just turn the potentiometer adjustment screw 20 turns counterclockwise then you good to go.. datasheet - www.olimex.com/Products/Breadboarding/BB-PWR-3608/resources/MT3608.pdf
Datasheet quite clearly states that although the OUTPUT voltage can be up to 28v boost, the FB(feedback) pin can only support 6v as an absolute maximum. What does that mean? It means you MUST power the module from LESS then 6v, and then adjust your output boost voltage. That is how you set them up. Powering from MORE then 6v during adjustment(as in this video - 10v) WILL kill the device once you turn the trimmer so that the potential on the FB pin rises above 6v. This will cause an internal breakdown of the IC, and subsequent catastrophic failure of the device. Basically.....user error if you cook them. Remember also, that any buck or boost regulator is generally fatally unstable if the input voltage AND output voltage are the same. So, for a boost converter, the input voltage MUST be LESS then the wanted output - ideally by at least a couple of volts to be safe. A buck-converter MUST have an input voltage that is HIGHER then the wanted output - like a standard Linear regulator.
I burned the first board the same way you did... thought that was odd... did some googling and saw your video and seems that the comments below are correct. Turn the pot 20x counter clockwise and start with low voltage... raise gradually and trim the pot and it works.
These are NOT faulty. But they do have a slight flaw in the design. I've sold thousands of these on my ebay shop and recommend to all my customers to set them up with 5V. People assume that turning the pot clockwise will turn the voltage up, because, since when did you ever turn anything to the left to turn it down?! Set up on 5V and turn pot anti-clockwise. Really, you should edit or remove this video as it's completely misleading.
For those who may be watching. I assumed by the info from the purchase that they supported up to 24v in. However through trial and error I have found it to be a max of 6v in
The annoying thing about these units is that they are supplied with the voltage set to minimum. This means if you connect more than 6v to the input they go up in smoke as the Vref pin apparently can't take more than this voltage. To avoid mishaps start with 3v or 5v then turn the pot ANTICLOCKWISE until you get the volts you want. I have only ever used these with 3v or 5v input usually to provide 9 or 12v and never had any problems since then. Their failure mode is Vin = Vout and the adjustment no longer works (as you have popped the Vref pin) which is exactly what you found. You were definitely popping them by applying 10v to them straight out of the packet then you were turning the pot in the wrong direction which moves the Vref towards the supply voltage and pop!
BEWARE! You don't necessarily turn the pot screw anticlockwise. BTW, WTF are you thinking writing anticlockwise when the proper term is counterclockwise? Anyway my point is, there are different versions of this board, and some, like a few I have with a USB-C input port on them (otherwise the schematic/function is the same) are different and you need to turn the pot CLOCKWISE to increase the voltage. However, the failure mode isn't exactly Vin= Vout, rather it is Vin => 6V, with the pot adjusted nearly all the way to the top of its travel to minimum voltage, which is not a lot different than what you stated, but the specifics might matter. This is only true of the generic modules that don't have any additional pull down on the pot wiper. There is a fix some people have found, by putting a jumper between the pot wiper and the resistor pull down to ground that can keep them from burning up unless the input voltage is higher than spec. Some measurements might be in order, to determine whether the modules you receive do have clockwise vs counterclockwise voltage increase, since that makes a BIG difference above 6.0V input. Even so, don't doubt yourself, a lot of these generic modules are defective out of the package which is why it is a good idea to order more than 5... 5, because they often come in strips of 5, so if you order more than 5, you usually get specimens from at least two different strips, decreasing the chances that some manufacturer defect propagated itself onto both strips' modules.
Hi guys. If you wanna make this work check first variable pot resistor to be set on minimum 3.3kOhms and connect on input 3,7v li-ion batery. If voltage raise above 10volts you are fine. Do not forgot voltage between input and output must be higher than 3V.
I blew up a few of these too. But before you call an item defective you should know which way the pot needs to be turned, not just "turn it the other way". Also since it is a step-up device you don't start testing at 10V I imagine? and you can bridge the +- output with your fingers? When you re buying them for 0.5 USD a piece you might as well figure out how they work... they're sensitive devices but work great! And god bless China for selling them to all of us at this stupid price! They handle 10W without blinking!
mendebil - I also like these modules and bought a few for different purposes. I killed one too, but managed to repair it. Pretty small chips but doable. (See my other comment for a link to my video about the repair).
Ya, GOD bless china for taking away millions of American jobs but I admit that I am forced to buy the garbage they make because the American companies can make a few more pennies per item. Wonder how long before they start making our children ? lol
You will kill them by applying more than 6v to them out of the packet. I only blew one up before i realised this and now i input at most 5v and they all seem to be very happy. The reason for this is that they come set at minimum and the Vref pin can only handle 6v.
@@vidasvv Nope, American greed and postage rate disparity is the problem. People want to sit on their butts and pull in big money when they are doing nothing valuable. It's not the hard working american labor pool but rather the adminstrative, executive positions, the environmental regulations and so on pushing costs through the roof. The result is, you can get these boards under 25 cents each in bulk but if a US company were making them, then they'd want at least 4X as much plus it would cost more in shipping cost than they are worth. 2nd problem is our trade agreement that subsidizes postage such that you can get 10 of these delivered from China, for cheaper TOTAL cost, than it would cost you or I to mail the EMPTY envelope they came in, to our next door neighbors.
