Understanding Trailer Dynamics like Sway is a really important. If you look for it, there are many articles and some great YouTube videos discussing and demonstrating the many factors and issues. There is some really good info out there, and some (we’ll call it) sketchy info too.
I have 2 favorite (currently) videos on the subject of trailer towing dynamics. The first shows trailers in a cross wind that we referenced in this previous article Dangling By A Single Trailer Safety Chain. The second is a video posted by L2SFBC – Robert Pepper showing some great insights about trailer sway.
Excellent Trailer Sway Video
Everyone who tows should know something about “Trailer Sway” (I prefer the “Trailer Stability” description). In the video below, Robert Pepper gives a great overview with easy to see demonstrations. So today, with permission, we’re looking at the video in detail, then injecting some perspective.
Seeing trailer sway in real life is always a little scary. You just never know if it will increase? Or if the driver will get it under control. In this video Robert shares some experimentation as well as real footage of sway gone terribly wrong.
One of the things I really like about Robert’s videos is the analytical approach. For instance, at the beginning he talks about how the models are and are not representative of actually towing a trailer. From the speed of the treadmill to the mass of the bits he is using.
Applying Engineering Perspective
Through my engineering career I have always loved learning. In more recent years, with the rise of YouTube and other sharing platforms, we get the privilege of seeing the work of many really smart people. On the other hand, we also see a lot of (let’s call it) eye-crossing physics going off into the weeds. Who should you believe? There’s good, and there’s misleading, so be careful.
The topic of “Trailer Sway” attracts quite a few videos. Many are just promoting anti-sway gadgets without giving sound validation. Skip those. Several also use a treadmill and tethered car (like in the video above), but fail to bring real understanding to the audience. From an Engineering perspective, the treadmill model is great when the results acknowledge the limitations.
In the video above, Robert demonstrates a lot of things, then relates them to real life. He shows things that work, but doesn’t go out of bounds trying to explain it. Most important, he is not a sales pitch for products that claim a fix for sway issues. Kudos for that.
So, the video is here, with permission. In this article we’ll refer to several bits within it, then add some perspective — and, ask some questions.
Dissecting The Video On Trailer Sway
First, I want to say I really appreciate the video Robert has produced. It’s a lot of work, I’m sure, and he did a great job demonstrating and explaining. The items below are NOT a criticism at all. It’s far easier to critique than to create, so the things below reflect a 20/20 hindsight coming from watching and thinking about what he created.
Also, as we learn, we know how to ask better questions so we can learn more. That’s really what I’m getting to with this article. Watch the video first, then read below and look back with the notes.
(Times noted below refer to the minutes:seconds of the above video play. YouTube video clips are not super exact, so I apologize if the notations are not perfect.)
Near the beginning of the video, 2:37 a weight is moved from the center of the trailer to the back. Note that the placement of the weight did NOT instantly start instability. It took an action, or external disturbance to set it off. We see this same thing in many places through the video.
That may seem obvious, but in life, people often tow trailers that are on the verge of stability and think it’s fine. Well, it is fine, until it’s not. We don’t know exactly what we’ll encounter. While it is probably not a giant finger touching the trailer, disturbances do happen.
Means Of Disturbance
The means of disturbance in these demos is a physical displacement of the trailer to the side. In reality you rarely get a definitive physical push as he does there. (Maybe in a police chase “pit maneuver”.)
We do, on the other hand, get all sorts of “disturbances” that are not as obvious. The wind for one, and turbulence from another vehicle for another. Those are the obvious, but there are more, like steering input from the tow vehicle, a dip or bump in the road, a corner or off-camber turn, etc..
I would love to see this same demo using a bump to excite the trailer. Just putting something on one side of the treadmill for the car and trailer to roll over. Or, actual tow vehicle steering input, rather than physically displacing the trailer. How about a fan or blower to create a side wind? It is my experience that real disturbances which cause horrendous accidents are much smaller, almost trivial in comparison to a hard physical displacement of a trailer to the side.
