Most treadmills have a guard or plastic covering on the bottom that stops the tread from pulling something underneath it. From pictures, the peloton treadmill doesnt have this... so I would guess if anything gets between the back of the device and the floor, it'll pull it under the machine.
That's what it looks like to me, too. That's a terrible design. It's just the right height to suck a small pet under the machine and crush the animal.[1] Or a foot. Looks like Pelotron wanted a nice clean look, so they left the guards off. That looked nice in the Jetsons series in 1962.[1] But not in real life.
In industry, this is called an unguarded pinch point. Classic cause of injuries and amputations. It's an OSHA violation in an industrial operation.
Every other treadmill I've ever seen had substantial guarding around the pinch points.
OSHA: "1910.211(d)(44): 'Pinch point' means any point other than the point of operation at which it is possible for a part of the body to be caught between the moving parts of a press or auxiliary equipment, or between moving and stationary parts of a press or auxiliary equipment or between the material and moving part or parts of the press or auxiliary equipment."
I think this is clearly negligence, I immediately saw the danger at a glance of the image in the article but had to look twice to confirm that they left off the guard for the bottom. They should be sued and the designer should find other work preferably not involving anything that could pose a danger to human life, maybe floral arrangement as long as they don’t find purple, yellow and red an appealing combination.
The large gap with an exposed high speed belt is an obvious hazard and is completely unnecessary. There’s an argument for not letting your children play with treadmills unsupervised but there’s no defence for a design that would allow a child to be sucked into the machine while it’s in use if they happen to wander in to the room and fall behind it. It’s an easy to imagine scenario since they might try to approach their parent from behind and be struck down by the moving edge of the belt and then pulled under.
Well some parents make mistakes or unpredictable situations happen. That is hardly unusual in a house with pets or kids. It is expected. Of course you should design with that in mind. A bunch of warnings that no one reads is no excuse.
Thats very harsh to the designer. I know many and nobody would design machine to kill. Its obvious the management decided to cut few hundred bucks to get higher bonuses.
This thing needs to be banned first, company investigate second. Bet there are email trails from angry engineers telling exactly what most of us say here. The excuse that only few people get killed so its all kumbayah is the most despicable thing a CEO can say. He needs to resign immediately.
The picture you link to is of a pair of Truform Runner treadmills[1]. I have no idea of how safe they are, but they are notable for not having motors which means that they wouldn’t pull a child under them on their own. Could they injure a child? I don’t know.
Perhaps someone that has used one and can comment on it. I’ve been thinking of replacing my current treadmill with one of these. I don’t have small children or pets, but I’d like to get a safe treadmill if I’m going to buy new one.
There are many models of slatted treadmills and in my shopping I haven't seen one with a guard
It could be a coincidence, a lot of treadmills period don't have guards, but really I think the answer here is maybe making sure kids can't activate it
I was expecting the child standing on it to be the one hurt since at full tilt those could easily catapult a child with deadly force..
> the answer here is maybe making sure kids can't activate it
That doesn't help if a kid or pet are caught in there while the adult is on the treadmill, especially at higher speeds or when reaction time is slowed due to wearing headphones.
The answer is to install protection. Having slats instead of the classic band doesn't make it impossible to protect that pinch point, rather just makes the machine bulkier (longer) and less aesthetically pleasing. There are plenty of mechanisms and sensors that can detect something being caught in there and immediately stop operation but they cost more.
See the video above, with the small child being slowly pulled under the treadmill. The machine not being able to detect *it lifted off the ground repeatedly while operating" in order to come to a full stop immediately is the kind of thing that happens when half-assed engineers meet half-assed managers, thinking they can make a "full-assed" product.
Yeah, no... there is no "one answer" unless you want to say "don't be a negligent parent and let your kids near running power equipment"
Why are you acting like this is a binary choice instead of a case where multiple mitigations can try and help make a bad situation slightly less worse?
There are multiple stories of children dying on treadmills without ever leaving the top surface of them. Mike Tyson's daughter died after getting caught in a rope hanging near the machine.
-
At the end of the day with you on it or not, if your kids are near the running machine, it's already a dangerous situation and the most important mitigation called "parental supervision" has failed. No other mitigation will change that.
I wonder how long until I get the requisite "have you seen how easily kids escape supervision!" I sure have, and that's why having a system only an adult can activate goes a long way.
I'm a fully grown man with one and I'm certain if I were to put just a single foot on it at full tilt I'd violently injure myself. I can't even imagine what it'd do to a child. A guard wouldn't change that
Who said anything about this being a binary choice or about negligent parents? I mentioned the adult on the treadmill to highlight that the answer is not "making sure kids can't activate it". The device is expected to be used with unsupervised children or pets around it so a simple child lock is not nearly enough. The back of the treadmill is always unsupervised during use so you either make it safe, or you clearly instruct the owners to only operate them in strict isolation from children or pets.
