This blew up! Let's open it to find out why 

“Sooty Skidmark” sounds like a 1950s blues band bass player.

"That thing you're using is going to break." Most people: "Oh no, I'll have to find a replacement." Clive: "I can't wait."

maybe that black stuff is plastic melted from the four posts that center it, or some kind of glue, carbonizing and conducting because of the heat

So funny. UK people get excited over the smallest skidmark in their shortcircuited gadgets. Come to mainland Europe, where we do not have those pesky 3A fuses in the plugs. Over in Europe, the short circuit will be able to dissipate much more energy before the 16A circuit breaker pops.

We use a couple of Peltier-effect dehumidifiers in a small motor/mechanical cabin at the observatory. During the peak summer months, they pull about 2L/week out of the air in there. We have a larger conventional dehumidifier in another part of the "stack" which does most of the work. While the little Peltier-based units aren't terribly efficient, they require no maintenance and just keep on trucking. For our application, the desiccant ones would just be too much work.

Looks like the plastic supports in the corners have melted and carbonized and provided a conductive path to start the flashover... I would bet that if you had a new one and either cut off those supports or put a bit of heat resistant kapton over the plastic that the heater would have a much longer lifespan.. For heaters that have already been used, you might be able to disassemble them with somewhat more care than you exhibited in this video, and either sand/scrape/chemically dissolve the melted plastic and then perform one of the steps above to keep the plastic from melting onto the element..

The reverse breakdown voltage of a typical LED is only in the range of 5 volts, so in reverse conduction in this circuit the LED will only have to dissipate around 5mw peak, which won't bother it in the least. People often assume that reverse conduction is bad for semiconductor junctions, but that's just because the most common case with reverse conduction involves much higher breakdown voltages and much lower source resistances, which together make for incandescent (or even pyrotechnic) levels of heating. So there's nothing particularly wrong with running an LED straight across the mains like this with a high value resistor, other than the existential pain of the circuit being very inefficient.

you could grind the edges clean again with a diamond wheel, and then apply either silicone grease, or silicone sealant, or ceramic paint to seal it and retain the sealing at higher temperatures the device functions at. A middle step that may or may not help or prove smart, is to etch the edges with an aggressive acid to dissolve any metal tracking, and wear away the sharp edges such that electric field emission points are smoothed and rounded to reduce the risk of arc/flash over.

If the substrate is ceramic, the only way of removing the tracking mark would be to grind it off, scraping it with a metal implement will just exacerbate the problem.

I suppose when you put an unsealed heating device in a thing that soaks up moist air, that device might just get moist in itself and go bang, because electricity loves it when substrates get damp and provide a path for current to have a wander across... :P

Oh c'mon Clive, you know we want to see you try to clean the sooty skidmarks off then power up that heater in the open to see if it works...or goes bang. 😅

As a controls engineer, I am adding 'sooty skidmark' to my lexicon of phrases to use when faced with working on buggered equipment.

The 3 pin degaussing thermistors used in colour CRT TVs often used to fail in the same way.

"Sooty Skidmark" was my stripper name.

I shall forever refer to capacitor discharges and mild electrical shocks as "zappy-poos"

What are the little right angle pillars that hold/centre the heating element made from? It seems mighty suspicious that it's failed/failing on the corners of the heater right were those pillars are.

...There's "Dupli-Color High Heat Ceramic paint" that can be sprayed on car exhaust systems or brake calipers - claims to be "able to withstand temperatures up to 2000 degrees F" (approx 1000 c). Might work if non conductive type chosen...Bit dubious it can cope with 1000 C....

You are the man I should have been. My late brother and I used to make stuff with valves, then transistors came in and we made a superhet radio. Then he got married and I went to college and got married. Please keep doing it for my brother and me. It's fascinating.

The high temp plastic wasn’t high enough temp. How very unfortunate.

5:36 🎶🎤😏 "I just dropped in... to see what condition my air dehumidifier was in..." 🕺🎵

Get another one, and break those four plastic things that 'align' the PTC element. Use gloves etc. - you don't want to touch the ceramic, the oil of your hands will track the same way the molten plastic has... Maybe you could wrap the plastic legs in kapton tape, but better to just be rid of them and be careful to align the element on reassembly.

Can see that those plates were passing current onto the corners only and cooking them corners of the heating block. Seems the point of failure is baked into the design, oh that was bad Duncan. Thanks for the video Clive, hope the show is going well.

you could try grinding off the sooty skid mark. it's a sintered block of ceramic so if you remove the shorted bit it might still work.

When you said "it is defunct now" I immediately flashed back to Monty Python, Dead Parrot!! lol

Don't pick on Sooty, He's a harmless little bear !! Lol. 😂

The OPTIMUM way to dry pink silica gel is to pop it in an ELECTRIC oven ( NOT gas , these produce water vapour ! ) ... until it turns DARK blue .... tip it in a clean jam jar ... and it will keep for YEARS , nice - n - dry ........ but , at least you have a few bits - n - bobs for future usage ..... ( tried - n - tested ) ............... DAVE™🛑

This makes feel better about using the oven to regenerate my Silica Gel

5:36 - Yeah, yeah, oh yeah, what condition my condition was in... 😆IYKYK I'd been sorta-considering getting a US version of this, if available, but now I'm giving it a hard pass.