Emphasis By Accident
As an emphasis of the demonstration, he did not point out that when the weight fell off at 2:47, (which I’m sure was unintended). Worth noting, the trailer went back to stable even though it was quite out of control just prior to the weight falling. That tid-bit emphasizes what he was saying, quite by accident.
In the video at 2:08 or so — Thank you for explaining the “Scale Speed”. It is not actual speed, and I’m really glad he points that out.
For all in the USA, a speed of 90 km/h is about 56 miles per hour. 120 km/h is almost 75 mph. I’m glad he has 2 speeds 5:40, but I’m left yearning for more info. I would love to see the effects of lower speed also. I often claim that conditions for instability on the highway, will not necessarily manifest as instability when driving in town.
Almost all trips with a trailer start with slower speeds. Everything feels fine, so we proceed. Even when conditions for trailer sway exist, we might not get it without extra factors like speed or a particular kind of disturbance. See the video at 2:40 when he places the weight on the back (creating conditions for instability), but the system remains stable until the disturbance. Would that also happen with the same conditions and same disturbance at slow speed? 30 km/h? (19 mph) or 60 km/h? (38 mph) The discussion at 5:40 emphasizes this.
In real life, people travel long distances without trouble, then suddenly, there is a horrible accident. Other people don’t ever have an accident, but talk about having trailer sway they have to “fight” all along the drive. These are important as we think about knowledge we get from experimentation and bring it to our own trailers.
Size of the Disturbance
With this same demonstration, what if the disturbance was really small? What if he gave progressively larger disturbances to see what it will recover from and what it will not? I’ll suggest some conditions that will go catastrophic with a large disturbance, won’t with a small disturbance. I’d love to see that demonstrated to see if it’s true or not.
And, going back to Speed, how much of a disturbance does it take to cause catastrophic trailer sway at higher or lower speed? While it’s an interesting question, maybe it’s not all relevant in light of the Kinetic limitations of the experiment.
There is one limitation that Mr. Pepper did not explain which makes a demo on the treadmill quite different than life. The kinetic energy (mass energy in motion) for the car and trailer on the treadmill is near zero. They aren’t really moving, the belt is moving. In reality, when you’re driving, your car and trailer have a lot of kinetic energy, which complicates dynamic interaction.
These tests on the treadmill really model the vehicle interactions with the surface (road), not the dynamics with vehicle kinetic energy, wind drag, etc.. Of course, kinetics do come into play when instability goes haywire even on the treadmill, but that is primarily side to side without the fore-aft component.
Please don’t misunderstand what I’m saying. That does NOT invalidate the testing, it’s just another limitation of the experiments. Something to keep in mind as we transpose the results from model to real life.
Note also that the weight of the trailer 1900 kg is about twice that of the car 1000 kg. That’s a big difference in Kinetic energy. That’s kind of like pulling an 8,000 lb trailer with a mid size SUV — something the SUV is probably not rated to do. Again, just keep it in mind. I believe the demonstration is valid, but these limitations are important. Also see 4:43 + where he speaks to weight differences. The delta in weight is a negative with respect to “control”.
Ball Weight / Tongue Weight
A discussion about too much ball weight begins in the video at about 2:55. I am really glad he covers this. Here are a few more points to think about.
- In the demo, the amount of weight right at the hitch is extreme. What percentage of the trailer weight is on the ball?
- What is the load change for the front versus the back wheels of the tow car? (More on this below.) Also, how does it act with just one weight added to the tongue?
- I’m sure it’s a limitation of the video, but they did not let the condition persist. See 3:21. There is a confounding effect with the tether at the front of the car. I don’t know if that is helping the situation not get too out of control, or if it is keeping it cyclic? I’d love to see this experiment with a longer tether.
