Just to address some of the repeating comments: 1. I didn't come up with this analogy! I read about it in an engineering book that related it to a stress reduction technique used in the past. You can read more about the theory in Reference 3: Roark's Formulas for Stress and Strain, Chapter 17.2, page 777. I thought it's a crazy idea so I decided to test it. I uploaded the PDF for you here: drive.google.com/file/d/1JE9jCCAGj7MGXMqZQSjOo8KS8XlC5y4D/view?usp=sharing 2. Yes, the method is used for tension members specifically not for beams but my home setup does not facilitate tension testing so I had to improvise. The point is to show that removing material can increase the capacity. 3. Of course this is not a peer-reviewed study so the results should be taken with a grain of salt as there is a lot of variation between the samples, they are way too few, and the testing environment is not strictly controlled. 4. Not many holes are elliptical in practice. In theory this should work with circlar holes (cables, pipes, ventilation, etc) as well, but the capacity gain is probably much less, if any. I used an elliptical hole to make the gains more drastic and hence more interesting for a YT video. I encourage the comments pointing things out, this is great! I like the idea of community notes, I hope it comes to YT as well. Cheers!

Why did you demonstrate potential flow theory on a beam loaded in tension when your test is a beam with a transverse point load? They're separate loading conditions that require different analysis. Furthermore why didn't you repeat your initial test multiple times? Wood can vary significantly in strength so using only a single point of reference does not make for an accurate test

Interesting idea, though your slit in the first beam has a lot sharper edges (cat eye shape) than the ovals in the following 3 cases. Also given that you shared stress model for the single opening, it would be nice to see what your modelling software predicts. As for the conclusions, note that your intra-sample variability ( 1929~2140kg) are consistent with standard wood variation and rend the results of your experiment inconclusive. Finally note that the top fibers being crushed before you reach failure mean you are observing variability in wood fibre separation ( delamination ) rather than stress propagation. I am looking forward to seeing a followup, keep up being curious ^_^

Wood is far from an ideal material for these tests, since it is grained and also nonhomogeneous. For a more sound experiment you would have to repeat the test many times due to the variation in grain patterns. Since the model (beam under tension) differs from the experiment (bending load), preferable would be a more homogeneous brittle material under tensile load, such as concrete made with small aggregate. Also good to know is that ductile materials are not affected by (static) stress concentrations, since they deform locally at the site of concentration and redistribute the stress evenly throughout the zone. A ductile beam with a notch or hole is weaker, but only because of the lack of material. Smoothing out sharp curves and corners won't strengthen them in the same way it does for brittle materials, at least under static loads. I really liked your video and I think it would be really cool if you made another one that shows the effect for ductile vs brittle materials.

Your loading doesn't match you stress analysis. So tell me again how you can increase tensile strength by reducing the section?

Really like the fluid flow analogy.

The same idea is very important in the fatigue design of parts. Where comparingly small decrease in peak stress can increase lifetime several times. I had a patent application with this idea in the construction industry for fatigue sensitive parts.

Top quality 👌 really cool analogy and I love the new style with bench testing. Keep it up!

That is pretty cool, thanks for sharing. Charles

This is an excellent video.

Excellent use of the word "comprise"!

In terms of a video, I think it would have made the point better if you started with the hole in the middle. Otherwise we are sitting through the whole video expecting it to increase the strength relative to the whole beam.

Interesting. You had me there though. I originally thought you were going to propose, that a beam with relief voids was going to be structurally stronger than a solid beam. For anyone who is going to point out the cost weight benefits of non-solid beams. Yes, I know, I have the ability to look at cranes, bridges and aircraft wings.

I am not sold. Wood is ridiculous for it's inconsistency. To make it at least somewhat scientific you would need to make more than one test with just one hole. Even better: use a solid such as engineered plastic or something.

Wow that’s amazing insight, very counter intuitive but brilliantly shown, well done.