Maybe the reason is already mentioned in all these comments. But I guess the reason for blowing up these modules is that the wiper pin of the 100 k pot is connected to the feedback pin of the MT3608. It is possible, that the wiper pin is loosing contact just for short durations. Then the feedback pin is pulled to ground with the 2,2 k resistor and the MT3608 increases the output voltage to more than 28 V. This destroys the mosfet, what actually results in a short cirquit and overheating of the IC.
If the wiper pin is connected to the FB pin, then losing contact would mean it is floating, not pulled to ground, or if the contact is very poor then resultant higher resistance to ground.
You shouldn't put more than 6V when setting the converter. The modules that you have I think have a design issue where the trimmer is used as a pot instead of a variable resistor. They've used a 100k pot in series with a 2.2k resistor which makes it very difficult to set the output when the wiper of the pot is connected to FB pin of the IC. You would have to keep turning the pot close to one of the ends of the pot before you would see any changes. Once set properly, you can change the input voltage up to 24V as long as the set output voltage is higher than the voltage at the input.
Hey, thanks for the comment! as you will see on the blue meter in the background, the input voltage is at 10v and 6v at some points, but yes, these modules are not ideally designed!
Why the mt3608 caught fire? By reading the datasheet, it is stated that it has internal overload thermal protection, so should it turn off itself before catching fire? Or the best case to prevent the pcb catching fire is to put an inline fuse at VIN ?
Input and output ceramic capacitors of 22μF are recommended for MT3608 applications. For better voltage filtering, ceramic capacitors with low ESR are recommended. X5R and X7R types are suitable because of their wider voltage and temperature ranges.
yes these are faulty, be wary when using them. i have a 9v battery run through these to output 12v, works out fine but once unplugging and plugging it for multiple times it just started to spark and burn. another faulty one was when the battery ran out and was recharged, i attached the booster that was already set and working it caught on fire.i broke 5 of these too faulty.
I might seem stupid but i would ask what's your power supply,linear or switching? i've run into similar issues when i was supplying it using a switching power supply,the voltage seemed like it can't be adjusted. Apparently using a battery it worked like a charm! I can imagine using a capacitor in parallel on both input and output can fix many problems,mostly because any switching frequency on its input will mess up its own one,in which case it will be filtered out.
I've used dozens of these exact same modules and had no issues whatsoever with a single one. That said, they are very susceptible to blowing up if the output is shorted (even on very low o/p voltages). I found this out during destruction testing. Non have ever failed in service.
Les McQueen - When they fail it is most often the internal mosfet that shorting out. Although the modules are cheap enough to just replace, they can be repaired. In a recent video (see other comment) I show how to repair a failing module.
I looked at the internal circuit of the IC and I think because of the external pot being set too high, being a multi 30 turn pot before applying power you need to turn the pot clockwise till it clicks - clockwise lowers the voltage output and anty clockwise raises the output - I think What happens when the output is set to high, when power is applied the IC internal shunt resistor can not support the input surge and hence the smoke of death. The output voltage once set and turned off, Then when restarting needs to power the output via a stand by switch once the module is on then the output stand by switch can be switched On to apply the load. Also, note that whatever input voltage you apply will be the lowest output - so 2V in = Lowest 2V out - Highest = 27V. from a 2V input. -- So a 6V in = Lowest 6V out. - Highest 27V From a 6V input.
0) Get a fresh Module. ( Do NOT power it ) 1) Do twenty counter clockwise turns on the pot ( This lowers the boost ) 2) Connect 2 AA batteries (3v) on the input ( Start with a low voltage ) 3) Now check the output. It should be higher than 3v. If it isn't, do 20 more counter clockwise turns on the pot.
0.5) Check that the potentiometer is wired up correctly. Two of the pins need to be bridged, or the voltage divider the chip uses to set the output just doesn't work.
NO! These pots only have about 25 turns total, so if you do 20 turns, there is no reason to do as many as 20 more. You also have it backwards, you do not want to lower the boost. That is what causes the problems. You want to raise the boost above the input and then back off the pot a half dozen turns from that. Also, certain designs of the MT based boards, are turned clockwise to boost instead of counterclockwise. Best bet is measure with a multimeter to see what it's doing before applying power, and of course, give it under 6V input till adjusted, then try higher input voltage.
@@sazonswaggreviews4449 It will do that, if your 3V is constant, not a pair of 1.5V/1.2V batteries that drain too low during use, and of course, if your 3.nX voltage boost, does not exceed the current capacity of around 2A. That's 2A worth of boost. For example, if you have 3V in, and want 9V, 700mA out, that's 2.1A so check the temperature as it is probably too much to ask of this module.
stepping up voltage doesnt do any damage to the module,but when it comes to step down voltage,shits happen. Because when you try step down voltage this module setup trying to hold some of energy in itself in order to limit voltage, that s why it gets hot and circuit gets burn. I also burnt this exact module,while trying to step-down 12v -1.5amp tablet charger. it didn't burn immediately ,but after continuous usage, it get heated up and burnt after while. AND ALSO YOU HAVE TO TURN IT COUNTER-CLOCKWISE (FOR THE FIRST TIME) BEFORE USING
And counterclockwise is higher voltage, and you should never turn fully right. Because the PWM goes to 0%, witch means the MOSFET shorts the inductor output to ground.
