Thread Tools
This thread is privately moderated by mnemennth, who may elect to delete unwanted replies.
Jan 18, 2015, 09:24 PM
AMA 1033652
mnemennth's Avatar
Thread OP



We've all been there... Working all day on a new build, you finally finish, you give your new beast one final look-over and don't see anything wrong. So you plug it in and TZZZT! SNAPP! POP! and instantly, your plans for new-found aerial bliss disappear literally in a puff of smoke.

You tear it down and find a blown ESC caused my a motor with a too-long screw into a winding or a little blob of solder that you SHOULD have caught before you put the new shrink-wrap on, but DIDN'T.

Now you have to tear it apart and replace the toasted ESC before you can even see if the motor is salvageable. You already know you're going to be ordering a new one shortly; the only difference is whether you're going have to wait for it to come in the post or if you ordered enough to have a spare and will need to replace it.


Nothing exciting here; quite the opposite, actually. We want testing your new build to be as UN-exciting as possible, after all.

This is a little gadget that electronics hobbyists have been using for nearly a century; an incandescent light bulb as a current-limiting device to provide power to a device under test.

The wattage of the light bulb is chosen such that if there's a dead short in the device under test, instead of allowing massive current to flow and letting out the Magic Smoke, the bulb will glow and only the amount of current needed to illuminate the bulb will flow. This only needs to be enough to allow the device to initialize; not enough for it to operate at full power.

This can prevent fatal current surges from ever hitting many electronic components; saving the part and allowing you to get back to troubleshooting the original failure instead of having to replace burnt-out parts and THEN troubleshoot.

In our case, since we're dealing with a small quad, we need 2-3A with a 3S pack to be able to power up, arm, and start the motors with props on at just above idle. As luck would have it, a common everyday 1157/3157/4157/3057/4057/3357 automotive taillight bulb will flow approx 2A through the heavy filament, and another 450-500 mA through the lighter filament. If we connect them in parallel, we get a Current-Limiter in the 2.5A range.

The important part is to use a dual-filament bulb rated 25-27W and 5W.

DIY - You can build this $4 device that will keep you from ever smoking your quadcopter again!!! (3 min 7 sec)

CAUTION: Use a little common sense!

This video shows me sticking my hands in running props on a 250 quad as a test exercise. Remember this is a TOY-size quad with plastic props. Even with the torque-limiting provided by this device, you can probably still get hurt by larger quads and Carbon or Wood props.

If you build a SmokeStopper™ with larger wattage or multiple bulbs for higher current to be able to slow-spin up larger quads, it will DEFINITELY allow enough power to flow through your motors that you can get hurt doing this!!!


The concept is simple: Make this device; build it with the power connectors YOU use (Mine will use the super popular XT-60s) to make it so convenient you won't NOT use it. Then, EVERY TIME you work on anything where you move or work on anything electrical/electronic, you plug this in between your battery and the aircraft or vehicle to test for shorts.

If you have an old 3S battery that you just use for bench-testing, leave the SmokeStopper™ plugged into it so you don't even have to think about using it. Or... make a long set of leads to power from your 12V Workbench Power Supply; you'll NEVER be without it! Just remember not to use higher than 3S/15V power for testing; you don't want to blow the bulb if you do have a short.

A 4S pack at approx 50% "STORAGE CHARGE" appears to be safe; I've been using one that way for several months without any significant difference in usage from a full 3S pack. I HAVE blown the bulb using a freshly-charged 4S pack, so I recommend against it. You can see why here:

This device has undergone some evolution since I first released it; the advent of Micro and Nano quads has resulted in a similar evolution in smaller, more sensitive electronics being used for all model aircraft.

Nowadays it isn't safe to do initial testing with the High-Current tester shown here, though you should STILL have one for testing with all accessories before plugging the battery in direct.

The ONLY TIME you should be using the HIGH-CURRENT(25-27W filament in circuit) Tester is AFTER testing with the LOW-CURRENT Tester; this will protect your smaller electronics as well as your ESCs.