The consideration at 3:50 is wrong and it is often done on wing profiles. There is no reason for a particle to speed up exactly to cover the same horizontal distance, there is of course an increase in speed, but not to that value.

Guitar builder here. I always wondered about this when making transverse struts that would bear the load of the string pressure on the bridge. That part of the guitar top, which acts as an monopolar oscillating plate supported by beams, needs to bear load (about 80 lbs) but also be as lightweight as possible (unsprung mass?). Could this be an improvement? I guess an experiment is about to be born.

I like engineering I can understand without resorting to mathematical notation that I have never understood. Because of that, I subscribed which is something I have never done before until I see multiple video's

Pretty much one of the first optimizations we did in structures class, a cantilevered wing spar with distributed load.

Dude, do more testing please

Wow !!! That is sooo counterintuitive - and really set me thinking . Video saved - I shall definitely revisit this . What a surprise … thank you for making this . I almost can’t get over that This would imply you could strengthen a joist after a plumber has put a pipe through it by drilling extra holes !! I guess in buildings strength may not be the limiting design case however, where absolute deflection under a given load may well be more critical, and indeed must be lower than a prescribed amount so as to prevent damage to attached brittle materials, and the more nadgered a beam is, the more it will deflect (long before failure). Presumably the coupons with extra holes drilled are deforming more for a given load ? Otherwise why wouldn’t all plumbers do this as a matter of course, and more importantly it be built into building regulations ?

very very interesting!

We need to consider this material, wood, is a composite structure, and have different properties depending on the direction. A test with metallic will be interesting.

You did a great job on this video and obviously put a lot of work into it. Nice! Don't get caught up in everyone correcting things. They tend to do that on the internet.

I don't need to finish watching this video to know this is wrong. Your hack job on cutting the holes is a good representation of your experiment.

5:51 - 5:59 Than you. "Natural variability in the wood" was my first thought.

I don’t see that the strength increased with the addition of holes compared with the original lumber, rather compared with the piece with the first hole drilled into it. And it should be noted that you need to compare strength to weight ratios, as that is really what you’re comparing. Less weight due to more holes drilled - as opposed to the original solid piece, which being heavier, was also stronger.

Wheres the test on the drilled beam seen in the thumbnail? I'm not sure the term "flow" on a static material when the reaction to the downward force imposed is omni directional radiative with longer or shorter felt-force vector arrows, but it is a fascinating experiment

Horizontal shear is a failure mode most common on short, deep, heavily loaded beams (bending members). Round holes can relieve the strain paths. Holes with reentrant corners are crack propagation points in any material.

I think the point is... All real-world construction will require holes in structural elements and thoughtfully placed additional holes may improve the strength. (Not that a beam with holes is stronger than a beam without holes)

At end supports vertical shear stress higher . In center 90% of stress is in outer fibers .

Interesting analogy, I didn't know this approach. Would the beam perform better or worse if the holes were drilled near the bottom instead of the center?

Without multiple test of sample one and two it does not make the samples three, four and five very convincing. Wood being a natural product means there are lots of inconsistencies in it's strength even pieces from the same board because of knots and variations in growth rings that could have been caused by injury to the tree while growing or any number of other factors. Besides the oval holes where inconsistent which would have stressed the board differently for each test.

I think some are being overly critical in the comments. Yes the experiments would have to be a lot more detailed to have scientific validity, but the overall conclusion is correct. Also, although technically the wood failed first at the press contact point by delamination, we were still able to see the failure on the tension side of the wood block. The stress flow evidently is much different from the examples given, as this is a bending load. However, this was addressed in the video and it still holds that the ellipse concentrates stresses on it's sharp edges, and that the holes may help distribute the flow more evenly My takeaway was that the main point of this video is to show the counterintuitive result that taking away material can make the structure stronger, which is absolutely correct. And i believe it's not reasonable to expect a super detailed experiment on a simple youtube video like this. The use of potential flow as a theoretical justification for why this works is also correct, even if the loading condition isn't the same. And the experiments illustrated your point even if they weren't perfect. I enjoyed this video a lot! I think its also important to say that this does not suggest that adding these holes is optimal or good or desired in a real structure. Real structural solutions often have better ways of reducing stress concentrators. This result, however, does show up a lot in real life -- not by intentionally removing material to make a structure stronger, but by adding material and unintentionally making a structure weaker -- engineers need to be aware of this kind of thing

