Mechanical Advantage is a Myth

We all know friction in a mechanical advantage block and tackle pulley systems can reduce output, but I haven't heard anyone talk about stretch. I am always using semi static or climbing ropes, which have a lot of stretch. We test all the way up to a 13:1 with dynamic, semi static and 0 stretch dyneema. We did NOT use multipliers or brakes to simplify this already long experiment.
*All numbers are in kN. Trying to keep a lot of data easy to look at!
This unlisted video of our 81:1 (that doesn't work) inspired our "9:1 is not 900%" video • 81 to 1 pulleys for br...
►►►DYNAMIC ROPE
►1:1 pulleys n/a - 1.34 INPUT x 1.19 OUTPUT / 100% vs Actual 88%
►3:1 with pulleys - 1.13 INPUT x 2.26 OUTPUT / 300% vs Actual 200%
►3:1 without pulleys - 1.19 INPUT x 1.99 OUTPUT / 300% vs Actual 167%
►5:1 with pulleys - 0.89 INPUT x 3.84 OUTPUT / 500% vs Actual 431%
►5:1 without pulleys - 1.39 INPUT x 2.38 OUTPUT / 500% vs Actual 171%
►9:1 with pulleys - 0.85 INPUT x 5.92 OUTPUT / 900% vs Actual 696%
►9:1 without pulleys - 1.12 INPUT x 2.38 OUTPUT / 900% vs Actual 212%
►13:1 with pulleys - 1.01 INPUT x 5.04 OUTPUT / 1300% vs Actual 499%
►►►SEMI-STATIC ROPE
►1:1 pulleys n/a - 1.62 INPUT x 1.26 OUTPUT / 100% vs Actual 78%
►3:1 with pulleys - 1.31 INPUT x 3.46 OUTPUT / 300% vs Actual 264%
►3:1 without pulleys - 1.72 INPUT x 2.48 OUTPUT / 300% vs Actual 144%
►5:1 with pulleys - 1.10 INPUT x 5.15 OUTPUT / 500% vs Actual 468%
►5:1 without pulleys - 1.18 INPUT x 2.67 OUTPUT / 500% vs Actual 226%
►9:1 with pulleys - 1.31 INPUT x 7.48 OUTPUT / 900% vs Actual 696%
►13:1 with pulleys - 1.4 INPUT x 6.98 OUTPUT / 1300% vs Actual 498%
►►►DYNEEMA ROPE
►1:1 pulleys n/a - 2.33 INPUT x 2.07 OUTPUT / 100% vs Actual 89%
►3:1 with pulleys - 1.34 INPUT x 4.04 OUTPUT / 300% vs Actual 301%
►3:1 without pulleys - 1.47 INPUT x 2.97 OUTPUT / 300% vs Actual 202%
►5:1 with pulleys - 1.67 INPUT x 6.71 OUTPUT / 500% vs Actual 402%
►9:1 with pulleys - 1.47 INPUT x 9.62 OUTPUT / 900% vs Actual 654%
►13:1 with pulleys - 1.51 INPUT x 8.54 OUTPUT / 1300% vs Actual 565%
►►►THOUGHTS:
*** % wasn't that different but my input is so much greater not fighting stretch
***9:1 is better than a 13:1 and might be worth using, but a 5:1 is the most optimal to get as close to theoretical as possible
This is a very helpful article overtheedgerescue.com/pulley-...
👉 Learn and SHOP at www.hownot2.com/shop
👉 Best EMAILS on Earth: www.hownot2.com/signup
👉 SUPPORT US and get gear discounts hownot2.com/support
👉 10% off ROCKY TALKIE by clicking www.hownot2.com/rocky
Its a Myth!
01:34 Dynamic Rope
5:04 Semi Static Rope
06:53 Dyneema
09:45 Applying the science!
12:01 Conclusion

[@HowNOT2]

What we learned in this video was vital in making this episode a possibility! kzbin.info/www/bejne/fqrCiGRmrbKlea8 Check out our new store! hownot2.store/

[@matthewluttrell9413]

When you have a long rope and your doing short jerking pulls, you'll get lower peak values at the other end. If you statically load the long rope you'll get rid of that difference (for the most part). When you're jerking on the rope you'll see that the elasticity effects the force that the other end experiences. I loved this video and think it does a great job at demonstrating how mechanical advantage really works!

[@michaelm8624]

Yes, this would have been the better conclusion

[@diamondflaw]

Yup, under dynamic loading it’s acting more like a mechanical capacitor and smoothing the peaks.

[@davidpetrela383]

I was going to mention a similar thing. I would think that if you loaded a stretchy rope quasistatically, that is to say infinitely slowly such that dynamic loading is negligible or zero, the force would be the same as an infinitely non stretchy rope. But, some of the mechanical work would be consumed in stretching the stretchy rope.