The LOW-CURRENT version uses ONLY the "Dim" 5W filament; this is enough to power up the FC, and initialize the ESCs, but will sometimes glow if your ESCs have a very loud "Beacon" or initializing tone.

It may also glow if you have a high-power VTX and/or LED Bling lighting. If it does glow or cause a reboot loop (Boot, Smokestopper™ glows, shutdown, reboot), unplug the Bling lighting first and retest, then the VTX if necessary.

The HIGH-CURRENT version is really only safe to use AFTER first testing with the LOW-CURRENT version; it will allow you to arm and bring up throttle to just off idle even with props on, which as you can see in my video, is plenty to do basic response testing on the bench.

There are two ways to make this work. As the light bulbs usually come in a pack of two, for simplicity's sake I usually recommend making two testers - one LOW-CURRENT version that uses just the 5W filament, and another that has BOTH filaments in parallel for the HIGH-CURRENT tester.

This is the EASIEST way to ensure you're not testing with the HIGH-CURRENT version when you should be testing with the LOW-CURRENT version.

Alternately, you can make one tester with a single bulb and a SPST switch. Wire it so that the 5W filament is ALWAYS connected, and the switch ADDS the 25-27W filament for HIGH-CURRENT mode.

Obviously, you want to make sure your switch is CLEARLY labeled so that you can tell whether you are set to HIGH-CURRENT or LOW-CURRENT mode at a glance.

Here, have a picture (and slightly less than a thousand words ):


If you are going to use a wedge-base taillight socket with pigtails, you have to figure out the wiring on your particular socket. There are essentially two ways to wire one of these sockets, but the color-coding on the wires is different from one socket to the next so you need to trace it out with a continuity tester/multimeter. Sometimes you can easily see by looking which contacts are connected together for the common (This is the one we use for XT60 BATT IN + on the previous diagram) wire as sheen in the photo above; sometimes it is not so easy to tell.

The most common wiring on US vehicles will be the "Standard" wiring shown above; however, as many listings on fleaBay are by China-direct vendors, there's no way of knowing which you'll get for sure until it arrives.


The 5W filament provides JUST enough current to power the quad at idle and allow the FC/RX to initialize; in most cases enough to initialize the ESCS and arm as well unless the ESCs have a really powerful initialize/arming tone. If this is the case, it will manifest visually with bright flashes as the ESCs sing their little songs.

This is PERFECT for first testing, as the maximum possible current flow is still low enough that even if there is oh, say... a dead short in the RX, it will NOT allow enough current to flow to burn intermediary wiring or to damage the FC.

It is ALSO the best to use as a current-limiter for flashing your ESCs. This is a VERY real concern WHENEVER flashing any version of BLHeli, as the flashing process can cause both HI and LO side FETs to turn on at the same time, creating a dead short that ONLY appears when power is applied. Flashing the wrong firmware to an ESC can also cause this.

I've tried dozens of common bulbs to find the optimum current capacity; for most any QUADCOPTER or PLANE, 2.5A is a good safe range for initial spin-up testing; that is why I finally settled on the 1157/3157/4157.

Remember, the high-current version of this device needs to be sized to the aircraft. This one works well with small 250-sized quads running 1806 sized to 2208 sized motors. It will slow-spin up a single 3548-950KV with a 14 inch prop on my 72" Piper Cub. I've tested with my 2212-1050 equipped 450-size FPV Spider; it will arm and just spin the motors above idle nicely.

Even if you have a larger craft that can't slow-spin up on the current it provides, the single 1157/3157/4157 SmokeStopper™ can STILL save you from smoking stuff if there's a random dead short present.

This is a device that you NEED on your bench, no matter WHAT you fly! Or DRIVE, for that MATTER!

If you try to use this on a brushed nano or micro-quad, it will still probably flow enough current to blow transistors off the board if there's a short. Look for a 5-7 watt bulb for this application, or use only the smaller filament from a 1157/3157/4157.