I've played a bit with these & had them die when trying to go below the input voltage. There is another model that integrates a micro-USB connnector for 5vdc input. I've had them powered by 4v up to 11.6v (12.6v 3S with 2 diodes). They work great for boosting but also appear to be able to output a lower voltage than the input, although I only tested 11.6v input down to 9v output. They are also marked MT3608 but are shorter & have wires visible on the inductors.
I got the same version of this DC/DC converter as you and it works just fine. Also I think the layout is better on the version you're talking about (the small FB resistor is much closer to the IC and the output capacitor is closer to the output pins). In my converter the output capacitor was dead shortcircuited so I replaced it by a 1uF capacitor with the same package (1210) and now works just fine. This little converter is perfect to run fans or stuff that need more voltage from a USB port.
Hi guys, could you please suggest some similar module, which will be able to invert 12v to 19.5-20 and give 2-4A - need this to charge the laptop from 12v battery and immiting UPS 220v inventor (which consumes the same as my laptop). Thanks
You can use XL6019 or XL6009 boost converter for under 4A power. Or you can use (05KJ) 150W 10A boost converter, that's the most common boost module people use to charge laptops.
one Buck convert after another buck convert can cause problems. Edit/ I used couple of those modules in past years and none of them did catch on fire, I use them to often boost batteries 4.5v to 12v or higher.
Of the 5 units I have ordered, only 1 worked. 2 units burned and 2 had faulty pots, in which, output voltage varied from 3v to 6.6v depending on the input voltage. For the 1 unit which worked, I have set the output to 12v which remained constant no matter what input voltage I put. Used an assortment of input sources including a bare li-ion batt at 3.6v, powerbank at 5v 2A, a 5.1v 1A iphone charger, 5.1v 2A ipad charger and an 11.1v 3s li-ion battery pack. All these inputs had a constant 12v output.
For your initial input voltage, only set it to 5v.....then adjust the pot to a higher output by turning anti-clockwise. These units are often set at 9v or lower, so if you input 12v it cannot handle it and will burn out. Current limiting on the input supply could also save them. After the initial use and set, they can be set to the voltage needed and they will be fine. Put simply, if the input is higher than the set output, they burn out...cannot handle it. Believe me, i burned several before figuring it out.
You are prematurely writing the MT3608's obituary! Let me start with some background. How these DC to DC converters work is as follows. Basically the input is used to charge up the inductor. The inductor is the put in series with the input voltage and supplied as the output voltage. Since voltages add in series the output voltage is higher than the input voltage. This switching is done by the IC, in our case the MT3608.This is the DC-DC Boost converter. You also get a DC-DC Buck converter where the inductor is charged by the input and only the inductor, but only the inductor is then applied to the output, giving you a lower voltage. This happens at very rapid switching (in our case 1.2MHz) so that you get a relatively reliable output. How the voltage is set is with the use of a feedback resistor, in our case the 10 turn pot. This feedback is used by the IC to charge the inductor to the correct voltage. You cannot use a Boost converter as a Buck converter, since the input voltage are inductor must be in a different configuration. You do get modules that can do both Boost and Buck, but they will usually have two different inductors. It can also be done with MOSFETs, but an extra inductor is probably more economical. Why the modules burn out is that they are not Buck-Boost modules, but only boost modules, therefore the output must be greater or equal to the input voltage. Not quite sure why you were able to set them to a lower voltage than the input, but they probably did not add a safety in the event that someone tries to lower the voltage. I am not sure what you meant by your last statement that all your modules were defective , since I saw at least one of the modules working perfectly and you even powered an LED strip with it. So I would like to know based off what do you declare you modules faulty. Also note that in order to reach the rated values of the modules you might have to add heat sinks. Since operation might not require one, they might not provide it, since it is extra cost. Please also beware that you will not always be able to run the modules at the rated values. They do often oversell the operational conditions! Also for future reference, the Chinese mentality differs greatly from the traditional Western mindset. If you are going to order cheap items from China I would just like to keep the following in mind. If you order a larger quantity of items from Chin, they do not check that all the stock is in perfectly working order. Quality assurance is expensive ad they will rather want to provide you with a cheaper product. I have never had large batches of defective components, but be aware that you might receive defective components.
Hey, that's some good information, the one that did work, i couldn't adjust, so none of the boost converters were usable as they were advertised at. I did however get another batch and they did work better, but had an extra link on the pot in the back, I'll have to make a revisit video.
I've had several of these modules, the reason they burn up is if you go too low with the pot which operates counter to what you think, so turning the pot clockwise reduces the voltage output. These modules also use a 25 turn pot, at no stage did the op turn the pot more than 5 turns in any direction (before swapping direction). They are temperamental but do work.
That's not an explanation for why they burnt out. Turning the pot down does (or should do) nothing bad, the IC will simply switch off the internal MOSFET. I suspect the power source in this video (also a switcher) had an inductive output impedance, so the MT3608 module needed a decent bypass capacitor at its input.
Hi, are you using them now ?..I have no problems with them but they are a bit twitchy...so I use the chip itself on projects and take the inductor and the diode of the modules and it still works out as a very cheap boost solution in a small form factor.