[@kenbrown2808]

I agree. the bounce applies all the force to the first load cell, and the stretch absorbs the impulse before it is applied to the second load cell. it's why dynamic load ratings are WAY less than static load ratings.

[@colt45lmer]

just what i wanted to say if he wanted to get proper values he would have to statically load the stretchy rope to get. If he wanted to do this he should use a tree or whatever to do this vertically than he then apply his body weight statically to the system

[@patrickradcliffe3837]

I suggest revisiting this strictly using the winch. It seems you are just measuring shock loads with just pulling with manpower. You want a repeatable same force applied during each test.

[@elmeradams8781]

He used to do that exact test with his old bolt breaking machine. Then he switched to hydraulic.

[@davidseslar5798]

As a whitewater boater, I'd say this is pretty realistic for a small paddling group with a pin (except for the large number of pulleys available)

[@arlodillon1659]

The purpose of the test was explained at the beginning.

[@ianbelletti6241]

Exactly what I was going to say. Stretch affects shock loads, not constant loading. With constant loading you will reach a point where maximum stretch is reached and it will no longer absorb load changes. Also, when you lock a rope into a specific load on the pulling end, the stretch will cause the loading to equalize across the system and the output will be more accurate. What you're doing in these tests is showing how the stretch of the rope equates to the shocks on a car.

[@FallLineJP]

Or just use a progress capture on the pulling end to ensure the rope has time to stretch evenly all the way through.

[@error.418]

"Mechanical Advantage is a Myth" is definitely the better "get views" title, but I suppose what you're saying is "Mechanical Advantage is Oversimplified" and I love your investigation into real-world variables and results. Great work!

[@matthew568]

Right. You can still pull a boat or a truck with a simple pully system.

[@NoExceptions109]

Kinda of rubbed me the wrong way too. But "multiplying stretch with mechanical advantage" has nothing on "mechanical advantage is a myth" for clicks

[@NoExceptions109]

Stretch and friction

[@error.418]

@@NoExceptions109 Totally, it's fair play on KZbin, but nice to drum up some convo about it

[@jr.6199]

Ryan stated what he means, in the first 30 seconds. But if you're justifying lying, and calling it fair play for $$... Lookup how "Fair Game" was used by the admin of VP Dick Cheney (Haliburton profiteer), Scooter Libby outing a US spy for revenge, cause her husband Joe, called out Dick's lying about Iraq buying uranium to justify their invasion. Don't say lying is okay... for profit. Ryan knows that and is why he tests, for truth.

[@johnnylavoie]

Dynamic loading and peak force measurement are somewhat incompatible here. The springiness of the rope act as a damper, widening the peak force in a broader peak, while the force shockwaves may also be interfering which each other. You should do the same thing, but while keeping a constant force (maybe do 2kN and 4kN)

[@DieselRamcharger]

do it vertically. use weight. this is a bunk science on purpose with the intent to be dishonest. {click bait}

[@christopherforsyth5284]

@@DieselRamcharger definitely. That jerking pull is useless.

[@goskascience]

For the next test, I would suggest one of the follwing: - constant pulling force (such as a weight in free space) - progress capture device, again providing static load to the rope

[@guilhermetorresj]

If you apply a constant force, you don't get a video to post... lmao the results will be a lot different, and a lot closer to what intuition expects from those pulleys.

[@TheTomco11]

Also calibrated force gauges. It doesn't matter how much stretch is on the rope, in a 1:1 system the force at both ends is equal

[@punkazecke3881]

so a ratchet

[@BrandonRadtke]

To reduce the length of rope needed, you could extend the pulley block away from the anchor tree to just a couple of feet from the other end. This way, the back and forth between the pulleys is only a few feet of rope instead of 100+ feet. Less stretch and less elastic losses, and less rope length and weight.

[@therflash]

There's one more variable that you threw in by accident, it's the duration of the pull. A single straight line should have equal values, unless you're yanking on it quickly, without giving it the time for the forces to equalize. With the 13:1, the rope is so long that the pull takes a while to travel all the way to the end, and if the pulls are so sharp, you aren't using all the pulleys, you're just fighting against the inertia of the rope. Slow pull the next one?

[@pedroferreira3371]

But of you think about it, that is how you would normaly pull things by yourself on a wall or just in general.

[@therflash]

@@pedroferreira3371 True, it is still a variable though.

[@theaykay]

The straight pull with no multipliers also shows a big difference in the scale readings. This might be worth re-visiting after checking that both scales read the same and calibrating them if they don't.

[@olesvren2906]

These test are for a system he can take into the woods to test in real rock, so it is quite close to intended use. Though he should do a test with capture so that the system has time to equalize

[@iura0]

@@pedroferreira3371 except that normally you would be building up that tension in something (like a highline) with it's own stretch, and not just a solid piece such as the linescale.

