Building Prototypes Dan Gelbart part 10 of 18 Flexures

[@joshdrexler8773]

I worked for a company that made middle frequency vibrating X-flexures as an alternative to mirror galvanometers. The product never found a market. One unexpected headache was persistent spalling at the point where the flexure was clamped. No amount of clever clamp design, such as the curved (hyperbolic?) clamp ends you demonstrate, ever effectively solved the spalling problem, limiting "infinite lifetime." Another problem was that no matter how high the Q of the flexures, there was never any adequate way to isolate the flexures from external sources of vibration, in real world applications, to the desired level of noise rejection. On the bench, yes. In machinery like a helicopter, with dozens of high magnitude vibration sources, not so much. Also, obtaining consistent flexural material from vendors, batch to batch, turned out to be non-trivial, since blanks needed to be not only dimensionally consistent but also alloy-composition consistent and internal-stress consistent as well. While nowadays I suppose the problem of designing flexural machines is readily solved in computational software, back "in the day" designing cross-flexural vibrators to precisely hit a combination of frequency and inertial targets was an arduous exercise.

[@wwjjss33]

Thank you so much for sharing these important details. I take it the spalling is avoided when the flexure is fabricated monolithically, but how would you approach it now? Was it a surface phenomenon of the flexural material? Or could it be circumvented with a polymer clamping shoe? No one else has ever drawn my attention to this issue--again, thank you.

[@edwardangus9436]

Interesting - I would anticipate fretting fatigue failures at the point of fixture, but not spalling.

[@LAB67_World]

Look into Prof Jon Hopkins at UCLA for methods to reduce external sources of vibration, multiple methods to do this.

[@dgelbart]

Did you use Nitinol for the flexures? I can see it happening with spring steel.

[@VKRenato]

Thank you AVE, it's Excellent channel.

[@captcarlos]

I second that, thanks AvE!

[@proletariatworker7622]

I third that AVE.

[@-Sean_]

I fourth that, thanks AvE!

[@MrMultizaf]

@@-Sean_ I fifth that Thanks AvE

[@steelcannibal]

It's skookum as frig!

[@OccamsRazor4eva]

This stuff is pure knowledge gold. I hope Dan gets cryogenically frozen and reanimated when technology has evolved so that future generations may learn from him! 😉 The world must be a very interesting place when you get to play with laser optics at the micron scale. This makes me want to give up my current profession to pursue something similar. It strikes a perfect balance between the creativity of prototype design, culminating in seeing your creation actualised physically through self fabrication. Shalom & Kudos to you Dan.

[@mxcollin95]

Mark Slatem absolutely! This guy is and his mechanical creations are amazing!

[@douro20]

This guy has done extremely well for himself. His company revolutionised the world of printing in the early '90s with its development of the first single step computer-to-plate production system. This technology directly engraved the aluminium printing plates with a high-powered laser, completely skipping the process of photolithography required for traditional digital imagesetting. The process saved both time and money in large scale printing processes and even allowed for recycling of used plates.

[@michael-rommel]

Finally!!! I foud a video, where somebody explains the reasons behind the geometry!!!!!! Nice!!

[@pirke86]

I found this channel by coincidence. I love the movies you post, Construction principles was always my favorite subject in engineering school. Keep up the good work!

[@conraddiaz2274]

ELEGANT complexity! Love it.

[@peterthinks]

If you put your aluminum flexure in a tub of water would that act as a buffer so it would not drift? The amount of energy you would have to put into the water to change the temperature would be much greater than the energy needed to distort the metal. If the water was frozen you could also take advantage of the energy required for phase change.

[@evanlane1690]

Excellent video and information. Thank you for sharing!

[@praetorious]

Hello Dan, When you demonstrate the sensitivity of the aluminium flexure to radiated heat from your hand, is it not possible that some of the deviation shown by the capacitance sensor is due to changes in capacitance from your body?

[@dgelbart]

praetorious The flexture is grounded to prevent this (you can actually see the grounding wire wrapped around the sensor cable), but there still could be some small electrical noise effect. The majority of the effect is radiated body heat. Problem is worse than simple linear expansion, as non uniform radiative heating distorts the part.