Here's what we need: an 1157/3157/4157/3057/4057/3357 taillight bulb, 3 leftover ESC wires, a pair of the connectors YOU use all the time, some Red & Black heat shrink tubing to match (+) & (-) on the connectors and Epoxy to pot the wires onto the bulb once everything is soldered. The big heat-shrink is optional if you pot the wires properly. (FYI - any of the above bulbs bulb will also work; the 3157 is just what I had handy.

The important part is to use a dual-filament bulb rated 25-27W and 5W.

You can get the 1157/3157/4157 bulb anywhere people use cars. ANY auto parts store or major department store like WalMart will have them. If you don't want to cart yourself out to the store in person, you can still find them on eBay, but they'll cost more due to shipping.

This search on eBay will omit MOST of the LED lighting clutter that has overtaken fleaBay:

This search omits more of the LED listings, but returns fewer results.

Both searches include pigtail sockets you can use along with.

Similar search for 4157 bulbs:

These bulbs are all over fleaBay... there's no need to search for other bulbs as substitutes.

The first step I'm going to assume you already know how to do; making a main power input for a quad. On the XT-60s, it helps keep the connectors properly aligned (the plastic housing tends to get soft during soldering) if you plug the female connector into the end of the male connector before soldering. Let the soldered poles cool completely before moving on to the other pole and again before unplugging the connector.

Here I'm tinning the (+) pole of the female connector; note that I've kept the male connector and the wires on the other end to act as a heat sink; again, to keep the pins aligned properly so they plug/unplug easily.

I originally made this using thin 20ga wires from a small ESC; I found those were not strong enough to stand up to tugging on the XT-60 to pull it apart. This version, I'm making with 16ga ESC wires.

Don't forget to put the heat-shrink tubing on the black wire BEFORE you solder it to the (-) pole on the female connector. Note that I'm placing the wires so they overlap; this it to help keep the Red wires oriented correctly.

Now a little clear heat-shrink to keep the red wires in a "Y" configuration...

...and pull the Red wires up behind the loop of the black wire to keep things tidy.

Use a thin blade to lift BOTH contact wires on one side out of their slot; it doesn't matter which side you start on.

Now pull them both out straight...

...and bend them over the housing to the other side, next to the contact wires already there.

Now, tin both pairs of contact wires together like this.

Don't forget to place the big heat-shrink tubing BEFORE you solder the Red wires!

And here it is, all soldered up. I've found that the contact wires here can be a bit weak to hold up against handling, so I'm going to pot the wires with Epoxy.

You want to use plenty of Epoxy... need to completely encase the wire to provide strain-relief or they're going to break in no time. If you don't mind the look, you can omit the Heat-Shrink; I just like the look of the finished product better.

I prefer to use Epoxy for this purpose over Hot-Glue; light bulbs ARE a heat source. You can probably get away with Hot Glue if you're just using this to protect your gear at first power-up; it should never be on long enough to get hot. However, if you intend to use it for motor testing or as a discharge load (As I intend to do), you should pot with Epoxy so it doesn't matter if it runs lit up for extended periods.


Well, there you have it. Make the Smokestopper™ and use it with all your aircraft and even cars and boats - no model you have should ever suffer from Magic Smoke Syndrome; at least on the bench. Out in the air, out on the track, or out on the water... well, there's only so much we can do with preventive maintenance.

The point of the SmokeStopper™ is that it allows you to power up everything and detect faults that aren't easily detected with a multimeter. FETs can turn on while flashing, regulators can act as a short if reverse polarized, reversed capacitors will usually heat up and make the bulb glow first rather than just blowing off the PDB.

If it glows bright and stays on, you have a dead short or something (like high-power Bling lighting or big wattage VTX) that is drawing massive current.

You troubleshoot by disconnecting B+ to subsystems in this order: Bling lighting, VTX, ESCs, FC/RX one at a time until the light goes out. If you have reason to suspect a device, like a particular ESC, of course try disconnecting that first. It's pretty straightforward.

