary!
Wow excellent work
I'm happy to watch your videos. I suggest you evaluate lamping collars. I propose ot splitting them as they are off the shelf but with a slit tangent to the bore. Them when tightening the clampingscrew tends to grip by wrapping not just squeezing. I've been designing parts this way with more reliability
I have always thought that the nut that holds the toolpost is behind the camera - you live and learn.
Really nice form tool cut at the end. That much length of cut it is really easy to get a lot of chatter
Another shoptalk! These are always a pleasure to watch!
Case hardened shafts are nice for that kind of applications. One source for those are old printers. Especially old IBM line printers
Excellent, detailed content as always. Thank yiu sir.
that last cut was pretty interesting sir.
I was expecting a whole lot of chatter, note the ultra slow speed though, I'm guessing not slow mo.
If my memory serves me correct, he did something similar to demonstrate the rigidity of the setup when he converted this to solid tool post.
That was absolutely beautiful, Stefan.
Obviously too late to say that but my lathe had a 4 tool toolpost with a 8 position ratcheting mechanism that consisted of a toothed plunger on the compound slide and 8 serrations on the bottom of the toolpost. I made an adapter plate with 60 serrations on the bottom (for 6 degree indexing). The newer Tripan toolposts have a locating pin to lock them in position so after locking it at a full index position (as all positions in between are possible with the one way ratchet, and a circlip lifts it off the ratchet when the bolt is further unscrewed so you don't have to do a 360 in case you accidentally go to the next tooth) you can just indicated the toolpost directly and drill the hole but since mine had a parallel key going through the center of the locating bore I put a boring tool on the lathe chuck to mill a reference flat vertical to the lathe axis. I then set that parallel to mills Y, found the center of the bore and mill the key slots across it without touching the round boss of the adapter. In literally 30 minutes I have an appointment to have some someone blast through a few adapters...I'm actually really proud of that design because it can combine the speed and versatility of the multifix with the easier to clean, easier to make adapters and higher tailstock clearance and visibility of the Tripan.
Always a pleasure.
Most certainly I appreciate your videos.Thank you
In statistical studies, particularly for QC and 6 Sigma type studies, we consider quartiles and median figures, rather than mean, to help account for outliers.
Always interesting stuff Stefan. Thanks.
Excellent content and I always learn something!
Another nice video Stephan, thanks!
I was looking forward to this! Multifix seems like a great system!
I found the last video segment interesting. A tapered cut, looked like low speed and a wide chip. Could you provide details on the type of material being machined, doc, and more on the tool doing the cutting?
As always, interesting and usefuls tips. Thanks. Stefan.
Interesting . I am not sure if you have tried the Moore & Wright style fixed end telescopic gauges , i find they are a step above the style that both plungers move when feeling for the size . Cheers .
I had some brand-X gauges - didn't like them, bought some Mitutoyo - not better! I could see lots of burrs under the microscope - no better than the Chinese. I think the problem is some of the players are just putting their names on things. I used some old ones once - nice an smooth - worked fine - but he wouldn't sell them.
I took the Mitutoyos apart - smoothed things with some pencil stones - still not right. Not sure you can fix the cheap ones - and the expensive ones may be cheaply made today. I think the tolerance of the pistons vs tube is sloppy - not something fixable.
TOT didn't seem to fix his either.. In production shops they just don't use these anymore, so it could be no one cares.
Great video! Really enjoyed the variation of topics. Stay safe my friend.
Tolles Video 😀 wieder viel Neues gelernt 👍Gruß Robin
Thanks Stefan, you've given me some ideas! I wanted one of those Multifix toolposts for my lathe but in the end they were just too expensive for me. Maybe when I get a larger lathe...
Most older german lathes come equipped with the Multifix holder. My Weiler, for exampåle, had the Hahn&Kolb version of the post. Now, some far east toll holders are NOT compatible with the "original" Multifix tool post. So, do not assume that the tool holder is compatible if the size designation (A, Aa, B, ...) is the same.
