Regarding ECF Safety Specification For Metal Tube Mods

[warlike36]

Since ECF has detailed info on the failures/explosions, why not post the info...like batteries and units involved ? And any and all pertinent info ?

Is there info that may help me avoid those same conditions ?

I would like to know and think it would help me as a consumer and user of mods.

I meant no harm.... and I personally will not recommend said mods to anyone, but at the same time I have no idea which mods are actually involved, I have tube mods that use one and two batts, and have vents, but that to me is half the story.

If I have a hard time finding info, then a new vapor simply will not find it. Why not have a thread that is factual on all of the incidents where things went catastrophic ? I also think that said mods should be re-called. If I own one, I will happily send in it for safety modifications. If we are at "cliffs edge" then I feel these actions are necessary.

Also, let me say I support what you are trying to do here roly, I just want it straight and in simple black and white.

 

[Rocketman]

I would watch your vendor's site for a new and upgraded safer model, with the one you have offered at a 'sale' price. That would be an indication you should start worrying.

If your vendor is still offering totally unprotected cells just for stacking then they probably weren't involved. The ones that had failures with injury probably have removed the unprotected cells from their web sites.

 

[rolygate]

You hit it right on the head Roly. As an engineer with over 30 years experience I would like to add that in industry, a design must be minimum 100% over spec for safety if the thing is going to consumers or be used by the general public.

Hah Jan, if you wanted to talk about safety factors, design loads, cycle life before fail, fail rate per 100,000, safe failure modes or whatever around some of our mod makers, they'd think you were talking Chinese. They have no concept of such things.

 

[pumasforpets]

Hah Jan, if you wanted to talk about safety factors, design loads, cycle life before fail, fail rate per 100,000, safe failure modes or whatever around some of our mod makers, they'd think you were talking Chinese. They have no concept of such things.

I've worked with professional engineers for quite some time. That arrogant attitude is extremely common among them. It's the same arrogant attitude that allows you to think nothing of making design decisions for an industry without bothering to back them up with anything of substance.

Because I said so is not now, nor will it ever be good enough.

With such a large community, there are more than plenty talented folks to review your work. Post up that chinese gibberish you're sure no one will understand so others can review it. That way we'll know if it has substance or was pulled directly from thine ...

 

[Rocketman]

Backyard testing, just to get a handle on the scope of the problem/solution would be better than guessing.
Even NASA started out in the back yard (of a weapon test range). Refining the parameters of a scientific test takes some assumptions, proof of concept, and some data collection. We aren't anywhere close to determining margin of safety, predictive failure rates, or safe failure modes, not just yet. Final safe design recommendations? Baby steps.


Let's blow something up. On purpose

 

[Retriever]

I would really like more details of what mods failed, and the improvements the maker was implementing, if any. We have big battery mods failing, stacked batteries failing, medium sized batteries failing. What next? It has totally ruined my experience where mods are concerned. I will stick with my KR808's - until one of them blinds me.

Byt the way - Thank you Rolygate and SteelJan for your efforts in this important area. We need this info!

 

[pumasforpets]

Actually rocketman, that's in my (eventual) production plan. 2 protos then 15 test units with 5 of them set aside for failure testing with as much documentation as I can muster. Not sure about others, but I really think it's important to see it in action...not just in theory.

 

[cbrociuos]

Actually rocketman, that's in my (eventual) production plan. 2 protos then 15 test units with 5 of them set aside for failure testing with as much documentation as I can muster. Not sure about others, but I really think it's important to see it in action...not just in theory.

destruction testing is a great idea. its probably the only true way to see what works, and what doesnt. im sure posted videos of said tests would be one heck of a selling point too. this way customers would know you have designed your product using a "scientific" approach and not guess work. im very interested in seeing some videos, keep us posted.

 

[Rocketman]

Actually rocketman, that's in my (eventual) production plan. 2 protos then 15 test units with 5 of them set aside for failure testing with as much documentation as I can muster. Not sure about others, but I really think it's important to see it in action...not just in theory.

An analysis technique that might be useful is a 'fault tree'. Every point in product being analyzed for the chance of causing a critical single point failure. Then every one of those is analyzed against every other point for synergy in creating a critical two point failure, and so on and so on.

For example, a protection diode to prevent reverse current can fail. One failure mode is an open circuit. Probably end of the line for that branch of the tree. But it could fail shorted. Still not a 'single point critical failure mode' but may contribute to a critical failure if batteries where subsequently inserted backwards possibly causing an additional circuit failure. The combination of these two failures could result in a critical failure, when the third failure mode, a short between the upper cell case to the mod metal tube causes that cell to vent when the fire button is pressed with a shorted boost Mosfet and shorted protection diode.

When analyzing corrective measures each should then be evaluated against all previously identified failure modes. A hot spring may not protect against the critical failure scenario above. The choice of the number of branch levels depends on the worst case result of a critical failure. For a complex system like the (retired) Space Shuttle system, post accident failure analysis took many months. Every point in the system had previously been analyzed for failure modes, by hundreds of engineers and scientists, but still failed, twice.

 

[rolygate]

I still feel the need to know from where the 2in x 5/32 x 3+ came. I think an exact spec like that can not be arbitrary.

Sure. However this is a discussion point for within the Suppliers forum, and I just can't duplicate all that here.

I devoted all of Saturday and part of Sunday to answering this thread because I felt it important for members to understand the reasons why we have to do what we are doing. Honestly, I am afraid my time allocation is about done here, because I have done my duty. I spent more than one whole day on this. It is not reasonable to expect me to now pursue two duplicate lines of debate, since I now have to get into the specifics with the manufacturers. I do apologize.

 

[tj99959]

Ty for giving us your time Roly. While we may dissagree on specific points, we really do understand the intent.