[@stevebarratt888]

I think bouncing is having the effect of systematically increasing the input force more for the least dynamic cords. Also you could pick up a short length 1.5 to 2" thick wooden dowel to make a pulling 'handlebar'. I'm using a portable fingerboard similarly for tightening my longer slackline setups. I can get way more force without the finger pain.

[@SafetyBoater]

Maybe a rigging plate could work.

[@kevinmcdonough9097]

It definitely would, but I appreciate this because this jerky motion is how I tend to tighten things. Now I know the dynamic rope is eating up a ton of force.

[@DustyLambert]

I think we also need to consider static vs kinetic coefficients of friction: it takes more force to set something in motion than to keep it moving.

[@lleberghappy]

Efficiency of pulleys in a system is a compound problem seen in the video. For example, if 90% of the force is redirected from a pulley. If you haave a 5:1 system. The ropes will pull at: 100%, 90% 81%, 73%, 65%. These forces are added up 309%, so 40% of the systems total pulling force is lost in friction. So.. pulley efficiency in series is pretty important. A dynamic rope basically adds rope to the system upstream under tension, which means you'll have less opposing force and therefore less pulling power. I'm a engineer btw, and I do enjoy this backyard science, and trying to explain the results.

[@gallonofcactus]

This channel is amazing. Thanks for all your hard work and dedication. You are making the world a better place. This is like myth busters of rated gear!

[@brainstroem1]

i think it is not there yet :) but it could be, i the channel gets even more popular and budget increases :D

[@fishman241]

I'd love to see if you get closer to the exact ratios with slow/constant force pulls.

[@msilva3084]

I would imagine it takes time for the stretch to travel through the system. Giving it time to settle would probably have some effect. That would just be more for curiosity though as what he's trying to test is the efficacy of the setup the way it's used in the field. I'd also be curious to see the old 81:1 setup tested with he new line scale 3s. But that's a lot of work for what would probably be a predictable and not very exciting result.

[@xidarian]

I'd guess that it has a large effect on it. Imagine lifting a weight with a soft bungee cord. You could yank really hard and not lift the weight if you didn't get all the stretch pulled up and start lifting the weight but once you had the weight lifted and the bouncing stopped you'd be pulling exactly the weight of the bungee and attached weight. If he used that capstan winch to load the line slowly he'd get much more consistent results regardless of stretch but he'd have to pull more line through on ropes that stretch more. By yanking he's loading a bunch of kinetic energy into anling length of rope and then letting it dissipate between yanks. I understand that this is how he intended to use the system so I get why he did it this way but it's a bit disappointing that he didn't repeat the test with the winch.

[@theonly5001]

@WungusBill I would go for a small ratchet system in which he can gradually increase the force. He still can generate force spikes by forcing the Rope to one side and thus creating a massive force with very little input.

[@408murph]

Thanks for another great video. The physics 101 graphic at the beginning of video assumes frictionless (as you noted) but also a constant force being pulled with zero stretch line. Bouncy dynamic inputs are damped out by the stretchy rope, less by the semi-static and even less by the Dyneema. I'd be interested to see the Input v Output when using the portable winch set up as it's less bouncy. Appreciate the video and look forward to more content from the drop tower and testing of sailing components.

[@_joac]

Will be nice to see calibration of those linescales. Worked with load cells and is important to do the calibration process often if you are putting a lot o stress on them.

[@elmeradams8781]

Maybe, maybe not. You could check their 0 point by just turning it on. (I'm no engineer, but have used load cells on man lifts). They get calibrated upon install by setting a zero weight in bucket then max weight in bucket. The stuff in between is linear. They do look allot different then the line scales though so don't bank on my info.

[@georgebridgewater2573]

Love the vid Ryan, very informative. Hope there's one delving into multipliers too at some point. One thing I think would help: A bar chart at the end displaying all the different % results from each rope type and pulley setup. Every result on one graph would make it much easier to compare the results I think, rather than trying to keep them in my head.

[@glynroberts4802]

It would be interesting to repeat the tests with different total length of rope per configuration. Suspect doing 1to1 testing with very short length of rope incrementally increasing the length would give you a factor reduction in efficiency per meter length. Probably best to use mechanical clamps to remove variance in knot tying between each test.

[@stevebarratt888]

I was thinking this also mostly. same setup, different rope lengths. Clamps might be a bit overkill/impractical for their level of benefit.

[@samsokach582]

Awesome content my dude! You guys are constantly growing and learning and sharing it all with us. Much appreciated!

[@francescovaroli6893]

Haven’t seen that many comments appreciating the fact that this is helpful to understand why it’s soo hard to pull someone out of a crevasse. Yes there’s stretch and yes I am yarding on the rope just like he does here. So now thanks to him I’m considering separate static for glacier travel even with a dynamic for the climb.