PTC heater modules aren't very much to buy would be a interesting little project to buy a better one and fix this unit, might have to design and 3D print some brackets but still.... FUN!!!😁

Thanks for the show Clive 🍻

I wonder if this could have skewed your results in the dessicant testing video? If the heater gradually has changed specs, maybe the maxtemp has gone up or down over time? As for the led, I wonder if they’ve used a neon in a previous version, and just put an led in its place in the updated version, without changing the schematics (other than a different resistor value).

I think about you a lot. You explained my 2 electric shocks from a heavy duty sewing machine and a hissing spitting phone charger. Good man. Good man.

Oh bugger, i bought 1 of these, to use in the car in winter. Worked really well.

Very interesting thanks Clive 😊

The corners are also a high voltage stress point (point discharge), so that might be a contributor. Rather than scrapping it, you can get PTC elements from the usual suppliers, or, how about one of those metal - jacketed power resistor elements (with say an NTC probe and simple load controller?)

5:30 It's fjording for the pines...

I would love to just sit down and disassemble things with you to learn how they work.

400 grit wet & dry to get the grot off and a strip of kapton (the yellow, high temperature, high voltage) tape to prevent the plastic carbonising onto the PTC material, and you're good to go.

The honeycomb thing that failed due to arcing/flashover: you have a bunch of these dryers. Stack two honeycombs that haven’t yet started to fail/carbonize in a working unit and then… maybe you’ve doubled the distance it needs to arc so maybe you’ve 4x or greater extended the life of the heater???? Maybe.

The same happened with my hot glue gun after a few years, although it might have blown more violently by the looks of what remained of the PTC heater and its surroundings. It made quite a bang and it's worth mentioning that there was no fuse in the gun to make for the lack of one in the plug like you have in the UK.

Clive have you seen the videos on RU-vid for a electric device to remove itching from a bug bite sting? They use a small ceramic heater maybe 4 or 5mm square and heat it up for 4 to 10 seconds on the sting/bite area, this is said to trick the body to respond to a burn and not the sting. I guess it only gets to about 120 to 150 degrees f "48 or 50 C". They plug it into the smart phone for power and use a ap for the timer. What if you build a unit that uses a push switch and USB power to save us Money as they charge anywhere from $50 dollars and up for these devices. If they even work.

I really love your videos. 🎉 But I'm also a bit scared of these appliances. It's a Chinese version of Russian roulette 😂 What would happen if the fuse in the plug would have been a 16A, or maybe a fake one 🤺💯 In 99% I would expect a louder pop and nothing more, but still I see a slight chance for a fire or even smoldering with lots of nasty smoke that is more mortal and can also destroy the whole flat.. I remember your video with the Car 12V Charger that had a different component design that was changed by the Chinese factory without noticing the company that had designed and ordered these products. So you never know wether safety critical components have the right specifications.. (also best example, the 2KV Y-Capacitor) Keep the good work up! ❤

It is always fun when troubleshooting a widget trying to guess the failure mechanism prior to beginning to debug.

Necropsy - autopsy is done to your own species (i.e. a human).

I wonder why they used a square shape for the element and not a round one. It would have avoided the corner problem and secondly fits better to the fan.

It looks like the plastic brackets in the corners react with PTC element and there is some transfer (plasticizer seeping out?) and providing path for the arc?

Like everything which contains some form of heating elements, from the east, it sooner than later fails. The assemblies that contain resistance wire, such as glow heaters and hot plates are the worst. The west used to use proper Kanthal resistance wire, which is much too expensive for products from the east. But of course you can't get anything else now, so don't buy glow heaters any more because you'll have to buy a new one every two to three months.

Clive they used to use Mica as a high heat electrical insulator its not hard to cut into a shape instead of the i guess its a ceramic material

I have just got one of these to keep my 3D printer filament dry! Do you think there is anything that could be done to help keep it alive? (better airflow into the unit? or reduce the power somehow) Or could there be another module that fits? 🤔🤔

Use a circular coil around a quartz rod full of 3-6mm quartz/glass beads or boiling balls. Use a PID for setting the temp and the fan can run constantly at a low speed? Peak performance would see you using cubic zirconia beads for higher thermal conductivity with an air filter on the fan but both those are somewhat overkill.

I didn't know flat ceramic PTC heaters are a thing, I expected to see bare resistance wire.

Hi Clive, skid marks ,does that mean they are crap? Many thanks from Nr Chester.

The tracking may have been assisted by something emitted by those plastic corners when heated.

Round the corners. Its the sharp edges that started it.

I was waiting for this to be run on the Hopi throwing off the power factor.

Sooty & Sweep: SkidMark was the Zeppo/Pete Best of the trio.

could you adjust the circuitry on the others you have to prevent this happening to them? Another vid potentially? These vids of yours really chill me out, proper stress removers hahaha.