[@fariedonshaheed2823]

Your videos are very informative, and I’ve learned a lot from them. Could you please tell me the thickness of the long-range flexure mechanism?

[@dgelbart]

About 0.5mm thick.

[@fariedonshaheed2823]

@@dgelbart Thank you for your reply. How long are the nitinol sheets? Are they made from Nitinol SE508?

[@dgelbart]

@@fariedonshaheed2823 I just got them as strips about 12mm wide, they also come as small sheets about 150x300mm. Don't know the type of Nitinol.

[@damienmiller]

Did you build the capacitive sensor yourself?

[@dgelbart]

I did not build the one in the video, but I built many similar devices. They are easy to build and are great. Interestingly, the accuracy is not affected by the flatness of the object, even if you measure to a curved surface.

[@jeetenzhurlollz8387]

Awesome..i learnt so much in 12 minutes!!!

[@TheNefastor]

When you say that flexures have no wear / infinite life : does that take into account metal fatigue from repeated cycles of bending the pivots ? I admit I'm a newbie when it comes to mechanical engineering (I'm an electrical engineer). Is the bending so limited that it does cause fatigue ?

[@dgelbart]

The fatigue limit depends on the metal and on what % of the elastic limit you are using. If you design it right, life is infinite. Think of the hairspring in a mechanical wtch.

[@TheNefastor]

@@dgelbart ah, I see. Thank you !

[@dgelbart]

@@TheNefastor Don't worry, I'm an electrical engineer too, but I often impersonate a mechanical engineer.

[@TheNefastor]

@@dgelbart we're quite a lot. I think it's because we all eventually realize that we can't just make circuit boards, but also need to make something to hold them in place. That's how it started for me, and eventually it evolved into robotics and DIY CNC machine tools...

[@RBAERO]

Make Flexures with a water jet the tapered cut the water jet make is not a problem?

[@dgelbart]

On thick plates the taper of a waterjet is very low (and you can cancel it out by slowing cutting speed). On precision flexures, like the aluminum block in the video, I drill and ream all holes at the end of the flexure arms first, then connect the holes with the WJ.

[@abdh621]

please upload all parts for the Flexures. I am learning a lot from you.

[@puddingpimp]

Wow, these flexures are really inspirational! Thanks for demonstrating. Regarding the laser alignment technique; I've used this before for aligning other fixtures. Here's an improved method I use: If you place a mirror on the fixture and several mirrors on the wall you can measure even smaller graduations as the angle error is multiplied by two each time the ray returns to the fixture. At some point this becomes untenable due to the losses and lack of flatness of each mirror, or because vibrations turn your fixture into a seismograph. The alignment isn't that difficult because you're only ever aligning the tail end of the laser path back onto the target mirror; unless you knock an earlier mirror or the test fixture, then you've got to start all over again. You can probably get much better results using a vibrationally damped optical workbench and first surface mirrors, but I'm not a millionaire or research scientist, so I used broken up CD-ROMs or bits of makeup box mirrors which worked ok for 5 or 6 multiplications which was enough to make vibration the limiting factor. 2**6 = 32 multiplication * 12 path lengths * 20m, to get the same angular resolution without this technique you would need a 7680m pathlength, and you wouldn't be able to see the laser spot anymore.

[@jeetenzhurlollz8387]

hello, if the screen that is receiving the laser pointer is inclined? Will the pointer movement be multiplied the more the screen is inclined close to parallel to the laser beam?

[@puddingpimp]

Yes. Beware that the magnification will only be in one plane and it will also magnify the aplanarity of the screen.

[@jeetenzhurlollz8387]

indeed...good point

[@dgelbart]

puddingpimp The magnification will actually be the same in both planes. Also, no need for several mirrors, just a larger mirror. Point the laser pointer to the distant mirror.