Interesting counterintuitive results! I wonder if this would work the same if the initial hole was filled by a bolt to attach another member.

Isn't one of the advantages of the flying buttress? The others are mass reduction, increased distance of load from the base of the building, reduced liquifaction of subsoils, and a graceful aesthetic. Gaudi's cathedral, La Sagrada Familia, Barcelona, comes to mind.

Interesting, but why make an elliptical hole when 99.9% of the time the shape cut through a beam is gonna be circular?

Interesting. I would like to have seen multiple tests with a single hole and with no holes just like you did with the last 3 samples to get more normalized baselines.

And locating a center hole in a beam at midspan is the BEST place to put the hole to reduce the loss of strength. If you locate the hole near the supports, you will see a very dramatic drop in strength.

That’s really interesting, I guess the holes let the wood flex more which increases the fracture threshold

It would have been interesting to apply topology optimization in the stress analysis to compare the resulting geometry with the optimal fluid flow theory discussed in the video.

Stop drilling cracks in aircraft alluminium skins can be thought of similarly? Thank you Most satisfying.

It is to small a veritable wood 😂 need more tests for me to believe

So, are the holes, if we are to take the results on face value, redistributing the stress in the material so that while they take away from the overall capacity to take load, because the beam is already compromised in a very specific area, they move stress away from the point where it will inevitably fail first?

If anyone has ever seen what electricians do in order to run cable then you would know that single holes are drilled through many studs and joists in your house (assuming that it's made of wood). 😅 If relief holes accompanied said holes then you would have a stronger structure. 🤷 I think it's at least a cool concept. Thanks for sharing ✌️

Misleading title when all the drilled wood fails over 100 pounds before the default 2x4. Although the strength increase you bring to our attention is intriguing physics.

Yes. But the holes have to be in the same location as static load nodes are. However, making extra holes is not a good approach with transient loads. Wind and snow transients usually defy the benefits on building construction applications.

So the takeaway is, that you shouldn't put a hole at all in beams under tension. But if you have to, make the area around it weaker also. Crazy chaotic video on so many levels. 😅 I have to forget about this now and go back to my workbench.

So this is why the toilet paper never rips at the perforations!

Thanks for this interesting video and the great analogy. Yet, I am a little bit confused by how the shape of the obstacle should reduce the velocity around it. Given than flow J is equal to velocity v times cross-section area A, the only thing that should matter for the maximum velocity of the liquid (i.e. the maximum stress of the material) is how much wood is left around the hole. Any flaws in my reasoning?

Works great in metal parts that need some flexing.

Yeah those samples are from the same beam, but they have knots in them and different grain / growth ring alignment, so not much luck with using wood for modelling this complex load situation here. Also confusing how all of the analysis is about tensile load but the testing is done with a bending load, which makes for a hybrid failure. Props for the effort, though I don't know what exactly I can take away from this.

Turn the oval 90 degrees and place two of them beside two small round holes stacked in the middle instead. basically an inversion of what you were testing. If the goal is to increase strength by removing material strategically, then goal post is beating the non-altered 2x4. None of your tests did that.

The results rely on the crushing snd failure of the wood before the final force measurement. In a solid plank the crushing force causes a long crack through the bulk. In a plank with a hole the crack only propogates to the edge of the hole and the bulk is compressed which makes it stronger in failure. In some cases the void can allow the bulk to act like a lever, spring, or damper. This design would be excellent in something like a bench since the failure mode can take more load after deflection, softening the impact to users. So! Use a bigger plate!!!!