[@matthewgough9533]

I have been so curious to see this series of tests done. Thanks for doing this!

[@jessebrighten6319]

You’d have better results with complex/compound systems. Put a 2:1 or even a 3:1 on the tail of a 5:1 and you can get some huge forces.

[@jakegotta4334]

👆that would cut out a lot of the internal system friction from ropes running back and forth next to each other - probably the biggest reason the 13:1 couldn’t out perform the 9:1

[@user-pz3oj7ql1f]

Boby: It's excatly same if you are dyslexic. me, dyslexic: no joke, that's static, and they are same, OH SHII

[@DxPru]

Now I wonna know what would happen with static load ...

[@FridayNightZombies]

Great video as always! I'd love to see this applied in more climbing-specific scenarios, mainly giving a boost to a second while belaying from above, as well as in crevasse rescue!

[@borosecrans]

After 8:1 the friction overcomes the efficiency of adding more parts. . .most of the time. Cranes will have more than 8 parts but use super efficient sheaves to do so.

[@ziggyx27]

As a sailor first and a climber second, I appreciate this video. Could you test some sailing-oriented line for us? I think most of it will have a dyneema core for strength and being static, but covers for this core vary widely depending on friction needed, pulleys being used, the handfeel, and workability of the rope. An interesting experiment would be to see how much extra force we get on a jibsheet for every extra wrap around a winch we take. More wraps takes more time to do and undo, but also lets you hold hundreds of pounds of force. Too few wraps means your winch will slip against the rope, and you won't be able to pull the sail in anymore.

[@skysocalbalen6075]

So a wrap will just increase the friction. It will highly depend on the type of and thickness of your line, the size and the material of the winch.

[@williamplovanic329]

Hey, have you ever thought about testing a petzle ASAP fall arrestor for use in rope soloing? Its been a topic of debate around the shop recently!

[@jakemonaghan1659]

When I pull my soft shackles to 10KN+ on my theoretical “15:1” system I found I can maximize the force of my “1” by putting on my harness and clipping to my belay loop. Like a saddle on a horse!! Helps especially when you have no homies to pull with you. I’d need an extra linescale if I want to do a similar test like this one though. Thanks for this info I’ve been waiting for some good MA testing!!!

[@levio1463]

Do those pulleys have ball bearings? You should look into sailboat hardware, they make some pretty trick stuff. The nicer stuff rides on ball bearings and has almost no friction, even under load.

[@bockjess]

Agreed. I've got 5 to 1 in pulleys plus a block on deck for 6 to 1 on my main sheet before the winch with sta-set line (dyneema core). Curious what it would show.

[@HowNOT2]

The 3” smc pulleys do have ball bearings. The others dont

[@levio1463]

@@HowNOT2 Harken makes some blocks that are rated for 44000 Lbs, can you test one of those? ; )

[@SonnyKnutson]

Fun fact, I am pretty sure ball bearings are a Swedish invention :D

[@petesmith1]

This is exactly why we use Dynamic rope for climbing, Semi static rope for where it is possible to fall, or a super Static rope where there is zero/no chance of falling. Lets see the difference of a FF1 and a FF2 on each type of rope. I think it's an eye opener!! Keep up the great work.

[@iura0]

Also, do the readings depend on which way you link up the linescales? Because this time you had the pulley system attached to the top of the "yanking" scale on one end, and to the bottom side of the "stationary" scale. Of course this wouldn't matter for a slow loading, but might be one of the confounding factors when yanking.

[@johnjohnstone8506]

There are two separate issue going on that are incorrectly getting linked together here. These are (1) how impulse loads are generated on a system and (2) the sum of all forces acting on a system (each rope end) incorrectly not equaling zero. These actions exist independently. Regarding the force generated from the duration of the dynamic loading. This deals with Newtons second law. In this case you placed kinetic energy into the rope by pulling on it. The amount of time the rope takes to absorb this kinetic energy determines the peak impulse force placed on the rope. The more stretch, the longer the time duration, the lower the force. The less stretch, the shorter the time duration, the higher peak force. This is why the static rope saw a larger impulse load. The second issue is the question of “will a rope that is fully constrained by one end ( with no other external forces restraining it) and loaded at the other end see different force values at its end”? Newtons 3rd law states no. The tensile load throughout the entire segment is equal at all points regardless of the load being dynamic or static. This exact rope problem is littered throughout intro level physics text books and commonly appears on midterm exams. With regard to the different measured values at each rope end it is not that Newtons 3rd law is incorrect, it is more likly that the way you set up and performed the test failed to collect representative data points, for all forces acting within the system. There is probably a couple reason why your load cell values are different Load cells only measure force in a predefined axis. Placing the rope horizontally causes the anchored load cell to rotate more when loaded the force vector of the rope was not always concentric with the anchored load cell axis. The way you held the input load cell when loading the system placed the force aplied ro the rope directly inline with the loading load cells axis . Because the load was more in line with this load cell a higher value was recorded more often. With regard to the end of the rope you were not continually measuring the total force placed on the rope. Form most of the time you were only measuring the vector component that was acting in line with the end load cell’s axis. This would not be an issue if you had a load cell with an extremely high sampling rate and where then downloading this data into a laptop running data acquisition software. As the duration of the peak impulses can be very very short. However it is highly probable your load cells were not sampling at a high enough rate to capture the peak data point at the exact instance it was 100% aligned with the end load cell. I would recommend re doing this experiment and see if you get more accurate data. Anchor the rope overhead and perform this test by pulling straight down with the force being transferred straight through both load cells axis. Crank the sampling rate to the highest value you can and try and apply the dynamic force at a rate that will not exceed the sampling rate of the load cell. It would be interesting to see the results I love you channel and it has lots of great information