> that's why having a system only an adult can activate goes a long way.
And I'll just say it again so it's very clear: it doesn't go nearly far enough.
But I want to see a guard behind and under the treadmill so it can't grip and drag anything under it. I want to see sensors that detect when something was grabbed by the belt past a certain point, detect the (sudden) change in drag, detect the inclination of the treadmill, detect and only operate when someone is on the treadmill so that putting "just a single foot on it at full tilt" is simply not possible. When the failsafe is triggered it should cause an immediate but relatively gentle stop and maybe even slowly reverse the direction of the belt for a short time to release whatever was caught.
My cheapo gardening tools make sure they only operate when both my hands are tightly gripping the handles, they detect when something is wrapped around the spinning parts and stop to prevent damage, some even detect if you hold them at the wrong angle and cut power.
The are plenty of engineering solutions to to cover 99.9% of accidental and maybe even intentional misuses. But they cost $10 more and don't look that nice. So killing or injuring a child as a cost of doing business is preferred.
You literally start your comment by acting like what I described is useless then follow with
"The answer is to install protection."
Sure sounds like you're saying "the answer" is install protection.
When there is no "the answer" other than "kids and abrasive belts moving at 6 miles an hour don't mix"
-
Thats is why I said this is about negligent parents because it is! That child on the top of the treadmill could have easily grasped the nice tactile looking knob on the side of the machine.
They would have been instantly catapulted off the machine at full speed and no guard would have helped.
Like a Tread has a safety key clipped to you, your "reaction time" would literally be turning your body.
But even that is a red herring: if you're letting your kids near the machine while you're using it and unable to react to them even getting near you, let alone on it, again you're being negligent and no mitigation will change that.
That's not saying there should be no safety features before someone starts misconstruing this comment to say that...but the machine by it's nature exposes its most dangerous surface. It's like if your cheap gardening tool required holding it by the blade to use it, the treadmill belt has to be exposed in large swaths for the thing to work and presents an immediate danger to children.
Like seriously, think of a treadmill as a chainsaw you run on because it's not that far off in terms of power or how many people get hurt by them every single year.
Again treadmills injured over 20 thousand people in one year. 20 thiusand ER visits... if it was as easy to make a treadmill "99.9% safe" as you're handwaving it away to be with grandiose psuedo-political pandering... I'm sure someone would have thought of it.
I literally started just by saying that I have no idea how you came up with the binary choice claim (still don't). Then I clearly highlighted that the answer isn't simply to implement a child lock like you suggested. That's a very small part of the answer. Like a positive attitude is just a very small part of the answer for curing disease.
> When there is no "the answer" other than "kids and abrasive belts moving at 6 miles an hour don't mix"
Kids don't mix with a lot of things that still fill our houses because we designed them to be as safe as they reasonably can. Even your lowly power outlet isn't just 2 hanging wires with a warning post-it because someone smart enough thought it'd be a bad idea to expect humans to not make mistakes. The live wire buried in the wall, holes covered with plastic, fuses or GFCI/RCBO, etc. to keep you safe. Most of your belongings are designed to enforce some safety measures. But a treadmill operating by itself and dancing around the floor while sucking in a child is OK because the parent should have been there?? That's a bad product and an indefensible position.
Engineering for consumer devices has to be done with safety in mind and compensating for the mistakes that the customers will inevitably make. I'll leave this here:
> "It is believed that at least one incident occurred while a parent was running on the treadmill, suggesting that the hazard cannot be avoided simply by locking the device when not in use," the CPSC said.
So with your answer being categorically classified as "not the answer", the big part of the real answer is to install many protections that Peloton and others skimped on. Some of them I listed and would have prevented anyone from being thrown off, or sucked under of the treadmill unless they were going out of their way to achieve this.
> if it was as easy to make a treadmill "99.9% safe" as you're handwaving it away
It is easy to make them safer than they are now. Manufacturers just choose not to because it's cheaper that way, because regulation doesn't force their hand, and because the occasional injury or death is the cost of doing business. And I say this as someone who saw every step of this, from designing and building (up to and including medical devices to give you an idea of my upper limit for safety and caution), to setting the company strategy, in a time when there was no "move fast and break things, we'll just send an OTA to fix it".
You had multiple opportunities to understand the point I'm making very clearly but chose to focus on covering your weak position. That's too bad.
You started this comment with another bit of nonsense so I'm going to stop reading and save my breath.