Turn pot counterclockwise until voltage goes up. I had same problem getting voltage increase going clockwise many turns but then turned back counterclockwise many times until I actually got 5.0v out from around 4.2V in (got 5V out using one 18650 easily 2.5V - 4.2V in) . Also, as I purchased these to use with two 18650s for input voltage 6.0 to 8.4 to give 12V had no problem getting 12. I went back to 6V and actually got up to 27.7V max with no heating problem and ran 6v in at 18v out also with no heating problem. I didn't have the burn out problem going clockwise but didn't push the limit with high voltage. I have no idea why you did but I'm going going to test this procedure. I believe these should perform well. Note that as voltage in becomes greater than voltage out setting, the voltage out will follow the voltage in but be just a little lower. Thus set out for 5V and then put 6V in will get around 5.8V out. I tried two units and thought they didn't work and then looked up datasheet and found that NEED TO TURN POT COUNTERCLOCKWISE TO GET VOLTAGE INCREASE. Google is your friend.
I have 5 of these on order.. based on other videos i have seen.. i believe that these can handle up to 10W on the output. In your vid you probably want to restrict the current on your little PSU. Also the tech spec for the MT3608 shows that the max you can get out of it is 800mA at 12V with 5V input. The best efficiency of it is at 200-300mA at 5V and 12V out. Its not really designed to power high V or drain devices. There is no cooling on the chip at all.. You can drive a 3X3W LED at 300mA at 12V in a parallel configuration. This would be max that this chip would handle.. ie 12V at 850-900mA, even then i would stick a tiny 6mm by 6mm aluminium heatsing on it.
While you are correct in the sense that the IC designers didn't think it was a good idea to exceed 800mA, it is not necessarily safe to assume that ICs can't operate outside of specs on a datasheet graph as that was never spec'd as a max. You just assumed it and yet it is possibly true since there is evidence that these heat up before 2A. However IDK where you are getting the 3x3W LED at 300mA from. 3x3W LED would have approx 9.9Vf and 900mA, and this chip can do that, or at 300mA you'd be talking about a little under 1W each LED unless you mean they are special triple die LEDs but those tend to have the dies in series. Did you mean 3 x 3W triple die LEDs with each having a forward voltage of near 10V, so the module is pushing about 300mA at 30Vf sum of all leds in series? If so, yeah, that's about right for a 12V input to this boost driver.
The doc said, turn counter clockwise 20~30 turns before power it up!! During factory test, many set to top multiply. Thus, for higher input voltage, the output setting many be too much.
I blew up 7 of the 10 bought .My application was to go from from 12V to 24V.What I found is that the board offers no filtering to stop any non-resistive feedback .The unit run a resistive load indefinately with ease , however the moment I placed it on the non-resistive load they lasted a few seconds.Filtering may be the only means to reduce any feedback , as these boards are absolute bare bones.
Maybe there has been a revision but I recently have bought 10 and small no resistive loads have been fine, I also shorted it through current clamp to see what would happen and it easily held over 2.5 amp peak, aka temporary max current. She was noisy as hell on the high current load but other then that it’s been great as a small power unit for things like multimeters and data cable testers I was sick of buying batteries for.
These modules are shipped with the pot adjusted for low output. Before using for the first time, turn pot anticlockwise a good number of turns, and apply just 5 or 6 volts to input. Set output voltage, then modules should be ok to go with higher input voltage, provided input is always less than output. Dont exceed 2 amps on input.
I smoked one too. For me, once I got my magnifier out I could see that it was caused by the crocodile clips of my multimeter. The one connected to VOUT+ was a little too far north and broke through the laquer causing a short. On the others I have, I went straight to soldering in's and out's and no issues since. Hope this helps.
lol, yeah I was in disbelieve that he thought using those alligator clips was a reasonable way to hook that up, and after the very first one blew, the last thing I would have done is kept using alligator clips instead of soldering wires on. Where in the instructions does it state to slap alligator clips on the edge of a PCB? That is never the right way to connect anything. It was so absurd that he did it, that it made me skeptical of the entire rest of the video as I'm thinking "if he can screw up something that simple, I can't put anything past him".
I do not know what happen to my mt3608. So far I burn out 3 of them. Are they faulty product ? Even they are very cheap from Aliexpress few years ago but I do expect they work.
*TL;DR* These are not faulty. As mentioned by @VauxhallViva1975, here is how to use them: *1.* Even though it can handle a higher input once you have got the voltage output set, if your V in is greater than 6V, you WILL fry it when you try to adjust it. Only Adjust the potentiometer if V in is 6V or less. *2.* Turn potentiometer screw anti-clockwise to increase voltage. Clockwise will decrease the voltage. *3.* V in MUST be less than V out.
Except they are faulty - they were built without a link on the back of the potentiometer between the right and center pin. I have a bunch of these now with the link and they work perfectly
This SX1308 (b6286 MT3608) chip's a few problems. First, when the chip fails, it shorts Vin and ground. Second, for some boards, you need to unplug the load before switching on Vin. If not, Vout will always be Vin.
I have used quite a few of them and they work pretty fine as long as you stay within the load limits and not operate in its extremes. The multi turn pot starts to show changes only after more than 5-6 turns. I used it for 12volt input and 24 volt out put for a current of 350ma on load side about 8.5 watt and they work perfectly.