[@deadfishporter]

Fantastic! Sometimes I use a 3:1 or a 5:1 when working. I thought I'd be getting much more pull than your results show. I will have to take this in to account when I'm tipping trees over in the future 🤔

[@michaelsukut6101]

I was thinking the same thing mabee I will hook it to the winch now.

[@infbajuk]

Any plans on testing multipliers in the future? also, an interesting experiment would be side pulling a tight line, maybe in conjunction to a pulley system

[@eatalotmorebeef]

Good work. I would be interested to see if a ratchet, like a micro traction, would allow you to eliminate some of the losses from the dynamic rope.

[@davidseslar5798]

Based on past whitewater rescue training (and haul systems during cave rescue training), I can't imagine a scenario where a ratchet ascender or prussik would increase pulling power (decrease resistance or "losses") other than to provide better grip on the strand used for pulling. They are useful to hold the applied force on the line to the load when only a single line reaches from the haul system to the load, to ratchet a short-pull haul system on a long haul line, and to rig some higher force haul systems with minimal gear available. A ratchet also requires suitable anchors, additional gear or gear robbed from elsewhere in the system, and often a person to tend it. Prussiks are preferable to mechanical devices since their 'failure to hold' mode (assuming sufficient strength margin to avoid prussik breakage) is to slip; this provides the potential to dissipate some energy via friction and still hang onto the load. Mechanical ascenders will cut through the rope - based on pull-to-failure tests I remember on 11mm rope from years ago, ~1000lbs/4kn for toothed cams (Jumar/Petzl/CMI, etc), ~2000lbs/8kn for smooth cam ascenders (Gibbs).

[@matevzmarincic9932]

I was thinking about this whenever choosing between 3to1 and 5to1, thanks for sharing! Do you still keep collecting the results in a spreadsheet or similar? I cannot find it. I like graphs! - would love to see a comparison of both linescales' graphs for some more insight. :)

[@biggs8729]

So I think you should definitely do this test again but this time use a steady increase of pull to see if the energy at the other end of the system levels out with time. My theory is that you’re jerking motion put only a momentary pull on one end of the rope which didn’t have time to work its way through to the other end. Still not going to be a perfect result but I’ll bet it’s a lot closer. Awesome video thank you

[@wflynnification]

If using a loaded dynamic rope with a progress capture (like when you're hauling), wouldn't that eliminate a lot of the effect of stretch you're seeing? Because the rope would already be pre-stretched?

[@lighterguy82]

I thought you only counted the Xtra lines, like when you were saying 9 to 1, it would be 8 Xtra lines to 1 cause you would always have the 1 to start with

[@allensandven0]

You need a repeatable force method of tension using the recommended pullley / block system for rope / line size and if dynamic or fail point is your criteria then you must. Load vertically with a uniform dead load to create dynamic or live load

[@FallLineJP]

I would be curious to see if you get better results on the dynamic / semi-static ropes with a progress capture on the pulling end. This should let the rope get evenly loaded through its whole length, rather than just measuring stretchy-waves of pull force. Great video as always!

[@petersmythe6462]

"assume all objects are rigid bodies and friction is negligible" -every physics professor trying to murder you.

[@rustytygart405]

I’ve been a subscriber for a few years thanks for the great content

[@iura0]

Can you zoom in on the force graph to see the shape of the peak? Is there just one data point at the sharp peak, or are there two or more points and the peak is kinda rounded by them?

[@elitaube3273]

Alway have loved the content man❤️keep it up

[@CasperEngineering]

One way to get repeatable static pulls with systems like this is to add two redirect pulleys, one at the anchor you are standing next to and one above the anchor you are standing next to. Run the rope up and back down then use your body weight by hanging or standing on the rope instead of relying on your arms.