"It's a very small piece" when it literally would have prevented the exact incident linked in this thread that ended up with a child already abrading their face on the top surface before ending up under it.
You can't understand the basic details of the situation so your comments are not worth my time
Nothing is worth your time when you put ego first, and focus on saving face and being boorish after being called out on a bunch of half baked ideas. I even quoted something that I expected would grease the wheels of reason:
> "It is believed that at least one incident occurred while a parent was running on the treadmill, suggesting that the hazard cannot be avoided simply by locking the device when not in use," the CPSC said.
A child lock or quick reaction are not enough despite your insistence. Any kind of engineering background, or maybe just common sense, would tell you this. You even accidentally figured it out by noting that a treadmill could hurt an adult if they stepped on it at full tilt. This machine and many others have none of the safety mechanisms and failsafes that needed to make them safe enough to use in a home. They're just made to look pretty and save a buck, so they will kill or injure adults, children, or pets just the same.
Omitting essential safety measures and replacing them with a disclaimer is the hallmark of incompetent engineers. Like replacing the protection in a bathroom outlet with a disclaimer "don't use around water".
I tried taking the best possible interpretation but your comments don't show evidence that you have anything really pertinent to say on the topic, just opinions and guesses that don't stand up to reality. Might be worth it to take your time and understand what I'm saying and that way maybe you'll get something useful out of this conversation.
For adult use yes... But perhaps it should warn that no children or pets should be allowed anywhere near it at any time.
> > "It is believed that at least one incident occurred while a parent was running on the treadmill, suggesting that the hazard cannot be avoided simply by locking the device when not in use," the CPSC said.
It doesn't take long for a child or small pet to be pulled under and injured even when the adult is on the treadmill. From the video it looks like there's absolutely no mechanism to prevent this, or even detect that it's happening. The treadmill seems to be happily operating without any supervision of any kind, and is lacking even basic fail-safes that should kick in at least when it's lifting off the floor and hopping sideways while still pulling the child under.
When the government has to warn people to stop using your product if there are small children or pets at home you know you have bad engineering and bad management.
I stand by saying having the machine powered near kids is already game over. They should maybe add a child-safe mode that requires an additional pin to activate, guards are a red herring here
The treadmills at a nearby gym have a cord you clip to your chest. If you go off the back of the mill, it tugs on the cord and triggers the dead-man-switch.
Of course, I've never seen anyone attach the cord.
It has one, I have a Tread and it's referred to as the "key" and has the same type of clip.
Kids of course don't know to use an emergency stop which is why I'm going on and on about maybe not being negligent parents and leaving kids near a piece of power equipment that causes 20k+ injuries a year...
I have a Peloton Tread so I'm talking about something I own? I even mention shopping for one in another comment
What kind of garbage comment is this anyways? I have one "top level" comment on this thread, and every other comment is replying to someone who replied to me. As if I'm the first person on HN to reply to people who replied to them.
So much for "assume positive intent" right?
-
If you think guards mean you can leave your kids near a piece of exercise equipment capable of injuring a person 100 different ways be my guest.
If I seem passionate about it's because of the utter absurdity! If someone left kids playing with a chainsaw and they got hurt you'd lock up the parents....
But somehow sticking a just as powerful motor on a abrasive belt driven device you literally stand on doesn't ring any alarm bells? Chainsaws and treadmills cause the same order of magnitude of injuries a year.
It's a piece of dangerous power equipment. You're being negligent to let kids play near it. End of story.
It could be a far less dangerous power equipment, and should be.
The difference?
Far fewer people who make mistakes will end up with severe consequences.
Had that been done, this story would likely not be a discussion.
The fact that it is a discussion is why product safety is a thing. And it will prove to be smart business too. The fallout from this will be extremely likely to exceed the cost of more safe engineering.
Doing that engineering is cheap insurance for everyone.
Too many stories like this will eventually result in that engineering being a requirement too.
Nobody is defending the parents. I won't.
But, the high price paid did not have to happen too.
Producing products carries a responsibility to the general public. It is cheaper for the enterprise to say, "fuck 'em, they better watch out."
Stories like this actualize those external costs, and justify regulation to insure those external costs are not out of hand too.
Fact is anyone shipping power like this into homes knows stuff happens to even the best parents.
Part of making the money involves doing the work to minimize danger. And it has to be that way, or the carnage would be unacceptable.
These people short cut all that and now have a nice shiny bit of gore to their name.
Next time, do the work, because people are not perfect and sometimes shit happens.
Even better?
Had they done the work, they would be in a strong position to talk about safety around their gear, but they did not care, leaving them in a very poor, "who are you to talk?" position, deservedly so.