The two brands Stefan mentioned are OK. The AXA is made in Germany, The PwTools is made in far east under very rigid german control. There is a german phrase; "Trust is good, control is better". I bought my holders from PwTools as I was actually able to call the owner and ask whether the holders are compatible.
My last job we had one lathe set-up with a Multi-Flex. The only thing I did not like about them was the screws in the holders are very hard to find in the US. Not because they are Metric (or ISO if one prefers) but because they were 7mm or 9mm screws. I know they weren't 8mm. Of course the company was expanding their machining department and the new boss decided to restrict access to people he hired. I guess 45 years wasn't enough experience 🙄. But then the company started doing inhouse production work and the machining I did there was related to repair work
Repeatability of the C5 stop. Presuming your OD is the same on all the parts, would using a torque wrench not give you repeatable clamping force?
I quite like the British Dickson holders, but also Aloris because of its ability to do super quick tool changes. This is something cnc lathe manufacturers need to work on.
Hallo Stefan
Zwei Ergänzungen könnte man noch machen:
Es gibt noch einen kleineren Multifix als den Aa, nämlich den Aaa. Und man hätte das mit den zwei Stiften vielleicht noch erklären sollen. Dass der 2. Stift um 4,5 Grad versetzt ist, was einem zu den 40 möglichen Positionen des Multifix noch weitere 40 bringt.
Ja, bei den regulären Mfix gibts die nicht, deswegen hab ich die ausgelassen . Gibts nicht sogar mitlerweile einen noch kleineren als AAA?
Das mit dem zweiten Stift war mir nicht klar, man lernt nicht aus :)
Why do telescoping gauges have flat faces vs pointed, or even rounded to the radius of the minimum dimeter they are capable of measuring? It seems you have to wiggle them around to feel when they catch to take a measurement.
They have rounded ends - But still, they are not 100% self centering, it needs some getting used to it.
@@StefanGotteswinter ah, well still, a full round or point or wedge would save a lot of faff :)
I use a small Ifanger upr boring bar clamp for vee block clamping ,works well with no marking on the workpiece.
Interesting content as usual Stefan.Thanks.
A German using point for decimal instead of comma? You are being assimilated Stefan. Keep up the good work.
he reverted back in the second average :p
That is funny2,0
Thx Stefan! Great content!
Been waiting for this since you teased us in the last vid
Can you maybe add to the comments where you bought the tool post? I think I want one
Thank you Stefan, very interesting.
Thanks for sharing with us.
Gday Stefan, thank you the tips on telescoping gauges, I always enjoy you videos, I’m always learning, I think it was on Instagram you mentioned about the ikea lamps you use, mine arrived in the post today, I’m looking forward to seeing a bout better on the lathe in the future, thank you, take care, Matty
we have one of these german multifix toolholders at work. that spring always flies off on it. was ours made on a friday or something?
I guess that the telescopic gauges , where off at the end , creeping under 25.500 mm .
Due to the heating up of the telescopic gauge ..
Indeed the mitutoyo micrometer is well insulated , but the telescopic gauge wasn't at all.
reason its getting smaller (counter intuitif) , was because you measured a ID so when tips get hotter they touch earlier .
Is my throw at trhe repebilaty numbers .
I like such testing alot , weel done Stefan
grtzz from the Netherlands
Johny Geerts
very good video..thanks for your time
Great. Please show more how you use the Multifix tool post. Particularly interested in threading when limited to 9 degree increments. Cheers
With 9deg incremental rotation, assuming you set it up like Stephan did adjusting the boring bar to be aligned with the ways as a test, then you Always have 90 and 0/180 deg options for whatever op including external and internal threading.
What Stephan has done in line with his rigid, no compound slide technique is to carefully set that 0/90 orientation and dowel pin the 40 pos core block.
Even going as far as showing his first failed attempt..