Yes . I have 4 like that but a bigger one . Same problem as you have exactly . I was have to do a reverse engineering my model board . My first problem was the diod was short out and damage the ic My second was the resistance 10k was not work so it have these problem . In order to make them work again I have to re change all part on them except the conductor and VR
...I have those but it was a step down..there is another module with coils with description 1 dcv to 3.3dcv/ 3.3dcv to 5 dcv..for Peltier thermoelectric generator booster.. trying to lit a 5mm LED just by body heat..
did it already.added 82k resistor to 22k trimpot.this is for the version where the 3 rd leg of the precision trimpot IS connected and the voltage variation only occurs over about 20k. in the second variation the 3rd led of precision trimpot is not connected to anything and voltage variation occurs more evenly over a 100k normal trimpot
I got a toy that operates on 3V. I wanna attach a board with a bargraph and a single LED that is powered by 9V. Do you think this buck would suffice ??
If you're drawing 1.5 amps in the input, you're drawing a lot less on the output because your voltage is higher. 1.5*5 is 7.5 watts /16, your leds are drawing about .47A
Hi, Have you tried to reverse engineer to see if the circuit is as per the data sheet, the schottky diode is okay? the pot is the correct value or even connected on all three leads? The IC in the right way.. Please use a switch to switch it ON, not a clip lead.
The best you can do here is to remove the potentiometer and solder a fixed resistor considering the value of R2 which is from 2 to 3 kohm, and calculating with the formula V= 0.6*(R1/R2) as said in the datasheet. This chip is not designed for modifying the value of resistors while it's working.
I have the same problem, i am using it to power 12v leds from 3.7v 18650 battery and its flickering (seems like module is trying to shutdown because of overheating), however i was able to adjust output voltage.
Hmm interesting i have bought the exact same ones one of them the same chip went up in flames due to an accidental short circuit and I've had another go up in smoke for no obvious reason but the others (so far) are working just fine and as intended
I am guessing people have already explained why this is happening. I got my modules for roughly the same price and they are alright. By turning the pot clockwise, you decrease the boost until it is 1:1 (pass-through), if you do it sufficiently, you short the MOSFET and effectively this starts burning. By turning it counter-clockwise, you increase the boost. If I am not wrong, you can hear a very faint clicking from the pot as you keep turning it clockwise--- that's because it had already reached the maximum turns and was just jumping the wiper pin over the resistivity strip
Without resistive padding on the output side of the pot, when the wiper reaches that end it puts output voltage directly on the Feedback pin. If your input is greater than 6v and its not regulatung, ie at startup, that instantly kills the chip. Poor design for the sake of a protection/limiting resistor! Doesn't help the fact the pots are cheap and nasty either.
I had problems and they wouldn't power my arduino nano', so I put a 220uf 35v capacitor across the output and they work. Watching the movie I think you were adjusting in the wrong direction. I think if you wind it to far in the wrong direction it looses the thread inside the pot and it won't wind back in the opposite direction. Turn counter clockwise to increase voltage I think.
One on the output works pretty well, but we have been putting a second capacitor on the input aswell just like on most other voltage regulator boards. Just use electrolytic capacitors 220uf 35v connected across the + - at the ends of the board.
Put a capacitor on the output (at least 10µF, respecting the polarity if any). It burns because the current raised into the inductor doesnt have a path to ground when the IC opens its switch, and so goes through the IC and melts it. I agree on the fact that this trick should have been told by the seller, but on the other hand please document yourself about the circuits you are using before connecting them and watch them burning, or if you want to live dangerously put a fuse or something ! As far as I can see, the IC is so small i dont expect the output power to be much higher than a watt.
alexandre meyer Mine after 30 counterclockwise turns makes clicking sound once every next turn. Connected a dozen in two years successfully but this one wont boost from 3.7V to 5V. Why is that?
I bought 5 of these (ebay seller from China - exactly the same layout but different branding) the first developed a hard short across the input after a few minutes, the others would start "crowbaring" when connected to any load, and would not reach a steady output voltage if the load was connected before powering up the device. I found that if I added a series resistor (1/2W 220Ω) to the output they would produce a steady voltage, but soon the resistor began to dissipate a lot of heat, even though the circuit was only drawing 30mA. I gave up completely and went back to the XL6009 based device.
I tested the module by inputting a 8.5 V and as same as my module got burnt the IC within a minute . I am hopeless about this module . Bogus . I didn't take risk of the burning of IC . So I took XL 6009 module for booster and it is working fine . 👍
I don't know whether this would help or not. But, apperently there are some newer mt3608 with 330 inductor on aliexpress instead of 220 one on the older versions. I guess this might help in getting the module to last more.