 

[rolygate]

I would really like more details of what mods failed.....

Only two-cell metal tube mods with no electronic protection have exploded. We have no reports of any other types experiencing catastrophic failures.

....., and the improvements the maker was implementing, if any. We have big battery mods failing, stacked batteries failing, medium sized batteries failing. What next? It has totally ruined my experience where mods are concerned. I will stick with my KR808's - until one of them blinds me.

Byt the way - Thank you Rolygate and SteelJan for your efforts in this important area. We need this info!

Every possible variant of a lithium battery has failed, some more than others. Lithium cells are not 'safe' as we would normally understand that to mean. They are 'unsafe' in that you might expect to have a problem in one in 5,000 units or whatever it is, that number being orders of magnitude more frequent than for main street AA cells and so on.

However it is important to note that the majority of these failures occur while on charge, and the majority of those on-charge fails seem to be eGo clones. Other meltdowns are not uncommon, but no real cause for alarm. And if you always charge your batteries in a Li-Po sack (or a cookie tin) then you don't really have much to worry about.

I'd say the most serious issue otherwise would be ensuring that your chosen device absolutely cannot operate while in your pocket or purse, as that can lead to all sorts of issues.

Only 2-cell metal tube mods explode.

They don't, if basic safety features have been implemented.

One day a single-cell metal tube mod will blow up as that's what Murphy's Law dictates, but it hasn't happened yet. It will happen because:
- Someone thinks they have an AW red battery, but actually it's a counterfeit, made from a reject Li-ion unprotected battery.
- It shorts out to the body of a metal tube mod. They do that a lot, it seems.
- The metal tube mod in question is one of the sealed type, built by an optimist. Optimists in engineering are not your friend.
- Bang.

 

[Rocketman]

Roly,
I'm sure you've seen the drawing note: "Break all edges".

Maybe is doesn't translate well as the Chinese don't seem to be following it.

A popular replaceable battery mod that takes 14650/14670 cells has damaging sharp edges and a tight fit that seems to easily cut the wrapper on protected cells.

Something as simple as a 'standard industry practice' could save a few failures.
Here's an example of a 'near miss'.

Granted it is a single cell mod, but the concept and potential with a stacked battery mod is the same.

 

[tj99959]

Only two-cell metal tube mods have exploded. We have no reports of any other types experiencing catastrophic failures.



One day a single-cell metal tube mod will blow up as that's what Murphy's Law dictates, but it hasn't happened yet. It will happen because:
- Someone thinks they have an AW red battery, but actually it's a counterfeit, made from a reject Li-ion unprotected battery.
- It shorts out to the body of a metal tube mod. They do that a lot, it seems.
- The metal tube mod in question is one of the sealed type, built by an optimist. Optimists in engineering are not your friend.
- Bang.

And it will be because someone used a magnetic spacer to make a wrong sized battery fit in the tube.
A magnet simply can't distinguish between the metal top of a battery and the metal side of a tube.
In fact it wouldn't surprise me if that was the cause of half of the two cell failures.

 

[Rocketman]

- The metal tube mod in question is one of the sealed type,-------
- Bang.

Can we assume that a cursory visual evaluation by a potential user would indicate a rather sturdy bottom end? With little chance of popping first, and at a somewhat safe (?) pressure?

 

[SteelJan]

At the point where a vendor is purposefully engaged in product substitution all bets are off. Show design A, claim to sell design A, actually make design B, is a dishonest act.

I receive ecigs and mods from vendors wanting me to do a video review. The first thing I do is look it over electrically and mechanically and if I find problems, I give the vendor a list of things that are problematic, to give them a chance to respond, e.g. we know about it and it will be fixed in Rev 2, or they often ask for my pictures, data, etc, so they can go correct the problem and in the meantime ask that I not go forward with my video review. And I do so, because my goal is helping make ecigs safer, not about competing for who the most video reviews, lol.

Many times what I find is that after someone has designed a device and perfected it with their manufacturer, even gotten beta test units and etc, once they turn on the manufacturer to do a full commercial run, what goes out to the public does not conform to the original design. The manufacturer, which has been chinese in every case I've found so far, has silently cut corners. It can be substituting the battery(ies) (either loose batteries with a kit or ones that are encapsulated inside the device), putting an inadequate or faulty charger, or in one case both the internal encased battery and the circuit board in the mod had been completely replaced with cheaper items from the internal guts of another ecig the manufactured also produced. There may be initial batches that do conform before the corners are cut. Many vendors/designers/manufacturers are shocked at what I've found.

So I've come to think that even designers who are determined to follow the best of ecf's guidelines and also use their own design and testing experience to make the best product, can sometimes find themselves selling a dangerous or faulty unit.

 

[rolygate]

Roly,
I'm sure you've seen the drawing note: "Break all edges".

Sure have. To be honest this is the biggest worry I have about gas vents as it won't make the situation any better at all. But anyway...

Please PM me with that mod's name, I need that stuff.

[edit]
Sorry, forgot I block PMs except from Moderators.
chris[AT]e-cigarette-forum[DOT]com

 

[rolygate]

And it will be because someone used a magnetic spacer to make a wrong sized battery fit in the tube.
A magnet simply can't distinguish between the metal top of a battery and the metal side of a tube.
In fact it wouldn't surprise me if that was the cause of half of the two cell failures.

This is a good point. We may need to advise that spacers shouldn't be used unless they are taped or whatever.

 

[rolygate]

Can we assume that a cursory visual evaluation by a potential user would indicate a rather sturdy bottom end? With little chance of popping first, and at a somewhat safe (?) pressure?

To me, the only metal tube mod that I know can't explode is one with large holes in it of some kind. But maybe I'm less trusting than most.

 

[5vz]

http://www.roehrigengineering.com/Clevis Options/500 clevis.pdf