- What about variations in other factors? A small added weight to the front of the tow vehicle to keep the front wheels solidly down. A shorter (and longer) hitch extension. Though I love the demonstration, the simple statement “too much tongue weight” leaves questions. What is “too much”? Robert is right about “where does it end”, and about considering other things. I’d love to explore more. Certainly large semi trucks are stable with extremely high tongue weights, so tongue weight by itself is not a complete story.
— I admit, the above questions may be off into the weeds. For most folks it’s enough to know that increasing tongue weight is not the end-all answer.
While this Custom Trailer design we did has huge tongue weight (10,000 lbs when loaded), the key to stability is the tow vehicle.
- In the video tongue loading scenario, what happens with a weight distribution hitch?
Location of Weight Front and Back
In the video at 3:32 they begin a demonstration of weight location on the trailer. Awesome. But again, it leaves me yearning to inject some physics. (Actually, in a more recent video, Robert gets into this.)
Back to the video at 3:32 . . . What is the tongue weight compared to other conditions? What are the axle loads for the tow vehicle? Remember, in these experiments, the tow car (1000 kg = 2200 lbs) is about half the weight of the trailer (1900 kg = 4200 lbs).
Going on to 4:43, they change the weight difference, and we see a dramatic change. Something more equal in weight (trailer vs. tow vehicle) is more normal on the road. That said, the real emphasis for distribution of weight comes at 5:07 where the trailer is drastically heavier than the tow vehicle (almost 3x), but proper weight balance makes that work. This is super interesting.
Also, even when the large weight is not centered on the trailer (side to side) 5:18 to 5:26 it remains stable. I’d love to explore this more, including the measurement of load on each trailer wheel, since I have always argued that it’s best to have things central. Robert also thinks so, because he was trying to fix the shifting weight, but in this experiment, it didn’t seem to make much difference.
(Beyond stability, there are other reasons to keep the weight central.)
The video speaks to hills just briefly from 6:01 to 6:36. Excellent demo, which is spot on. However, it leaves two pretty big questions.
- How much worse is stability going down hill?
- What happens when braking? (Obviously, there are interactions with brake balance, yet in many ways, braking is much like going down hill. Also, sudden braking can be the disturbance to upset the flow.)
I’m not sure you can demonstrate either of these very well on a conveyor belt because of the kinetic energy limitations, but they are the questions that pop into my mind. How we could set up some experiments to mimic that?
Effects With Trailer Weight Placement
At about 6:40, the video changes venues to a static model. I think Robert does a good job of bringing it to a non-technical understanding. However, I want to point out that the demonstration is a gross exaggeration for illustration. It’s an important point, and I love that he makes it very visual. So, if your towing rig is not this exaggerated, that’s good, but realize some of the effect is still there.
At 8:02 Robert says “… that is why …”. This one of the few places where I take exception. Yes, it is a contributor, but it is not “why”. What he is showing can be a significant contributor in control of trailer sway, but traction with the tow vehicle is not “why” trailer sway happens. I think he’s saying that in a round about way, but it is not how it came out in the video.
Remember, trailer sway is a dynamic condition. It is easy to push the trailer sideways on the conveyor belt, but if the belt is not moving, the tires won’t want to move side to side. The moving belt allows it to move much easier. For the static model, the jostling of the hitch side to side is not the same as if the tires were spinning. Again, consider it, but please look beyond this limitation of the illustration.
Finally, he did not mention it, but some discussion (and demonstration) about weight distribution hitches would be very useful here.
Weight In The Center
The points in the video about trailer loading and how that affects the tow vehicle are important. There are a couple points to clarify.
As a minor correction, we don’t want the trailer loading right over the axles 8:05, we want it slightly forward so there is appropriate tongue weight. It’s hard to remember to cover all the nuances when you’re in the middle of recording a video, so we’ll give him some grace. I’m not trying to pick on the video, I just want to clarify. (And, he does cover it in the summary.)
So, when he says “… fairly equal weight distribution …” 8:10, referring to the tow vehicle, that’s not really what we are going to get. We just don’t want either of the more extreme conditions he illustrated. There will always be some change in front to back weight loading when towing versus not towing. The point is to minimize anything even remotely like the extremes.