There is just so much vaguery here that's all based on a false premise.
A machine that can accelerate your adult weight up 12 MPH. And has a grippy surface for you to run on it. Will ALWAYS be too dangerous to EVER leave your kids near it and you are NEGLIGENT if you allow them to be near it.
I never said it can't be safer, but it can not, by it's very intended purpose ever be safe enough to let kids be near it
This feels like kids found a loaded gun and instead of talking about how to deny access to the gun we can make the gun safer. Sure you can make it safer, but the root issue is the access because it is a naturally dangerous object.
It's as simple as that. We can spitball all the imaginary mitigations until the thing has a 3 x 3 exposed square for your feet and it will still be too dangerous for kids to be near.
All true, and that same machine should be produced safely, because the danger is known and how to mitigate it is also known.
They either were incompetent or did not give a fuck, neither is acceptable.
Secondly, because these are known dangers and mitigation are known, out there, expected, ordinary people may actually see more danger because generally set expectations do not match reality.
The reality being this machine is a bigger risk than may be expected, and it does not have to be. Should not be.
They can afford for it not to be too.
Any competent product safety people would have required the basics, which would take this story off the table.
This company does not have those people, and they should.
A loaded gun is inherently dangerous. Making it more safe, denying access, education all make sense, and the more of those things we do the better off we all are in the form of reduced cost and risk exposure.
Fact is guns are designed to kill things. We can only do so much, right?
Now, the inherent danger on a treadmill is different from a gun. Fact is treadmills are not designed to kill things.
That difference matters.
It can be made such that it rejects the danger condition mechanically, while still performing the task in a reasonable, cost effective way.
Fans are similar things. They are not designed to cut things or kill, but they can do that, unless designed to not do that.
Open, metal blade fans look cool, and are quiet, but people can get fingers into them easily.
Fans with fine screens on them reject the fingers, but pass the air. Fans with plastic blades improve things more. Bladeless fans, or flexible blade fans more still, though one can argue they perform a lot worse and may not always make sense.
All of these are primary, passive safety, engineered in to be a basic part of the device in question.
Doing that is the most robust kind of safety too.
No software, just physics applied to minimize harm. Always on.
For a treadmill, guards and paying attention to basic dimensions are primary safety, always on features.
Passive safety is best coupled with education, and it can all be enhanced further with software too.
Say we made a fan with software and sensors that can act quickly enough to stop the fan should people get fingers in the blades.
What happens when the sensor fails, or the software glitches?
Those are secondary safety concerns. Good to have, but not as good as primary safety is. Not always on.
You are talking about education as a safety measure. And you are not wrong to do so.
Others here, myself included, are talking about basic, primary safety, always on.
Education is not always on. Software and other active safety features are not always on.
Guards and basic dimensions being such that the machine rejects things it could do serious harm to are always on.
None of this is imaginary. None of it is trivial either.
It all adds up, and shipping a ton of these machines lacking basic, primary safety features increases the harm in the world, and for what?
A bit more margin, or other minor considerations.
Remember it all adds up.
Take a decade and a well designed machine with robust safety features and this one, and this one will injure or kill more people than the other one will.
That harm will happen because mistakes happen.
We can't undo dead and maimed people either. Talking about blame and shame, shoulda, coulda, woulda still leaves us with unrecoverable harm that was unnecessary.
Harsh world you want to live in!
Better hope you are not tired, or uninformed because in your world nobody gives a fuck whether you and yours live or die, or get hurt.
In the one product safety people work to make reality, people do give a fuck, and consider talking about close calls and minor injuries, when it all could and would have been so much worse, a very nice problem to have.
So, here is how that will all go:
On an A / B test, the safer world wins by a mile.
In that world we admonish the parents, who still have a live kid, and we all carry on thankful what could be done, makes good sense to be done, WAS DONE.
In this world, we admonish the parents, express our condolences and lean hard on these clowns to either step up or get out of the game.
In yours... well? Good luck. I suppose we can tell the parents they can just make another kid.
We do exactly what you said, AND we continue to improve product safety so that we minimize the harm in all fronts.
Read it again. Everything you say about parents is valid, education done, the whole nine.
The difference is you want to make it all about the parents, despite the fact that this kind of thing could, and does happen to even the best parents.
Because NOBODY is perfect there is a responsibility to design with that in mind, and in this case with this product, this company, that did not happen and it should have, same as the parents should have...
We have standards for all this because when we subtract either error here we end up with a far better outcome.
And that is why there willbe far fewer events overall, and for those that do happen, far less harm.
This is no contest, unless you somehow believe it makes more sense for people making products to not give a fuck about what happens to the people who use them.