[@886014]

I don't know if Dan Gelbart monitors the comments, but would be grateful if anyone could advise the dimensions of the nitinol strips used in the flexture featured at 5:30 etc? Are these Nitinol strips available commercially in relatively small quantities? I'm not having any real luck finding where I might get these stips, especially in relatively small quantities.

[@dgelbart]

Nitinol strips are about 10mm wide, 0.5mm thick. They are readily availavle but very expensive, better to look for it on Ebay. If no luck try Nitinol suppliers like memry.com.

[@886014]

Thank you, that was very much appreciated. I can only find the Nitinol as wire rather than strips so I will experiment with regular carbon spring steel and account for the lower elastic range. I hope you can be encouraged to publish more on KZbin. I go back and watch your videos again every 6 months and always learn something new each time I watch them. Flextures are a particularly elegant solution.

[@movax20h]

I really like these videos, and I watched them some time, ago but keep coming back. I was wondering, who invented Flexures and when?

[@dgelbart]

Flexures are a very old idea but the person who popularized them was R.V. Jones in his book "Instruments and Experiences" (a must read).

[@eduinmartinez5256]

Maestr buen día en qué se le da utilidad a esos dispositivos me causó curiosidad gracias maestro

[@johnhildebrand8635]

Thank you for the videos, most informative. I would like to attempt to build the two flexures described in your video. Could you post instructions as you have done for the laser pointer. Thanks again for the inspiration.

[@dgelbart]

+john hildebrand Sorry, too busy.

[@rodrigobernal2130]

Can you recommend a book on flexure design?

[@dgelbart]

+Rodrigo Bernal The best is R.V. Jones "Instruments and Experiences", out of print like all good books but e-book or library may have it. Serch on Google.

[@mitchberkson4235]

Why does the wiggle bar have a thicker part in the middle?

[@dgelbart]

Mitch Berkson To have stiffness in the longitudinal and the cross-flexing directions. In all flextures the trick is to make it fkexible just in the direction it needs to be.

[@richardbaker4690]

Thank you so much for making these videos. They are great! Could you clarify what say at about 3:15 about the temperature sensitivity of aluminum. Thank you again for taking the time to pass along such valuable information!

[@dgelbart]

Richard Baker Aluminum expands with temperature about twice as much as steel, so precision parts made from aluminum are more temperature sensitive.

[@chrismr3972]

@@dgelbart Is it possible to use the reversal method to self compensate for temperature change? Not for differential temperature change, just working temperature of the flexure. In much of the electronics I've been involved in a differential approach is used to compensate for things which affect two paths - the Aluminium flexure was "single ended".

[@dgelbart]

Possible, but since temperature effect is repeatable and predictable, it is easier just to measure the temperature and calculate the error.

[@still_curious6388]

@@dgelbart lol.. thanks you answered my next question in advance :)

[@davidgorczyca5675]

Hi Dan, thanks so much for your videos, I watched them all in sequence while in the airport and intend to watch again. I am curious to hear any vignettes you might have about using 3D printers in your work and would love to hear your thoughts on technique. Is there a place in your prototyping workflow where you have found 3D printing useful? I do not currently have access to a waterjet, and am working on 3d printing of flexures. Thanks for the inspiration!

[@dgelbart]

3D printing is becoming a useful tool, but unfortunately for plastic. Most of my life I worked in high performance/high accuracy stuff, were plastic is useless. However, 3D printer is great for making wax patterns for investment (lost wax) castings so I recommend having one, especially at todays low prices. Metal 3D printers still have payback only in certain application, even if you have the funds to buy them.

[@lekoman]

David Gorczyca In the event you didn’t see it, Dan just did a new video on 3D printing last week. Worth a watch!

[@mastervivi490]

i love the phrase "wiggle bar"

[@alexboensch9373]

Where would someone go about finding prints, dimensions, or any kind of resources about building one of these, specifically the first one shown made out of aluminum? Thanks in advance.

[@genehuang2478]

You are amazing

[@t76d777f6d]

Hi Dan, when you say the angle is the integral of linearity, are you speaking conceptually or can this be stated explicitly? Just trying to follow.. thanks!