If you can plug in power through the LOW-CURRENT version and it flashes a few times as the ESCs sing their startup songs but then settles down to a dull glow or goes out altogether, you're safe to try the HIGH-CURRENT version for spinup testing to just above idle. If that behaves similarly, you're safe to go with the battery connected direct.

*Puts on best Soup Nazi voice*


Every time a SmokeStopper™ lights up, a Model Aviator gets his wings!
Last edited by mnemennth; Jul 04, 2019 at 10:34 PM.
Sign up now
to remove ads between posts
Jan 18, 2015, 09:25 PM
AMA 1033652
mnemennth's Avatar
Thread OP

Recently it has come to my attention that HobbyKing has finally gotten around to copying the SmokeStopper™... only really NOT. Not even CLOSE to a copy; but rather a complete misunderstanding of the concept that will eventually prove to NOT protect as the SmokeStopper™ does.

All they did was put XT60s on a common inline fuseholder like this one then call it "Something like a SmokeStopper™ but not that".

There is a reason I chose NOT to use a fuse... fuses are SLOW. By the time that fuse blows in the case of a shorted output, or two FETs on at the same time, damaging current has ALREADY FLOWED for several milliseconds; more than enough to short a FET or make a leaky junction that will fail later. More than enough for a short in a VTX power circuit or FC to damage a regulator or the processor itself.

Their poor demonstration has NO WAY of showing that action as it simply happens TOO FAST. I don't know anybody who can SEE something that occurs in a few milliseconds.

Not EVERY ESC has a reverse-polarity shunt diode in it like the ESC they chose for this demonstration; and not every piece of small electronics does either.

I considered making this very device and rejected it years ago, knowing that a current-limiting resistor was the right way to go.

The lamp in the SmokeStopper™ is a current-limiting resistor... it PREVENTS damaging current from EVER Flowing in the first place. It is proven, tested and known to work.

No, it isn't an acceptable substitute. The video they made demonstrating the product is, as one might expect, only showing it protecting against ONE thing... not all the things the SmokeStopper™ protects against. And even their demonstration rig will NOT protect with every ESC, RX and FC out there.

I promise you, time will tell and unless HK redesigns with some form of current-limiting resistor, their ill-conceived copy will prove NOT to protect as the SmokeStopper™ does.

Do yourself a favor, and do as has been discussed in the thread many times over the last couple years... make a Low-Current (500mA using only the dim filament) SmokeStopper™ for first testing and flashing ESCs.

Make a SECOND High-Current version (~2.5A with both filaments) for testing with spinning up motors when you need to.

Bottom line is you'll be money ahead just in fuses not wasted... not to mention electronics the copy doesn't save, especially when you run out of fuses and decide to take a chance.

I really would have preferred if they made a direct copy of the SmokeStopper™... at least that would be a honest clone that actually does what it is supposed to do, instead of this product which is an outright lie in every respect.

110% home-grown Tech, from somebody who did it just for the good of the hobby. Accept no substitutes.
Last edited by mnemennth; Nov 28, 2016 at 04:14 PM.
Jan 18, 2015, 09:26 PM
AMA 1033652
mnemennth's Avatar
Thread OP
UPDATE: "Solid-State Smoke Stopper Inhibitor" is still NOT an authorized product, nor is the version RaceDayQuads and TBS/Team Black Sheep are selling using my SmokeStopper™ product name.

The plot sickens...

While Ben Eishen of Bengineering Labs has complied with my request that he not use my SmokeStopper™ trademark in association with his similar product, it appears he has sold that product, the "Solid-State Smoke Inhibitor" in whole or in part to RaceDay Quads, and they are still using my Trademark in their advertisements. As a result, I have been forced to contact RDQ regarding their use of the SmokeStopper™ name; see below the letter sent via their "Contact us" portal:

Originally Posted by mnemennth

RDQ Support -

Hello there; my name is Paul Krajewski. I am the originator of the SmokeStopper™. Your similar product seen here: has been recently brought to my attention; and as such, I find I must communicate with you to protect my IP rights.