[@niknik0815]

hey Ryan, thank you for that vid, that was super informative! Some other people already commented that shortening the rope probably plays a role. I know these tests take a lot of time, so maybe one of us has to do it when 2 linescales are around. The other day we broke some samples with a 5:1 behind a 5:1, pulling with 3 people and getting it up to 25kn in the process. I was surprised by that, because that means not much got lost in friction/stretch. We had big smc pulleys in the 5:1 and kept the static rope as short as possible (like 1m pulley length) then added a longer rope pulley with dyneema rope and smaller pulley wheels (you call them that?). Resetting took less than 5 min and we were able to break a lot of samples that afternoon. Your system of the electric winch is much better, but maybe some other people here can profit from that description :)

[@HowNOT2]

don't you only get 1/5th of your throw of the main 5:1 when you do a 5:1 multiplier? 3 people can equal 3kn of pull. The output if 100% efficient should be way more?

[@Aeronaughtica]

Great videos Ryan & Bobby, thank you!

[@MCKoolperson9]

Have you thought about using a progress capture pulley on the tail end? I'm not sure how much additional force youd be able to generate, but it would at least keep the rope between the pulleys tensioned between pulls so you don't have to pull that extra slack each time

[@OllyBoardy]

Do you think the 9:1 is greater than the 13:1 could be something to do with losing a percentage of the pull at each pulley meaning you want you most efficient pulleys first and the 13:1 has those small pulleys first?

[@nroose]

I know this is an old video, but just curious about what knots you used for the to attach the dyneema to the ends.

[@SunGodHorus]

How strong is the belay loop in the harness? I can't seem to find a direct answer for that question. I'd love to see you test some.

[@timkirkpatrick9155]

How often and by what method do you calibrate the scale? Why didn't you properly rig the rigging bar for the 9and 13 to 1 tests?

[@theralmackey8066]

As many below are saying, dynamic loading is wasting a bunch of the force/energy into friction and heat as the rope stretches. You will also find that putting the linescale currently on the tree on the opposite side (using an even number multiplier/pulling from the same side as the "load"), it will register even less, considerably less when using carabiners, since that first carabiner will eat a bunch of the force/energy as the highest friction point. As the force put into the rope goes up, the amount of that force required to even just start moving the rope around the carabiner will go up. . To prove out the actual advantage you should add the second linescale between the van/anchor and portable winch in that last break test.

[@joshuaknight96]

The relationship would be closer to 5:1 or 3:1 or whatever arrangement there is if it was equating total work on each side of the pulleys. Work being equivalent to force by distance (w=F•d) this also means we could add other variables onto this function such as the stretch of the rope. The stretch of the rope is ultimately the limiting factor when applied to multi pulley systems. That’s why cranes will utilize dyneema (amsteel) on a winch

[@jibblehardicardi3827]

Great vid and interesting results! Certainly using the capstan will give more control over loading, perhaps invest in some ground anchors to avoid "towing error". Regarding technique (and haters) I used to teach experimental technique as a Physics teacher and see no problem in what you do. You are looking for the big picture not the fine detail, I love it.

[@Mike-rx5uu]

You present a lot of awesome data, would be nice if there was a summary at the end. A chart of the numbers or something. Looking at a chart or graph, especially if you divide the actual numbers by the theoretical numbers, you can really see a sweet-spot right around 5:1 for efficiency. Unless you really need some extra capacity then 9:1 works.

[@rickkrause3198]

do you have any information about the power wench? Are rescue team is looking into them.

[@diegoconverslaverde9605]

Great video. Looking on the theory of mechanical systems, they consider no stretch on the rope, no friction on the pulleys and no weight on the rope, but also no weight on the pulleys plus a lot more variables, so as the video shows, the theory most of the times deviates from the reality. I usually use 4:1 system or 3:1, beyond that I think only adds weight and complexity to the system instead of giving advantage.

[@DanSutherland]

Would a vertical orientation also have an effect on results? My instinct thinks it definitely would, in part because of the jerking. I also wonder if angles on pulleys/gravity/leverage would have much effect.

[@jacobfurnish7450]

moment of silence for this man's spine

[@Mrwhomeyou]

Lol i like the set up in this video, there's no way I'm pulling at a constant and consistent way myself

[@ikesquirrel]

I worked with a tree company where the boss kept saying "this is a 13:1" or "this is a 5:1" or whatever, and they were not. The number of ropes is not how many times it is multiplied. Redirecting is not multiplying and the pulleys have to be in the right pattern, not just thrown together.

[@Puddyglum]

Most would never read the boring studies on the advantages of elevator music. Your channel allows us without a music degrees to see how effective the musical advantage really is.

[@ianproctor8069]

the stretch definitely makes a difference but only because the force is not constant, energy is being put into stretching the rope and that doesnt come out as force on the other side of the pully system. Once the rope is stretched out the force would look quite similar to the static rope if you had a constant load: perhaps hanging a weight on the drop tower with the reduction pully system?