We did that already and it was terrible. That movie already played out and the world moved on.
Back in the 20's it worked much more like you advocate here. The verdict was clear and does not favor the position you have advocated for here.
Even seems like a switch where if the feet come off the ground it would turn off. Granted injury would still be possible but maybe more road rash than actually getting sucked under.
Seems like if a cheap space heater can have switched like this I don’t see why a 34 billion company can’t focus on this to come up with something truest innovative and safe.
Regarding the photo switch. My first thought was that dyson is now counting dust particles 15,000 times a second in a consumer grade battery powered vacuum these guys could do something equally innovative.
For the record, I just looked at my (formerly commercial) treadmill and it has no guard. Looking at a google image search for "treadmill" shows maybe 1/3 do not have a guard.
Whether or not it is a terrible design, it's common.
I checked my treadmill and it doesn't as well. I also googled trying to see what they look like. They seem to be used on treadmills that sit higher off the ground.
My guess is that the Peloton design is more dangerous due to the treads, the thickness overall and how to high it sits off the ground. Perfect design for sucking things under.
Yeah what should I look for? Is it like this just a piece of plastic on the 'outside' of the tread? [1]
My proform doesn't have one? though i always wear the emergency clip on treadmills. cheaper treadmills though motors are super slow to ramp up and down. [2]
I just came back from my usual gym, a Snap Fitness, which has Cybex treadmills. Those have the belt running around a roller at the back, and then, on the underside, there's a bar or plate just past the roller, almost touching the belt. So if something gets pushed underneath, it gets peeled off the belt, instead of being sucked underneath. Simple.
Checked my NordicTrack 1750 and it has something similar. Didn't even realize that bar was there but it's good to know it is and what it is for. I'm not familiar with the design of the Peloton treadmill but it seems like most could have something similar.
Oh cool. Still would suck to get sand papered. I wonder if could make something like those construction cutting blade things, where if you trigger the bar with a lot of force it hard stops.
I always wear the clip i've fallen off a treadmill before didn't get great balance on the hand rails stepping off a sprint.
I mean, there is no regulation for it, but it seems like an incredibly risky design for the manufacturer. Quick release forks famously have "lawyer lips" that make sure the wheel doesn't fall out in case you are riding the bike without actually closing the quick release.
I always thought those lawyer tabs were so dorky... until a branch or something somehow unhooked my qr lever descending camino alto, at which point the lawyer tabs really were the only thing keeping me from a pretty nasty accident.
My treadmill doesn’t have a guard either. It does monitor the motor and if it suddenly stops it shuts down. I have observed an accident at a commercial gym where it also shut down on jam.
Early 90's read a long article on safety. First point front and center is safety is a primary system design issue.
Example: Old coworker mentioned working on a controller for an airbag. She said the watchdog timeout was set to a shorter period than required to fire the airbag. So if the firmware went off into the weeds the watchdog would fire before the air bag.
> Looks like Pelotron wanted a nice clean look, so they left the guards off.
But...the guards are usually on the bottom, maybe an inch or two from the edge. For anyone standing or sitting near the treadmill they either won't be visible or will only be a little visible--and in the latter case you can still make them hard to spot by making them the same color as the belt.
Only people who lying down on the floor or who are standing or sitting far away in a big room should see them (and again, coloring them the same as the belt can make them harder to notice).
Unless I'm misunderstanding the geometry (I've never seen a Peloton treadmill in person), I don't see how wanting a clean look would lead them to leave off guards.
Photoelectric safety switch between the back legs would’ve also worked (like a garage door beam), depending on how sturdy those legs are mounted for alignment.
As someone who has accidentally removed part of their finger by getting it trapped in the pinch point of a belt sander, ouch. Yes, this is exactly the same arrangement, and I can see it causing very severe injuries very quickly.
I don’t think this is exactly the same arrangement - the belt goes “all the way around” on this where the pinch point is between the belt and the ground rather than the belt and the internals of the sander.
Here the force is the weight of the peloton unit, rather than the motor pulling you into a fixed width gap.
That was my point, applying pressure to the object held against a belt sander is how you remove material. Those slats probably have some kind of tread on them which makes this a 500W, couple hundred pound belt sander capable of flaying small pets and children. If you have a strong stomach you can look up “belt sander injuries” to get an idea of what I’m talking about.
I was thinking this as well. For how much tech this thing seems to have, and for such a high price tag, they could easily justify putting, I don't know, two or three IR sensors down there.
Maybe but this is rarely the option we pick in the world. Thousands of children could be saved yearly by lowering the speed limits on roads but we don’t do it.