While I appreciate and applaud the desire to distribute the SmokeStopper™ as widely as possible, I must remind you that the name "SmokeStopper™" is a trademark belonging to my non-profit project seen here:

I have released this device under Creative Commons Licensing that other may benefit from it, but I retain all rights to the SmokeStopper™ Trademark and require that you link back to my original work when using my Trademark.

While this version of the device is clearly different from mine for purposes of manufacturing IP, your version is MUCH too high-current to be safe for initial testing as the SmokeStopper™ is intended to be used and I do not authorize use of my Trademark in association with this product which I consider to be unsafe for the intended purpose of the SmokeStopper™.

I have spoken with Ben Eishen of, and after explaining the non-profit nature of the project, and that I didn't feel I could endorse his product as it does not provide adequate protection for initial testing of a device in the way the SmokeStopper™ does, he agreed to change the name used in his advertising and remove the name from his silkscreen in future production runs.

A little history here: I spent quite some time experimenting with PPTC polyswitches back when I still had only the high-current version; I could not find a value that would properly initialize 4 ESCs at once while still limiting inrush current to 400-500mA as the Low-Current version of the SmokeStopper™ does.

This is why I went with the HI-LOW current design I updated with instead of a PPTC polyswitches. They have a steeper response slope than a filament on a 1157, etc light bulb; they ultimately simulate open circuit similarly to a fuse. However, they are SLOW; and they can permit much higher inrush current than the low-current filament of the specified light bulb and for much longer.

This is why I consider them to be unsafe for use as the advertised purpose of the SmokeStopper™.

That said - I am satisfied with Ben's renaming of his product to "Solid-State Smoke Inhibitor"; while it is close enough in semantic value to the original SmokeStopper™ trademark to still be a bit cringe-worthy, I would happily accept the term "Solid-State Smoke Inhibitor" used in your advertisement of your product and consider it to not be infringing.

If you have any questions regarding this communication, please call me. I'm pretty reasonable.

Thank you for your cooperation,

Paul K******* - AKA mnemennth on Tel: ***.***.****

I'll update here as things progress,


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~

History of this issue:

I have come to an understanding with Ben Eishen of that I hope we can both keep in good faith for the good of the hobby.

This afternoon I received a call from Ben Eishen of, and after explaining the non-profit nature of the project, and that I didn't feel I could endorse his product as it does not provide adequate protection for initial testing of a device in the way the SmokeStopper™ does, he agreed to change the name used in his advertising and remove the name from his silkscreen in future production runs.

A little history here: I spent quite some time experimenting with PPTC polyswitches back when I still had only the high-current version; I could not find a value that would properly initialize 4 ESCs at once while still limiting inrush current to 400-500mA as the Low-Current version of the SmokeStopper™ does. This is why I went with the HI-LOW current design I updated with instead of a PPTC polyswitch. They have a steeper response slope than a filament on a 1157, etc light bulb; they ultimately open circuit similarly to a fuse. However, they are SLOW; and they can permit much higher inrush current than the bulb and for longer.

Furthermore, that action pretty much eliminates one of the best features of the SmokeStopper™; use as a diagnostic tool for locating the source of the short if it DOES light up.

I explained to him these reasons why, and he seemed to understand, but he also stated he's had a fair demand for the product from word-of-mouth. I cannot argue with that, but I did suggest strenuously that for the good of the hobby he make sure to explain that his product can allow much higher currents to flow than my version (even more than the High-Current SmokeStopper™) and we've agreed to leave it at that.

It does get difficult to do the right thing from time to time.

Last edited by mnemennth; Jun 23, 2020 at 09:12 PM.
Jan 19, 2015, 01:17 AM
Registered User

It'd be worthwhile seeing if HK would pick up the idea and make them in a few sizes including, of course, a small royalty to you.
Jan 19, 2015, 01:47 AM
Registered User
Thanks for this! really great idea!
Jan 19, 2015, 02:59 AM
Registered User
Where the **** was this 2 days ago. Would have saved a brand new Naze32 from going up in smoke

I couldnt find the globe you used here but this one looks like its similar. 47534

Oops forgot the heatshrink and used hot glue but might save me from another Brian moment
Jan 19, 2015, 06:58 AM
Registered User

I started building mine using a socket (1157 really since that's what i have) for ESC programming. After reading this, I think I'll use a on-on switch for low high of single low filament and both filaments.