@simpleelectronics I had the some problem with adjusting. But i learnt that chinese said that you must turn the potentiometer 20 times to counterclockwise. And it should be work,during turning. It will start regulate voltage after while
Have expiring the same with my 10 each until I checked the potentiometer with a ohmmeter and realized it has to go CCW for higher output (as many other write here) . I guess it was a well thought thing to let it start with the lowest output from factory. The problem with this is just it can't go lower than the input. 😕
You have to turn 20x,30x or more to see a reaction (but in the SAME DIRECTION) and not 5 times forward and 6 times backwards! Until 1A should it work normaly (without warming) ... I installed more than 25 pieces each day and I had an error rate of 5% only
Gentleman, I got similar module (I say similar because layout of components are exactly as in the video, but there is no text in the back). My modules may have different R1, R2 etc. may be, but same MT3608 chip. I believe the problem is powering up this module with a switch mode power supply. Since the module itself is switch mode. I am working in industrial automation sector for 20+ years and we always have problems with switch mode PS connected in parallel or powering each other (connected in series).The details are not important but it is the case for the %80 percent of switch mode PS. I would recommend to use 2 pcs alkaline batteries as input for the first trials. For the first trials do not use lipo, sw mode PS or re-chargeable batteries since their discharge current is very high, in case you may burn yourself and module. Before connecting batteries I checked my modules, according to MT3608 datasheet. My module R2 is 2.2 K and R1 is the blue trimpot. I set trimpot arround 30.6 K. According to the calculation in the datasheet it will give 9 V output with 3 V input. After I set the trimpot, I connected input, and without any load I measured 9 V output. When I connected load (actually JYE tech DSO150 oscilloscope with 9 V power input), there is no voltage drop it is stable at 9 V. Here is my findings: 1. Voltage stays stable even if input voltage changes but not below ~2 V as it is the lowest voltage according to datasheet. 2. With or without any load setted output voltage does not change. 3. If batteries are weak or the load current is very high to the MT3608 (check specs), when load is not connected there is no problem. For example from input you can read 2.4 Volts and from output is 9 Volts. However when you connect load, module output power drops 4-5 Volts and input still seems 2.4 Volts. Input does not drop under 2 Volts however module not functioning. You may hear a sharp sizzling sound when you connect load. This is the time you should check either your load current or your batteries. I read on some comments that there should be capacitors, actually there are capacitors on the module. The brownish components in each side of the MT3608 chip are SMD tantallum capacitors. However if you have problems with feeding from another switch mode PS (Powerbank etc.) add capacitors, this may solve your problem.
Ok guys, I burned 2 and decided to move on to 6V, but I intended to change the trimpot for a regular potentiometer (100k) but now the output stays under 5V and gets hot really fast, any one tried this? Also it consumes 1A at the input!!!
XL4015 burns down your place - it claims to have short circuit protection, but that only applies when the current gradually rises through an inductor. When the external buck converter diode fails, it creates a hard short circuit and the XLSemi part destroys itself. We were lucky to have had the board in a steel enclosure, but the place wreaked for weeks.
just for information sake, the batteries were low. i left it charging and when i came back the mt3608 started boosting the voltage. but! i had spun that pot like 10-15 turns counterclockwise as well before it started boosting. after the batteries had charged so in all honesty idk what was the actual problem but now its resolved thank you for all your help and the people who left useful comments and great vid
Okabe Rintaru hi I want to do exact same thing 18650 to tp4056 to mt3608 to get 5 volt for Arduino did you get it to work regards Dan. I was also considering parallel connections from the output of 5 X tp4056 (discharge protected version). To one single mt306 so that it would use all 5 18650s ?
Another problem in this common design is that the R1 should be fixed and near to 100K, because the RV too low exceed the FB maximum allowed input... 22uH Isn't also the best value for that, only recommended when you will connect light load, which isn't you case.
first of all, pin 4 =EN need feed small current nA and should not connect direct to V in. second then pin 3 =FB with max 6v: need constant fixed resistor, not variable adjust or can not adjust while going on operator; this will drop voltage that going generate problem. third max ic=3 amp that mean 2v x 3=6 w, then the reliable for stable running is 3w, any over could give over heat and burn ic. after all , this module is good for small loader like remoter control, led like 2 AAA battery to 3.3v, or 5v to 12v, 12v to 18v with fixed resistor; other way it not good for test or heavier load tha 7 watts.
other users who buy and try to do the voltage folding without connecting any load to the output as much as I can see from the video footage its working but some of the modules may be defective or solder faulty.I bought 5 units, renewed the first module corrupt output solder did not work again.. If I knew that they were bad productions, I would go to higher modes..I would use 18650 dual batteries for the bluetooth speaker project to upgrade 7.4 volts to 12 volts and up.. I will not get a cheaper circuit :(((
um didnt that meter thing say 2.96A, these are not rated near that. What load are you pulling from this, just a meter, you probably need a min load or bleeder resistor?
It's 2020 now and those module still suck buckets. I have 5 of mine and two had faulty pot meters (replaced them and all worked fine) and one did nothing at all.
The pots are cheap and nasty, the go opencircuit when you adjust them. The feedback pin on the IC needs to see 0.6v at the correct ouput voltage. The best thing you can do with these is remove the pot, calculate thd voltage divider that you need for your output voltage and put a fixed resistor in place of the pot. The lower of that voltage divider is on the board as a 2.2k. The problem is the the pot wiper, when delivered is as one end ... the output voltage end!!! So, any thing more than 5v in will immediately put an overvoltage (>6v) on the FB pin and burn it out! As delivered, they are crap, but can be tamed if you know the issues!!!
i had the sameee problem in all i ordered soooo i solder a 10uf cap plug a 24v 20 watt led and it started adjusting no problem they probably need some kind of load to response
you have to turn the pot really slow counter clockwise for a good while until you find that sweet spot and it will function as advertised.if you change input voltage you have find that sweet spot again.
Yes, the wonders of cheap Chinese modules. I have one that takes 3.3v and 5v and boosts it over 20v, but does not adjust to 12v which is what I bought it for.
i replaced the100k 25 turn multi turn pot with a normal 100k normal 270 degree trimpot. the 25 turn trimpot is really stupid and you need a screwdriver and there is no knob for it anywhere
Mine's didn't work either ; applying 9vdc yielded 2vdc at the output after turning the pot 25 times. The circuit seems to behaving completely differently from the MT3608 datasheet by AeroSemi (an unheard of Chinese chip company). I don't think I'll be using these of my web cam and going on vacation, Basically, it's typical Chinese junk.