[@ivanlawrence2]

If stretch is the issue, wouldn’t compound MA have better results since there would be less material (shorter legs of cord)?

[@harryfine3655]

Excelente tema. Es importante derribar todos esos mitos comerciales. Felicidades desde la Ciudad de México a todo el equipo del canal.

[@nohands3255]

I tend to use a 4:1 with a 4:1 jag on the end. The key is constant pull with a progress capture. Next time I do it I'll put a meter on it and see how it compares.

[@heathvanaken7954]

This was very helpful, just like all of your videos. Thank you!

[@alexnone5720]

Some sort of progress catch on the final pull strand should help you build more overall tension/force in the system and give a more constant load on either the line or whatever you are trying to break. To get a better idea of how friction effects the amount of force this can be repeated by hanging it from the drop tower and adding the tractor weights to the pull side. Using a constant weight and changing up the pulleys from the 3" SMC pulleys, to the carabiner rollers to regular carabiners as well as testing a dynamic rope vs dynema to see how stretch and creep can affect the total force.

[@tonkmonster]

Best video! Love the pulley science! More please

[@johnliungman1333]

So, Ryan, I think you are totally right. But let me give my take on why. What you refer to as ”friction” is actually two things: heat generated between the surface of the rope and the biner, AND heat generated inside the rope due to deformation (as the outer curve of the rope is stretched and the inner compressed). The latter is why big pulleys are better than small ones, even when both have good bearings. Now, when you use a dynamic rope, the deformation is much greater (thats the whole point of a dynamic rope: absorbing energy by heating the fibers). While a static rope is much less deformed as it wraps around a biner or pulley. This is my hypothesis, anyone deeper into physics please correct me if I am wrong! Thanks for a great vid! ☀️ (Having read other comments about the need of a static force, I agree. But even having that in place, I think we would get less advantage with stretchy rope.)

[@guilhermetorresj]

The problem lies in those short bursts when you pull. If you apply a constant force, the stretch doesn't have the same effect. The stretchiness of the rope makes it spread the changes in applied tension over time. The best analogy I can come up with is with capacitors being used as a low pass filter in a circuit by smoothing out high frequency changes in voltage.

[@flyn2274]

it would be awesome to see a repeat of this with the same length of rope in used in all the tests.

[@halcyon510]

It's be interested to see how this goes with dynamic rope and a keeper, like say a petzl micro traxion. The stretch between the keeper and the output should be locked in after the first few tugs and I suspect you'll get better efficiency. This would be more analogous to a haul/crevasse rescue scenario.

[@lebulba]

Might be time for another buying guide now that you got the new site goin. Also new climbers like me would love your opinion on good gear to buy right now. Keep up the good work

[@devinholland2189]

Ryan consider a sailboat winch as a mechanical advantage. Theres an enormous one used for arborist work.

[@Only_extreme_]

Can you do a test on ice screws and while you are at it snow anchors where you literally only snow. I’ve seen it before and was wondering if it was super good enough.

[@lordofnothing.]

so the sweetspot for max power is somewhere between 13:1 and 5:1. would love to see 6,7,8 and 11 to 1 tested.

[@davidbrewer7937]

You have to add in the divided values of stretch as well as the effect of gravity pulling down on the rope... This is why by the time you got to 13:1 you had slack ropes dangling...

[@toastyherring]

It would be interesting if you where to apply a progress capture and compare using one early/(before the pulleys) in the system or last.

[@RomanFedorov]

2:00 According to my knowledge you have 3 wires and 2 blocks, so force should be 200% but you expect 300%. And you got 1.13 and 2.26 which is exactly 200%.

[@TheRetnuh1]

Dunno how this isn't top comment, he kept counting redirects as mechanical advantage. Every test was calculated wrong.

[@jasonbowers4388]

Have you tried an aluminum framed come a long for input power. I pull trees over with a 5 to 1 and a 1 ton come a long. Great videos. Always good info. Thanks

[@syborg64]

Theoretically, stretch shouldn't affect force, only work. Though it's understandable in an imperfect context, it would generate a measurable force difference. I think that difference would be a result of rope mass: as it resists acceleration it smooths out the force curve, with an equivalent integral but lower peak (which isn't the case for stretch, as a perfect elastic would have the same force either side of it at any point in time). I would suspect that on a continuous pull, the reduction from stretch would be less significant as the peak input force is held. You did provide graphs for the output LS3, but I'd be interested in comparing the graphs for input and output sides of the pulleys.

[@marceld6061]

Did you try pulling one scale directly hooked up to the other to see if you get any difference? I wonder if there is a 'calibration issue' between the two devices. Or run tests swapping scales end for end with the same line/pulley set-up to see if your readings are accurate. Also, like others have stated, either repeat the test vertically with a static hanging weight or use the capstan winch for a steady pull for each test.