Also this clarified to use bulb in line on positive. Instinctively on my first wiring up; I wired it for lighting the bulb.. Doh..

Last edited by ver2go; Jan 19, 2015 at 07:09 AM.
Jan 19, 2015, 01:16 PM
Waste Nothing
RickC_RCAV8R's Avatar
This just sounds too logical ! Any tools for aiding troubleshooting are good . Gotta try this one . Thanx .
Jan 19, 2015, 03:07 PM
AMA 1033652
mnemennth's Avatar
Thread OP
Thanks everybody for the positive feedback!

Matt -

I'd love to see them made and sold by a third party; I wont hold my breath on the royalty, though. My interest, as with my other popular projects, is to help other RCers spend their time having fun by doing it right the first time instead of fixing blown-up stuff or rebuilding.

If I could see one of these in every RCer's toolkit, I'd be ecstatic... means much fewer smoked ESCs. Dunno if HobbyKing would want to get into THAT business.

ingleburn -

Looks good to me! "Pretty is as Pretty does!" is my motto.

21W/5W is in the right neighborhood. It MIGHT be a little too low current to actually get the motors to spin up; especially if you have a lot of accessories. Of course, it's gonna be easy to find out - trial and error is how you figure out exactly the right light bulb anyways!

ver2go -

I recommend making one for flashing, and one for bench-testing. You want to make the one for bench-testing with battery plugs and only one setting, to make it so convenient you're never tempted to not use it.

For flashing and testing single ESCs, I still recommend the separate different-colored alligators; you can tell at a glance that you're using the right leg because of the color. That way you never have a moment of "Sonuvacrap, I had the switch in the wrong position" and your "helper" becomes a liability.

Really; these things are so cheap and easy to build that there's no reason NOT to have one for every type of aircraft you fly. It should be part of your everyday kit, like a voltmeter.

Let there be light bulbs!
Jan 19, 2015, 04:23 PM
Registered User
Good thing the sockets and bulbs come in 2 packs. Question about flashing when everything is wired up to PDB and motors. I'd power it through the power connector on the PDB would you recommend using the .5ma filament first (ie with with alligator clips to PDB) or just use the 2.5a with the battery connectors? Logic says try with low then move up as needed. But curious to your opinion.
Jan 19, 2015, 06:47 PM
AMA 1033652
mnemennth's Avatar
Thread OP
Caution is always the better part of valor.

Yeah; try the lower current filament first. I'm gonna see if I can flash the BLHeli FW to my Ice Blue 12s in the quad; I'll let you know how that works up.

Since I started using normal (3S/1400s instead of 3S/4000s and 2S/1800s) batteries on my Flying Brick, I find it is just crazy floaty with the 6045s, especially in wind. I want the active braking BAD.

Active breaking... that's how I fly. Maybe active braking will help.
Jan 19, 2015, 08:43 PM
The king of mediocre flyin'!
rikybob's Avatar

Jan 20, 2015, 02:55 PM
Fly 3D Mon!
3DMON's Avatar
Very handy! Thanks for sharing.
Jan 20, 2015, 07:17 PM
AMA 1033652
mnemennth's Avatar
Thread OP
Yeah; I know, right?

I shoulda done this years ago. No idea why it took this long for the light blub to go off in my head! Err, on... Umm... you know what I mean.

Teh caek iz a lie.
Jan 20, 2015, 08:41 PM
The king of mediocre flyin'!
rikybob's Avatar
Funny story. I saw this and haven't had time to build it yet.

Got up this AM and was rushing out to the club.

Was going to quick put an XT60 in place of a Dean's.

Well ya know what happens next, POOF!

Phreakin' eeediot!

If only I'd had this contraption....


Quick Reply
Thread Tools