Thousands of non functional products can ruin the company while this event could be explained with “Don’t allow children around running and dangerous equipment.” You wouldn’t allow small children around a running lawn mower or car and a treadmill requires the same level of respect.
All the lawn mowers I've seen at least have a safety mechanism where they stop as soon as you let go of the handle. And the dangerous parts are shielded on all sides during normal operation.
The peloton in the video has no protection to prevent dragging something underneath it, and just kept going as it dragged the kid under it. The motor stalled, locking the kid, then started dragging again as soon as the kid managed to almost get free.
This won’t work if your tall. On fast runs my ankles routinely left the back of the treadmill. Keeping your torso tracked in the “middle” is harder the more fatigued you get. A sudden e-stop will fuck you up if your running at 8mph.
I'd expect that they would just cut power to the motor, so it would not be an instantaneous stop. It would take a second or two. That probably wouldn't fuck you up.
Remember that on a treadmill you are running in place. You don't actually have any forward momentum, so you don't have to worry about slamming into the console at 8 mph. It's not like if you were running outdoors and someone suddenly put an obstacle a few feet in front of you.
You should just have maybe a little stumble, at most requiring that you use the rails to steady yourself.
Here's a video of someone running when power was cut to their gym [1]. The power goes out at 1:07.
It's probably worth actually testing this at various speeds working up to your normal maximum speed. I've done that (although I just do fast walks, not running) because one of the reasons I might be on the treadmill instead of outside is bad weather, which is sometimes accompanied by power failures, and I wanted to know if I should avoid the treadmill during those times.
It wasn't that. It was the torque and incessant "gotta go" programming in the motor. If the speed is stuck at zero mph and the motor tried more than 5 seconds without changing the belt speed CUTOFF immediately.
More than that, they could put multiple weight sensors on the platform to make sure an adult is on it in order for it to start and continue operating. As an added benefit, it could use the signals from the sensors to detect roughly where the user is stepping, how hard they're hitting it, how frequently, etc. That would be useful for gait analysis and other fitness-related stuff.
Additionally, a sensor on the motor could detect a sudden increase in power required to keep the belt spinning at a particular speed (such as something getting under it). This would trigger an emergency stop and possibly an audible alarm.
It's not very clear, but from the last two descriptions it sounds to me like it has a shutoff sensor that can engage by accident or leave a child under the "off" machine.
But how do you make a pinch point completely safe? Especially with a slatted belt that doesn't have a continuous smooth surface? Where it goes under the guard is still a pinch point. Much as I prefer inherent passive safety to active safety, an optical beam or two would work here - machine does an emergency stop as soon as the beam is interrupted.
Yikes. Sounds like they just didn't do a safety review.
Given that they are also apparently "opposing" the recall, I wouldn't trust them with a 10 foot pole now. They've put themselves in the same category as the Schlitterbahn, whose water slide decapitated a 10-year-old.
What I find most amazing in that video is that the treadmill stalls, detects an error condition, reverses for a second, then immediately goes full pelt forward again.
It’s crazy to think the treadmill has the ability to detect a potentially dangerous situation (the belt is completely jammed). But instead of shutting down, it does everything it can to keep operating and doing as much damage as possible.
Peloton didn’t need to add extra guards or sensors, they just needed someone to program the damn thing to shutdown when it jams. Rather than programming it with some dangerous and stupid auto-unjam sequence.
> Peloton didn't need to add extra guards or sensors, they just needed...
This is a very naive comment. Substantial damage can be caused prior to the system detecting an error condition; this is no excuse for not having guards.
I can't tell if this is sarcasm or not, but software should never be used to compensate for unsafe hardware. That lesson should have become standard knowledge after the Therac-25 disaster, which happened as a result of replacing a hardware interlock with a software one.
Software is used to make hardware safe all the time. Look in any factory with automation and you’ll find hundreds Programmable Logic Controllers (PLCs) running safety critical operations, and acting as interlocks.
Look inside a car, and you can find a dozen software based systems designed to make the hardware safer. E.g. Traction control, body stability systems, ABS breaking, software that decides which airbags fire in a collision, lane departure computer vision systems, air pressure alerts, engine control systems that look for unsafe operating conditions etc etc
Ultimately software can be made safe. Long and rigorous software development processes and audit process are what makes it safe.
Ultimately there’s nothing inherently safer about hardware safety systems, they’re engineered and suffer from human error.
Proper risk assessments are always required, and safety systems should be built to match the risk. There’s no one size fits all’s. In Pelotons case, they clearly failed to assess and address the risk properly.
I agree with all of your points here, but I think there is an important distinction to be made between passive and active safety features. All of the examples you give are active features, which means they require continuous error-free sensing to remain functional. Passive safety has inherent advantages.