The drill holes looks like the Adamas epaulet sound holes on the top of Ovation guitars. Just an observation but somehow there's a parallel with this video because of "flow".

I wonder if a more elastic material would be less subject to this effect than a morr brittle material - able to elastically redistribute load without the extra holes, reducing stress concentration points.

interesting results, what is clearly visible is that sample 2 failed differently than 3,4,5. 2 might have just been a bad sample or the different failure mode is really because of the extra holes?🤷. even though the setup is not perfect I respect you for trying it out, what i dont like is the short length of the beam, because it clearly impacts the results of 3,4,5,(1?)

So the releaf holes make it so compression and tension forces spread evenly on the remaining material?

How about the wood distribution strenght? All the wood surely have different fiber patern... does it have effect?

If the holes were smooth rather than ragged as it is in this test it might have a higher load potential

Replication studies please.

Im assuming that the test might have been better if the oval shape was perfectly cut and not hacked out? U left sharp edges on the internal curve which should have had a perfect radius and not jagged edges. Maybe drill holes at the top and bottom of the oval to start the shape with nice radiused shapes?

I feel like this is how fluting began😂😂😂

A more rigorous analytical approach where this kind of thing shows up is generative design

Why not drill circular holes for uniformity? Also, what about a single circular hole off center? It is a common rule of thumb that holes near or at the edge weaken the beam the most

Your experiment is absolutely incorrect. 1) You make the big hole differently every time 2) Different beams are different in load capacity because it's wood 3) Only 1 beam tested with no holes and with one big hole.

Try holes with aerofoil cross-section.

Well since all the holes grouped together are a horizontal elipse why didn't you trid eliptical horizontal holes?

Would be more meaningful with two point loading at third points to produce maximum flexural and zero shear stresses at midspan.

You're saying the Entlastungsbohrung is real? As many have said, you need to repeat your experiments many times over. And then you also need to calculate the expected value from your sample size and deviation to be able to state if something significant can be achieved. The mathematical "significant", that is.

Sus. How certain that the beam with single hole isn't a negative outlier in the test?

The first solid beam with just holes could go beyond 2300?

Why would pressure you’re applying be 90 degrees different to the flow lines? Would the flow lines not be representative of the downward force and therefore need to be aligned with the force direction? The video didn’t mention the discrepancy which makes it all not make sense and seem like either a mistake or an important detail glossed over.

One can be stronger if the pieces were already different before you drilled the first hole.

would love to see the effect on a steel member

I guess I missed why the first hole was oblong instead of round. I guess I also missed how there is any advantage to holes at all, being the first beam fractured at the highest level. Seems to me, no holes is the way to go.

Let's agree on the fact that there are no two identical pieces of natural wood. Thus, all this is just an anecdote: once i have cut holes an it held more

Joke from Soviet era: Engineers are developing the first soviet supersonic aircraft. But on all prototypes wings keep tearing of the fuselage. Chief engineer Mykoyan stays late in the office but he can't figgure out any solution. A lady cleaning toilets and rooms comes to do her job. She starts clean the floor. When she gets to mr. Mykoyan, she askes why he is there so late, so he explains her his broblem. And she replies: "tha's simple. just dril even spaced holes along the line of break..". So they try and it works. After success, mr. Mykoyan goes quickly to the cleaning lady to give her thanks. Ad he askes how did she came up with such solution. She replies: "Well boy, theese are years of praxis. Look at our soviet toilet paper, see there theese lines of holes - it never breaks there..."