I have worked 40 pos equiped lathes, with compounds, none of which have been dowel fixed, and you do find yourself having to adjust the position of the core block tooo often.
would 2 clamps on the V-block make a difference? If you put one on each end of the block it should pretty much cancel it out.
No Aloris in Europe? No multifix in N.America (except Adam Booth) it seems. Alway fascinating conversation.
enjoyed....great info
Always interesting, thank you Stefan :)
These are nice. I like my Dorian though. Very solid. Very accurate. Far easier to mount. Lots of block styles. Easier to keep dirt out of the tool and blocks.
Your videos are so interesting.
I watched a video about the mutifix toolpost that showed it made a difference in the direction you tighten the cam. Just happened to be the opposite way that you were tightening yours. It seems that the inconsistency of the way the cam was aligned. Wish I could give credit where credit is due , but I don’t recall the channel name.
Plot a histogram and it justifies chucking away the outliers that don't fit a Gaussian bell-curve.
Or just use the first measurement as it matched the average exactly ; )
[Found] At @8:48 you mention a podcast. Which one? "____ and Tolerance"?
Found it: "Within Tolerance" podcast.
Tom Lipton has a podcast "Within Tolerance".
Thank you for the multifix review, I have one in my basket ready to buy but out of interest most of us Myford boys use a Dixon/ Dickson quick change tool post. What would you say is there main disadvantage?
I had a Dickson for my Myfords but I gave it away because it was so bulky, the cutting force is too far overhung for my liking, the toolholders or bits have to be so shallow, and there is so little room to spare as regards clearing the tailstock and the chuck. I would opt for Tripan, ideally, but now I have bigger lathes, I only use the Myfords for instrument work and the like, so I've actually gone back to the Dutch Clog OEM clamp on my Super 7 & ML7 because it is so versatile and compact. With 16mm shank insert tooling the height is bang on, for most HSS I use Ifanger tangential tooling, which don't require packing, and for other tools, I ground a bunch of graduated packers from 3.0 to 4.5 in 0.1mm increments, and also a few thicker ones, and I write the height of the packer required on the side of each tool in indelible pen.
I worked out a couple of nifty techniques to quickly get the clamp surface dead level and I threw away the annoying spring. I also modified a deep socket to fit it with a sliding crossbar for the Super 7, and I cut down a cranked ring spanner for the ML7
On my bigger lathes I use Dickson, because that's what they came with, and second hand holders are readily available. I would like either Multifix or the American one with the clamping dovetail, but the Dickson (one genuine, one a good Italian Rapidue clone) are pretty solid and very trouble free for big lathe work, I find.
I just bought a T-slot style faceplate for my biggest lathe, and it would make a good stand for one of my Myfords. It weighs more than both of them put together. Come to think of it, it weighs more than I do!
Anyone reading this who has just upsized their lathe: make sure you have a long, strong "cheater bar" to do up the toolpost clamp nut. on a 400mm swing lathe it wants to be about 800mm long. On a bigger lathe, it pays to either use a commercial cheater bar, or if you make your own, make it with a heavy wall extension pipe to fit over the end to provide the necessary lever arm when you expect to take big cuts.
Lindsay Drabsch
Stefan I have retired from an engineering business after 50 years and this is not the way to use telescopic gauges .I have been using these gauges together with Swiss Tesa micrometers to measure bores .The way I was taught was to place the gauge in the bore at an angle ,lock it and pull it so the handle is parallel with the bore .The gauge will side to the exact size of the bore .The measurement will be repeatable to a micron on the Swiss micrometer provided the cylinder has sufficient wall thickness .This method of using telescopic gauges does wear the the gauges so I do have a number that are worn out .In business where mounting ball races or press fits it is necessary to get precision in the shortest time
You are correct in that my method wears the gauges in fact i have a box of worn out gauges
The precision that is obtained in measuring is worth treating the gauges as consumable
I use the gauges in all materials both aluminum and brass
It is important that the bore to be measured has a high grade finish and the wall is thick enough that it does not distort
The method I use is to complete the bore first while there is sufficient material that does not distort