Hello, this video will show you my quick look at the power supply: ruclips.net/video/Ve0msHFoyo0/видео.html just search for it on your favorite Chinese reselling site (ebay aliexpress etc...) and get the cheapest one!
Don't buy those power supplies, they are rubbish!!! Its the reason why the booster wouldn't work in the first place - the power supply he was using produces noise, which interferes with the booster causing it to oscillate and then burn out when it tries to boost the supply voltage.
Just calculated for module 2Amp maximum for input ( actually is too much) and depends how many times increase voltage with same times divide current on output from 2Amp. But keep in spare Calculate 1.7A in inlet. If want from 5V to make 10V 2 times output current 1A . If want from 5V to make 20V 4times output current -0,4-0,45A Better find other modull :)
Just had a similar module but one with an integrated micro usb connector. The thing was still in the package and bought it over a year ago. Finally found some nice use for it and before implementing I decided to adjust to the desired voltage. Mmm error, full shortage. Checked my probes, all ok. Then decided to check what got hot and limited the current in steps. At about 7amps, yup that high, the pcb felt a bit warmer but not that much. Not 1 component got hot. The only thing that would be causing this is a massive short in the pcb where the input was or the micro-usb connector. So I desoldered the micro-usb ... working as designed now. The IC itself might be good but implementation on the pcb or testing is a big no no.
Buck converters are not robust. You should expect that. You need to put a load BTW. Don't keep them idle when using even only with a multi meter connected. And you should always start them with the minimum voltage input.
Input from 2v up to 9v Out put from 3 up to 27.7v Test my model on 1amp pass 3.3v / 5v 5v / 6v 5v / 8v Test on 800ma 9v / 12v Test on 600ma 9v / 16v Test on 500ma 9v / 18v Better use input not exceed 8v And out not exceed 16v so you don't lose the power effective And yes it's better to read the data sheet for the IC model for more information as I did . It say 2amp max but it's only 1Amp And in high volt out more than 12 v your current need to be lower than 1amp for sure
So yes you can put in 12v and get 25v out BUT be sure to check your amp output! this module maxes at 2 amps but if you were to run it constantly, I would probably not output more than 1 amp, which will still be around 2 amps in!
Just turn the pot completely to the left before connecting to a supply. And while setting up only use 3V as input until your voltage is set. Remember: clockwise is down and counterclockwise is up! And never turn it to an output voltage lower than the input. Because the PWM goes to 0%, witch means the MOSFET shorts the inductor output to ground.
I think the lesson here is that, if you want to fool around with electronics, a decent bench power supply with fully adjustable current limit and output voltage is going to save you a lot of grief. I have been involved with electronics since the early 1960s and I still don't go powering something new from a lead-acid battery or a LiPo before it's been checked on a bench supply.
@@misc_channel It is very possible. In fact, it has happened to me before. How about you take a look at the ground plane of these modules and the input/output locations. Then you will see what I am talking about.
@@Mr_Wh1 Do you have one of these? Because you will see that on the other side of the pads is just a continuation of the pad. These is no coating for the crocodile clip to breach.
These units ARE actually faulty! A few people have noticed what is wrong with them in the comment section, what is your bet?
just turn the potentiometer
adjustment screw 20 turns counterclockwise then you good to go..
datasheet - www.olimex.com/Products/Breadboarding/BB-PWR-3608/resources/MT3608.pdf
Datasheet quite clearly states that although the OUTPUT voltage can be up to 28v boost, the FB(feedback) pin can only support 6v as an absolute maximum.
What does that mean? It means you MUST power the module from LESS then 6v, and then adjust your output boost voltage. That is how you set them up.
Powering from MORE then 6v during adjustment(as in this video - 10v) WILL kill the device once you turn the trimmer so that the potential on the FB pin rises above 6v.
This will cause an internal breakdown of the IC, and subsequent catastrophic failure of the device. Basically.....user error if you cook them.
Remember also, that any buck or boost regulator is generally fatally unstable if the input voltage AND output voltage are the same.
So, for a boost converter, the input voltage MUST be LESS then the wanted output - ideally by at least a couple of volts to be safe. A buck-converter MUST have an input voltage that is HIGHER then the wanted output - like a standard Linear regulator.
I burned the first board the same way you did... thought that was odd... did some googling and saw your video and seems that the comments below are correct. Turn the pot 20x counter clockwise and start with low voltage... raise gradually and trim the pot and it works.
These are NOT faulty. But they do have a slight flaw in the design. I've sold thousands of these on my ebay shop and recommend to all my customers to set them up with 5V. People assume that turning the pot clockwise will turn the voltage up, because, since when did you ever turn anything to the left to turn it down?! Set up on 5V and turn pot anti-clockwise. Really, you should edit or remove this video as it's completely misleading.