The conclusion in my opinion is wrong for all the listed parts by the other commentators. 1.) The material under test is non-homogenous 2.) the cutouts are different, thus the stress concentration points are different 3.) Flow can be used as an analogy only for easier understanding, otherwise it has no relevant similarity. The forces in flow diverging around a corner and the stress/strain concentration have no common ground. The only thing relevant in this whole video is that you can reduce the overall weight of the beam by removing material and still retain the majority of the load capacity. This is only due to the cross section. Cross section at the loaded points is the only thing you need to focus on. The rest is just nonsense.

This is just a rewarmed Bernoulli again. Let's see where it goes. Nowhere?

Looks like the big oval had less cross sectional area in tension on the bottom of the beam. Nothing to do with fluid flow.

So the result is, don't drill out any holes for the strongest result? Got it

2303kg

Strenght = surface area.. Given the same mass, body 1 with more surface area will be stronger than body 2 with less surface area. Ex: a 1kg/meter pipe of a given diameter will be stronger that a 1kg/meter solid bar!

Comparing the size of the center hole on the first sample and the additional samples: The first hole is obviously taller making the web on the bottom thinner. Not saying that that invalidates the experiment just that you need to have better controls on your experimental set up. Measurements on width of the bottom web. Radius of the notch all of those things can have a big impact on the results. Also flow is not really similar to stress. The top stress is compressive the bottom stress is tensile. Stress in a beam is zero on the center axis. Flow in a pipe is maximum in the center. So flow analogy is a poor representation for stress. By the way I am a retired professional engineer.

TLDR: Sharp corners bad, large fillets good.

Hello, I am a Timber Engineering faculty member at Virginia Tech. You make some good points, but you there are some probems with your content. First, your model is not the same as your experiment. Beams do not have a uniform tension force. They have a triangular stress distribution where the moment is greatest at the center and typically tension failure is dominant in brittle/semi-brittle materials like wood. I also have issues with your sampling of wood. Saying short samples from the same 2x4 have similar performance is not correct. Strength of wood is dominated by the placement of defects like knots. Locating knots in different places can radically change the strength. I also think you have a flaw in your sampling. Typically, 10-15 pieces are tested for material properties and more for connections / special cases. The flow idea is fine, but I think it is more of a visualization concept. I'm not sure if it is linked to fracture energy, which has the same idea of a more rounded curvature to prevent failure.

But WHY? Why does this analysis work?

This could simply be from allowing more flexibility which you do not want in a floor

Please normalize your audio. Loud bangs and your quiet voice do not make for a comfortable listening experience

Let’s look at the technical reasons for this … (explanation accelerates into hyperspace)

Yeah, but no. Just looking at your samples one can see large inconsistency in the grain and then knots also affecting it's shear strength. Then throw in the jagged inconsistent oval cuts in the samples. You should repeat this experiment with a wood species with a more consistent wood grain such as ash or sitka spruce. Also pay closer attention to finished hole it should have a smooth perimeter and consistent shape between the samples.

Yes, indeed very counter intuitive. I shall now the respectful when a see a bunch on idiots attacking a bean with drills

Its like a toilet paper....it never fails on the perforations!

Im SOOOOOO gonna start drilling relief holes in all my floor joist where the electrician & plumbers drilled holes to run their stuff. Im gonna copy your layout as you've presented yourself as a professional, and Im SOOOOOO gonna include a link to this video in my building permit.

It’s not counterintuitive to me. The oval is one of the strongest shapes against stress. Just think about eggs, for example.

What's wrong with NOT drilling holes? Still a 25% capacity gain over the elleptical slot. It's like punching a hole in a wall to make it stronger. Pretty sure that's NOT how it works. Not sure the fluid dynamics analogy works either, too small of a sample anyway. Still nice example of experimental thinking. Congrats!

And that was supposed to be the simple explanation?

The title of the video is "Can drilled holes make your beam stronger?" well you demonstrated that. However, wood being a fibrous material, those coloured plots don't represent the truth. There is a lot more at play.

3:50 please do not perpetuate this idea of fluid having to speed up so it's in line with the rest of the flow. It's plain wrong - that does not happen in real life.