Every treadmill I’m familiar with has a safety key attached to a lanyard that the runner is supposed to attach to their waist band that immediacy pulls the key and deactivates the treadmill if the runner falls.
I don’t know how this treadmill is still operating after the older child gets off. Maybe the key was left in place and the treadmill left turned on but that’s really f’ing dangerous. A modern treadmill should reset itself after a period of non-use and require the key to be removed and re-inserted. That’s in addition to a physical safety guard under the belt this treadmill clearly should have (see my post elsewhere in this thread comparing the size of the Peloton to another premium residential brand).
This Peloton is a death machine. I didn’t realize the CSPC doesn’t have the regulatory authority to force a recall.
ETA: just learned yet another thing. It’s really hard to physically reach the switch to turn it off. The switch is located under the treadmill. Wtf. See the 8 minute mark of this video:
I was a child once. I have distinct memories of playing with treadmills. Turning them on while sitting on them. Running Hot Wheels Cars on them while they’re running. The design of these treadmills is pure negligence regardless of safety keys or kill switches.
There are advantages to large rollers, a high deck and a heavy chassis. But these are all things that make this a more dangerous machine that other residential treadmills. Peloton should reduce that risk with a physical barrier under the deck. But even with that, this treadmill should still have a safety key and an accessible (for an adult) physical on/off switch.
My True 500 had nothing to prevent something from being pulled under it. Landice makes high end residential treadmills and their models don’t appear to either:
However, compared to those treadmills, the Peloton Tread+ has much larger diameter rollers and step-up height. It's also much heavier (455 lbs vs 340 lbs). It looks uniquely dangerous to me among residential treadmills.
Thank you for sharing these pictures. They clearly show the difference in height.
The Landice can be a threat for a pet or can grab/scratch a foot. The Peloton clearly/visible is higher and a small child can sneak or can be pulled under it.
Landice calls them deck slats and they are structural members that hold the two halves of the treadmill together. You can see that clearly on page 4 here:
(Aside, but I love that Landice puts their excellent technical documentation online.)
If they also serve a safety purpose, great. (My True has these too, but there's about 2-3" between the horizontal bar and the belt.)
The Peloton Tread+ must have the horizontal structural cross-members inside the belt. They'd have to make the side rails taller to accommodate a safety bar:
This is a different incident than the one mentioned in the article though. In that incident, the child was "discovered trapped under the machine and not breathing" and was hospitalized with a brain injury.. In this video, the child does get out from under the machine and walks away. Although the only difference between the two incidents is probably the ending... this child could have also easily become trapped.
Yeah, looks like the ball was what kept the machine propped up so that there was enough space for him to walk away. So basically an amazing lucky coincidence for him.
What? You can't really faze me with porn but I genuinely don't like watching videos of people getting injured.
This is why people who miss WPD on reddit kind of worry me a little. I understand the philosophical side to the argument and all that but it's probably the only stuff online that really sets my caveman-alarm off.
I dunno if removing WPD is really doing any good though. It wouldn't surprise me if it had a chilling effect on the NSFL label. Besides how are we going to figure out what level of warning is appropriate if we can't find anyone willing to see it?
My first instinct was to assume this to be parents doing a blame-game at Peloton for their own negligence.
But this actually makes sense.
A child or pet gets in there it really is game over. I can imagine even if an adult somehow slips up that it would be insanely painful and difficult to free yourself -- especially at a running pace.
They don’t really get to complain that the machine didn’t stop when they fell if they didn’t wear the clip specifically made to stop that event. You can’t sue the bike company because you didn’t wear your helmet.
Of course you can sue the bike company, and if your foot got caught because there was no chain guard, causing you to crash and be injured in an accident, you might even win. Even if you weren't wearing a helmet.
This is truly terrible design and I hope the machines are recalled... but there is more than a little onus on the parents here (at least in case of the linked video - can't speak to the other incidents). If you are responsible for children that age you should be actively evaluating your home for these types of risks. Letting them play on a piece of machinery like that unsupervised is neglect imo. From the time the little girl notices something is wrong (and is presumably and making a lot of noise) there is still no adult visible after a full 35 seconds has elapsed.
I think that's really the damning bit - in the video shown, you can see ripples in the belt still moving towards the kid just after the sister runs off. The belt's jammed, the user's dismounted, and it's still trying to win.
At some point the treadmill actually spits the kid out. It reverses the speed for like one, two seconds when the kid was jammed and everything had stopped, but then it starts going back in its original direction and the kid gets sucked back in before he has time to get clear.