Ahhh and I was wondering if we see a video where something blows up! 😁 Three of these PTC elements are often found in those 12V car windscreen demisters that one can plug into the cigarette lighter socket. 50W each as far as I remember. Wonder how much this one had.

It does look like a short distance to cover for the voltage applied. What is it in the UK 230-240 volts? It may fare better in 100 or 120 volt counties. I wonder how often the filter material on the bottom cover clogged up with dust?

I was expecting you to say, "Nothing a wee bit of toaster oven wire can't fix. Shall we?"

Needs a bigger fuse Clive 😅

If that fan doesn’t become an upcoming ionizer video soon then I shall be deeply disappointed

Thank you Clive now we know.

the ptc 'may' only be rated for a lower voltage.. 120v maybe?😉

I'd be more surprised that it initially worked, instead of wondering why it failed

At least you salvaged a few parts. Probably the reason going cheap.

It looks weird to me that the soot is around the corners is where technically the metal plates dont even touch it, instead the plastic? frame does. I am no engineer but the current path in this case looks weird. Maybe those brackets melt/burn, deposit carbon and the carbon itself flashes over?

Hey Clive, a lot of tools like grinders and drills fail because the capacitor pops. Can you make a video about their size/alternatives and why there are some with 2 and some with 3 wires? Often I cant find the original part for repairs.

What a shame I really liked the idea of such a device. Any recommendations for a better quality alternative?

If it can be reassembled, it might be interesting to compare drying times with and without heat.

I remember when you first opened one of them, it really didn't last too long

Moisture dripping down from wet silica, if you put dry silica into the cartridge it will probably last a lot longer.

It blew its load, was too high!

This device should have just used an incandescent bulb as the heater. Note: Metal filaments *_are_* PTC resistors.

I thought of a sealant on the edges, but I expect it could track over through the holes too. Especially if any dust gets caught in one and gets cooked.

I'd probably have called it tracking across the **insulator** 😊

Thanks for another great video 👍

What's the stated voltage spec on these? I'm wondering if they have pushed their luck and said it will take 220, don't know what the IoM is like but central Scotland it's usually the high side of 250, used to be filament light bulbs didn't generally last long.

Thanks for the update on this device as i do own one and i like to know if could potentially blow up 🙂 , I remember you saying on a live stream it may be a issue , haven't seen you done any lives in some time mind?

Another example of why you are so popular of the Tube Clive but you really should be wearing Blue Latex gloves for any forensic examinations 😀

I had a sooty skid mark the other evening but it did not flash over fortunately !

Thanks for the heads up. When mine does go pop, will using a dehydrator to charge the units work instead?

Does that element have resistance that changes in an odd way (rising at the edges faster) due to it's construction, so that there is excessive power dissipation around the edges at turn on,making it overheat and degrade faster round the edges? Maybe one switched on under a thermal camera would reveal something like that.

At first I thought it was wasteful to have a switcher supply just to run the fan, but it homologates the design so power voltage and frequency don’t matter. The PTC should be self regulating. The LED will be dimmer in countries with 120 volt power, but should still light. It seems like the LED should have been on the output of the switcher supply so brightness would be the same in all countries. I wonder if the PTC is less apt to carbonize I. Countries with 120 volt mains? There would be less E field across the edge and less likelihood of pulling debris across the gap.

I tried so many of these cartridge based solutions and i keep coming back to using organza bags and just regenerating my silica gel in the oven. I would love to find something more convenient, but they all seem to either not be very durable, don't regenerate properly or just cost way to much... I think i will try to 3D print my own next, maybe some high temp filament that won't melt. Any ideas? :)

It looked to me where the corner plastic locators are, caused the issue? perhaps shortening/removing the plastic stand ups may suppress the arcing?

Could moisture from the desiccant have caused the initial tracking?

I wonder what the reverse voltage limit of the led is? I don't think it's anywhere near the peak mains voltage. Should have put a diode in the opposite polarity across the led to limit it. Better still a retro flickering neon.

Clive, could you sand the sides of that heater chip to make it good again?

Conformal coating around the edges maybe?

I love touching the cap ( no surprises ) yep last resort test😄

More spares for the drawer !

If it’s desiccant you want to dry out and use again just stick it in the oven on a low setting and bake for about ten minutes or until it’s all dry. Source: my mother

I actually saw a packet of 6 AA Ni-MH cells for replacing Garden solar lights for only€2.50 i a shop here in Ireland call home-saver. Not sure if they are real but im going to by a pack today to find out... I think that's a good price for 6 of those cells... Ill certainly put them to use and finally get to use my Ni-Cd/Ni-MH charger i own but never had to use, "EVER" 😊👍🇮🇪 I'LL let you all know if they fake or if they actually the real deal. 👍👍👍

Why did they make them square instead of round? The sharp edges are flashing over.

my guess is....they need to get rid off old gen parts that are not being bought at all taking space in warehouse...maybe ;)

Always good.👍 👍 👍 👍

To continue using the cartridge. Rip the guts out, cut a hole in the base, and bolt a cheap hairdryer to the bottom. Afterwards put legs on the dryer for air clearance and switch on.

Thanks :)

Product dev 101 - Design things to fail!