[@mattfleming86]

My degree was chemistry, and I was one of those guys who took extra physics classes as "hour fillers" and for fun. It was a oet oeeve of mine that calculations always had so many idealized exclusions. Friction is rarely negligible, and unless you are flying REALLY high or going REALLY slow air resistance is always a factor. It's nice to be able to make approximations but the real world is always so much more complicated. Unless you are PTFE in space.

[@joshstagg148]

What's your progress capture for hollow braid UHMWPE ?

[@larryseibold4287]

Friction is a thing, and is largely an independent variable of dynamic loading. Using dynamic forces with a dynamic rope, is going to give dynamic results, where the system (rope) acts like a capacitor/inductor to smooth/dampen the input to the load. moving to a slow load increase (to say 2kN) will show the quality/effectives of the pully selection. If large great pullies are used, the rope selection will not matter. Using sharp bends in carabiners etc., will turn into a brake system at about 5:1 or less. Using static rope allows the "bite" with the people (walk distance to rope break rebound anchor) to be far shorter for the same pull length at the load (the working end of the 5:1 is only moving 1 ft for 5 ft of travel. 2:1 stretch drops that to 1 ft for 10 ft of travel or more due to cumulative loses). 2:1 and 3:1 are my favorite ratios if you have enough people to pull. It is not easy, but it goes much faster, with minimal gear. How do you hold that 5mm Dyneema cord (think four pairs of hands vs 11mm static line).

[@adamcook5192]

I was watching the dog thinking dang it’s super calm.

[@rickhaller3328]

This was very informative. Thank you for making this video.

[@dgardil]

The exact problem Aron Ralston (127 hours) faced trying to pull the rock (as he explain in his book). Stretch and friction played a big role!

[@340wbymag]

When someone tells you they are using a 3:1 pulley system (or any other numbers), they are describing the configuration of the system, and NOT the actual ratio of mechanical advantage. The efficiency of the system is affected by many variables.

[@jonflannery8984]

Serious question: hope someone can answer seriously, what if you were to capture the stretch in the rope not allowing it back into the system. As you jerk the input side you are taking up the stretch but then you let it return into the system. Is it still such high loses when you’re slow pulling with say a winch or a van where that stretch is continually being taken up…would you not eventually reach the point where the stretch is stretched and then acts like a semi static or static rope where your input more closely matches the theory or would there alway be loses to keep the stretch stretched. Sorry wish I knew more technical terms but hopefully it’s clear enough.

[@kylehart8829]

This actually causes most of the loss due to stretching. If you progressively load a stretchy rope and have a lot of patience you can get it to transfer very large forces, but if you jerk on it that'll never happen.

[@joshbrown3572]

I'd like to see you try a compound pulley set up? I bet it would dramatically increase your force.

[@anotherriddle]

Great testing! I'm not stoked about the title though and hope that's worth it for the views. I'll never understand it, but i'll live with it. Cheers!

[@johne1097]

The size of the pulley vs the diameter of the rope has an effect on the friction. That is another reason that the 5 mm Dyneema did better than the 10 and 11 mm lines

[@eduarddvorecky3731]

I wonder if you have let's say 5 guys pulling by one rope vs 5 guys pulling each by one rope will there be a diference? It's because when i pull one rope with other people i feel like my strengh isn't used as efficient as when i pull alone

[@snigwithasword1284]

Wow, I threw together a scabby pair of pulleys to hoist a kayak.. (paracord and 5:1 bushing not bearing pulleys.) I wonder if I was already past the point of diminishing returns!

[@bemk]

Maybe a cool test to do is to pull a line between 2 line scales twice, but switch the line scales around between the two tests, to see how much of a difference there is from line scale to line scale. I'm guessing it's small, but it'd give an indication of the quality of the scales as well as the error margins on these kinds of tests.

[@RileyOutside]

"It's exactly the same if you're dyslexic" -- me not realizing they were different until he said that...

[@Archer-bc6cv]

I have a suggestion for the data that I think would have been cool for this test and may be useful for future tests. Lets call the linescale you were holding A and the linescale at the end of the pulleys B. I think it would be really cool to make a graph of a moving point with the data from A on the x axis and the data of B on the y axis. I imagine it would make a little oval shape as you play through the time, but I don't know exactly and I think it would be really cool to see. Like at the start you yank on the line and A starts to go up. so the point moves right. but then a little later as the rope stretches B starts to increase so the point also moves up. But now the yank is over so A starts going down (so the point moves left) and finally B starts going down a little later so the point moves down back to where it started. I think for tests like this one it would give you some really cool visuals and you could even try to compare the oval shapes at the end