Yeah the Peloton does detect itself getting jammed. You can see it automatically start running backwards to unjam itself in the video, then start running forward again.
Crazy to think that someone thought that was a good idea.
This looks like a typical startup disruption approach where they wanted to build the hardware from the ground up and didn't ask industry experts why something is done the way it is.
Reviewing profiles of Peloton mechanical engineers suggests they employ experts, including from heavily regulated industries. It seems unlikely that this was built without discussion or research. However, it’s true that they misjudged the risk here.
In this case wouldn't it be a cultural thing for the management rather than the engineers? They're both human, but if the boss says get it done or get out, the standard is set from the top.
Certainly. It would be comforting, as an engineer myself, to think that management must not have considered the problem. However, the truth is that management, design and engineering are all quite capable of trying to avoid problems, potentially to great lengths, and failing anyway.
I have seen lots of trade mills without a guard there. Just googling image shows quite a few that don't have a guard there.
If the issue is the guard it's more than just peloton that then needs to revise their design.
What I do find more interesting is how the treadmill is still running even after the other kid gets off. There is a reason a lot treadmills have lanyard that shuts it off.
I don't understand how they released it like that. When my mom told me about kids getting stuck I immediately said "yeah under the back right?" And I've only seen Peloton in ads.
What a stupid design. They deserve to get sued into oblivion
Perhaps "most" is an exaggeration, since I have one which doesn't and the comments here and images found online suggest that many others don't.
IMHO that video shows negligence on the part of the parents. There was a camera there but no physical presence. Exercise equipment is dangerous if not used correctly, and although I'm one who doesn't believe in "helicopter parenting", I certainly wouldn't let my kid play with a treadmill with only a camera watching.
Yes, the video confirms what I expected from the description. And yeah, the video is a bit gruesome but apparently he was okay.
We have a Peloton Treadmill (now called the Tread+, but the original one is the same as the Tread+), in fact my wife is running on it right now.
If you do a Google Image Search for "treadmill", you will see most treadmills are belt-style, and often have a metal bar underneath the back, a few inches in. NOT a cover on the back -- the plastic cover is typically on the FRONT of most treadmills, which won't help with things getting sucked underneath at the back, which is where this would happen.
The Tread+ is a "slat style" treadmill similar to something really high end like a $10k+ Woodway, but other brands also sell slatted style treadmills.
It's definitely obvious how something can get sucked under the back of the tread, however I'm not sure how much better having a protective bar is; I really wouldn't want an arm to get sucked between the belt and a protective bar underneath, which is usually at least a few inches from the underside of the belt.
And of course the Tread+ is something like 450lbs; the heavier the treadmill, the more problematic something like this is likely to be.
Having it shut off when something goes under it does't really help; I'm not sure I'd want my kid to get crushed by a stationary treadmill either.
Like most treadmills, the Tread has a safety key which you're supposed to wear (but not everyone does, and of course kids playing on it are not going to). Actually, I just realized the obvious answer is, if you have kids, take the safety key off when you're not using it. But, mistakes happen, and of course something can also get sucked under when a properly trained adult is using the device as well. (One of our cats took a serious blow to the head from the swinging arm of our elliptical, but he was fine; cat.) I definitely wouldn't want anything going under the pedals of our Peloton bike either; metal pedal plus speed plus flywheel mass would be a REALLY bad time.
Back to the Tread, the belt stops if there's no weight on it for "a while" but I'm not sure how long; certainly it takes more than 10 seconds. And, again, you'd be left with a kid trapped under a STOPPED treadmill which is not much better.
There's no particular safety for turning it on. Typically I select my profile on the touchscreen but the knobs on the side to start the belt are active before starting a class. I think they might even be active when the tablet is "on" (tap the power button on the cross bar first) but before choosing a profile.
I suppose a workaround would be to require a login before turning it on, but of course this is a step most other treadmills don't have either.
Someone trapped under a stopped treadmill is much better than some trapped under a running treadmill.
At least with a stopped treadmill you’re not fighting with the treadmill the escape, and if the belt enters a free running mode when off, then it’ll make it easier to escape.
If you watch the video you can see the kid get trapped by the treadmill and completely stalls the belt. But rather than shutting downs, the treadmill just keeps trying to run forwards. There’s even a moment where the belt runs backwards for a second before it start running forwards again, re-trapping the kid.
Well, depending on the design, it might be possible to "come out the other side" .. but yeah, that's probably a bit like hoping to be thrown clear in a car accident. You MIGHT be better off, but statistically the average person will not, so wear a seatbelt.
Edit: confirmed (warning: video shows a child being injured): https://www.youtube.com/watch?v=onXNnlCYJ4Y