Trailer Sway Is NOT just Trailer Sway
Please note that there are 2 things at play: 1) Conditions that contribute to, or cause trailer sway; and 2) Conditions that keep control of the trailer as it tries to sway.
In the video they bump the trailer to the side, that is NOT trailer sway, and it is NOT the “cause” I’m talking about. A disturbance may excite it, but if the system is stable it settles. 2:34 On the other hand, if the system is not stable, it goes out of control. 2:42
The “Cause” I’m talking about is the sum of conditions that will not allow the system to return to stable. The “Control” I’m talking about is the sum of conditions that bring things back to stable, even though the “Cause” conditions and disturbance exist. That is very well illustrated at 4:18 to 4:43 because the trailer wants to be unstable, yet the control conditions now over-power the instability. The same is true with the hill demo at 6:01.
We want to minimize conditions that cause trailer sway, AND, emphasize conditions of control. Tow vehicle length, weight, axle weight distribution, brakes, speed, etc., are all things that help control sway. Trailer imbalance, overloading (front, back, or total), trailer load distribution, size, overhang, etc., are cause factors.
While talking about Cause and Control as separate is a nice way of thinking about conditions of the system, really, they are all part of the whole. And, there is plenty of room for confusion on what is a cause, versus what is a negative for control. The trailer and tow vehicle are a system. Making it stable and safe is the goal.
Cause vs. Control
To emphasize the difference between cause and control, we need look no farther than our article and video on Why Do Some Trailers Tow Stable? That video clearly shows a trailer that wants to be unstable — it wants to sway — but the controlling conditions won’t allow sway to take over. Vehicle size and other conditions keep it in check — even though the causes are actively trying to sway.
The reason I bring this up? Just because it doesn’t go out of control with trailer sway, does not mean it’s not a problem. Correct the cause side of the equation, then control required is much easier.
I’ll bring up another point — that sits as a pet-peeve for me. How many times have I watched a YouTube video where an RV’er says “Get this new Weight Distribution hitch and it will correct all your trailer sway”? That is totally naive. If your axle weight balance is out of bounds, then usually, bringing it back in line will improve “control”. Yes, sometimes that’s enough to make the control more than the cause, but if your trailer wants to sway, then you need to look at reducing the “cause” factors ALSO. Most RV YouTubers don’t think this through all the way. Sorry for the soap box.
6 Point Summary
I love the summary Robert gives. Starting at 8:18 the summary points are stellar. Watch it twice so you don’t miss anything.
Robert, you did a great job with this video. THANK YOU !!
There are more videos at L2SFBC on the topic too. If you want to learn more, try some of those.
Important Points – Understatements In The Video
Two really important points in the video come almost in passing. I want to emphasize.
5:56 “… the faster you go, the more likely you’re going to have problems with sway.” I want everyone to hear and comprehend all the ramifications of this statement. (Yes, it’s in the summary too, but I loved the way he passively states it at 5:56.)
6:30 “Conversely, going down hill you’re more likely to get sway.” A point most people don’t even consider.
I Want More !!
OK, that’s sounds very selfish. Yes, I really want to learn more. These experiments are wonderful, but they bring even more questions for me. As noted with the questions in the article above, but also the actual physics. The “WHY” for conditions that cause instability like trailer sway. We see the demo, and we can’t argue with the result, but what are the real “Why’s”? How are they different in life compared to the conveyor belt?
I have explored some of the physics and engineering answers to these questions, and I’d like to post about it, but I think the equations would be very boring. Plus it will take a lot of effort, so we’ll see about boiling it down to something manageable for an article or video.
Or maybe I’ll just go onto trying to answer my long “Wish List” of questions? That might be more interesting anyway. We’ll see.
In the meantime, these videos by L2SFBC – Robert Pepper give us some great things to think about.