[@dgelbart]

taiden If you have an angle "A" along a length " L" the total vertical shift is AL. As the angle keeps changing the path shape is the intergal of the angle over distance. Angle is expressed in radians. If you never took calculus there is no problem, just divide the path into short segment , measure the angle for each segment and calculate AL (which could be positive or negative) for ech segment. From these values you can draw the profile.

[@t76d777f6d]

Dan Gelbart I follow now. Thank you for the concise reply. Enjoy your weekend!

[@daviddevenport4865]

Beautiful!

[@falkon314]

Sir, thank you very much on this subject, very educational! Would you please recommend any books on flexure for further studying? If you have a book, I would certainly want to read. Thanks again!

[@dgelbart]

+Fang Kong best book is "Instruments and Experiences" by R. V. Jones, but it is expensive and hard to get.

[@falkon314]

+Dan Gelbart Thank you very much, I really appreciate it!

[@littlestworkshop]

Hi Dan, What do you think of using carbon fibre sheet for the flexures? Ideally unidirectional I guess, I realise they can't be spot welded but I can CNC carbon easily and I'm thinking that the parts could be bonded with a suitable epoxy.

[@dgelbart]

Can be used, bond line will probably open over time, but joints can be riveted.

[@bvsiness]

Excuse my dummy question, but what is flexure for? I found this and suppos this is related en.wikipedia.org/wiki/Linear_stage#Flexure but I dont get the point. Is there some application to look at? Thanks in advance

[@stauffap]

+bvsiness Yes. They are used in pulse tube or stirling cryocoolers for example. It's used to hold the piston in position, that oscillates back and forth. The friction is very low, there's no wear and there therefore no maintance required. This is almost required in cryocoolers since you can't open them. It needs to be completely sealed to hold the helium gas. Furthermore since the flexure bearings keep their position so well, they use it to maintain a tiny and constant air gap between piston and the outer wall, this also reduces friction and wear. Cryocoolers really would have a difficult time without flexure bearings. But i guess, there are many more applications. Since it can be used as a spring and as a guiding system/bearing.

[@karthick86c]

Thanks a lot Sir. I have been working on flexures and your lectures are an invaluable resource for my study. God bless you.

[@abdh621]

Thank so much you Sir.

[@bigjoshlevine]

Wow, so helpful. Thanks!

[@BHARGAV_GAJJAR]

I'm looking for someone to build my new flexure mechanisms

[@duilddome1730]

благодарю за знание

[@NOBOX7]

dan yuo should monotize your channel , your missing out on thousands of dollars , your content is gold

[@dgelbart]

I got enough.

[@bobbyintheshop]

@@dgelbart haha, love your videos, and there is even more gold in the comment section, but this is by far the best

[@GaryGrumble]

Take a look at Dan Gelbart’s video on Flexures. kzbin.info/www/bejne/hpLcoZaMe7Wmapo He is making joints such as yours in metals. He also has several other videos that are amazing.

[@platin2148]

I’d probably do that in wire EDM not the waterjet

[@andyvan5692]

so what are flexures used for, just looks like a piece to brag about, how intricate a piece of art you can make, as it doesn't look like it does anything!!

[@dgelbart]

Most precision instruments use flexures. Just look inside some.

[@plnmech]

Please pardon my ignorance but you have told me how build a device that I don't understand what it does an what it's purpose is, please educate me? I have watched all of this series so far and it's very informative and I like your presentation style.

[@the_grand_blooms]

Flexures are for really high precision positioning, so think of things like optics and scientific equipment. Normally precision increases the price/difficulty of manufacturing, but flexures are a way of having extreme precision from a device that is easy to make. One example is openflexure, which is a microscope where the x,y, and z ways are all one (3d printed) part, so the microscope has very precise focusing but is made using an imprecise manufacturing process. You can read about that here: aip.scitation.org/doi/10.1063/1.4941068

[@alexchiperi]

@@the_grand_blooms Thanks for the read.

[@charleshendry5978]

Thanks again!

[@neamtialin]

Ok I'll stick to file and hack saw for the moment