For those who may be watching. I assumed by the info from the purchase that they supported up to 24v in. However through trial and error I have found it to be a max of 6v in
The annoying thing about these units is that they are supplied with the voltage set to minimum. This means if you connect more than 6v to the input they go up in smoke as the Vref pin apparently can't take more than this voltage. To avoid mishaps start with 3v or 5v then turn the pot ANTICLOCKWISE until you get the volts you want. I have only ever used these with 3v or 5v input usually to provide 9 or 12v and never had any problems since then. Their failure mode is Vin = Vout and the adjustment no longer works (as you have popped the Vref pin) which is exactly what you found. You were definitely popping them by applying 10v to them straight out of the packet then you were turning the pot in the wrong direction which moves the Vref towards the supply voltage and pop!
El problema no es el voltaje sino el amperaje, si se le aplica más de 2A se quemará
BEWARE! You don't necessarily turn the pot screw anticlockwise. BTW, WTF are you thinking writing anticlockwise when the proper term is counterclockwise?
Anyway my point is, there are different versions of this board, and some, like a few I have with a USB-C input port on them (otherwise the schematic/function is the same) are different and you need to turn the pot CLOCKWISE to increase the voltage.
However, the failure mode isn't exactly Vin= Vout, rather it is Vin => 6V, with the pot adjusted nearly all the way to the top of its travel to minimum voltage, which is not a lot different than what you stated, but the specifics might matter. This is only true of the generic modules that don't have any additional pull down on the pot wiper. There is a fix some people have found, by putting a jumper between the pot wiper and the resistor pull down to ground that can keep them from burning up unless the input voltage is higher than spec.
Some measurements might be in order, to determine whether the modules you receive do have clockwise vs counterclockwise voltage increase, since that makes a BIG difference above 6.0V input. Even so, don't doubt yourself, a lot of these generic modules are defective out of the package which is why it is a good idea to order more than 5... 5, because they often come in strips of 5, so if you order more than 5, you usually get specimens from at least two different strips, decreasing the chances that some manufacturer defect propagated itself onto both strips' modules.
Is THAT why I've blown up so many of these that I thought were defective??? THANK YOU!!
Got a pack of 10, all 10 work just fine, once setup correctly :)
Hi guys. If you wanna make this work check first variable pot resistor to be set on minimum 3.3kOhms and connect on input 3,7v li-ion batery. If voltage raise above 10volts you are fine. Do not forgot voltage between input and output must be higher than 3V.
I blew up a few of these too. But before you call an item defective you should know which way the pot needs to be turned, not just "turn it the other way". Also since it is a step-up device you don't start testing at 10V I imagine? and you can bridge the +- output with your fingers? When you re buying them for 0.5 USD a piece you might as well figure out how they work... they're sensitive devices but work great! And god bless China for selling them to all of us at this stupid price! They handle 10W without blinking!
mendebil - I also like these modules and bought a few for different purposes. I killed one too, but managed to repair it. Pretty small chips but doable. (See my other comment for a link to my video about the repair).
Ya, GOD bless china for taking away millions of American jobs but I admit that I am forced to buy the garbage they make because the American companies can make a few more pennies per item. Wonder how long before they start making our children ? lol
@@vidasvv i dont think they are as good in making children as people from other continents 😅😅
You will kill them by applying more than 6v to them out of the packet. I only blew one up before i realised this and now i input at most 5v and they all seem to be very happy. The reason for this is that they come set at minimum and the Vref pin can only handle 6v.
@@vidasvv Nope, American greed and postage rate disparity is the problem. People want to sit on their butts and pull in big money when they are doing nothing valuable. It's not the hard working american labor pool but rather the adminstrative, executive positions, the environmental regulations and so on pushing costs through the roof.
The result is, you can get these boards under 25 cents each in bulk but if a US company were making them, then they'd want at least 4X as much plus it would cost more in shipping cost than they are worth.
2nd problem is our trade agreement that subsidizes postage such that you can get 10 of these delivered from China, for cheaper TOTAL cost, than it would cost you or I to mail the EMPTY envelope they came in, to our next door neighbors.
Maybe the reason is already mentioned in all these comments. But I guess the reason for blowing up these modules is that the wiper pin of the 100 k pot is connected to the feedback pin of the MT3608. It is possible, that the wiper pin is loosing contact just for short durations. Then the feedback pin is pulled to ground with the 2,2 k resistor and the MT3608 increases the output voltage to more than 28 V. This destroys the mosfet, what actually results in a short cirquit and overheating of the IC.
If the wiper pin is connected to the FB pin, then losing contact would mean it is floating, not pulled to ground, or if the contact is very poor then resultant higher resistance to ground.
You shouldn't put more than 6V when setting the converter. The modules that you have I think have a design issue where the trimmer is used as a pot instead of a variable resistor. They've used a 100k pot in series with a 2.2k resistor which makes it very difficult to set the output when the wiper of the pot is connected to FB pin of the IC. You would have to keep turning the pot close to one of the ends of the pot before you would see any changes. Once set properly, you can change the input voltage up to 24V as long as the set output voltage is higher than the voltage at the input.
Hey, thanks for the comment! as you will see on the blue meter in the background, the input voltage is at 10v and 6v at some points, but yes, these modules are not ideally designed!
Why the mt3608 caught fire? By reading the datasheet, it is stated that it has internal overload thermal protection, so should it turn off itself before catching fire? Or the best case to prevent the pcb catching fire is to put an inline fuse at VIN ?
it was a counterfeit defective chip - the datasheet does not apply here.
To rotate the trimmer screw, I put a piece of a tube on the tip of a screwdriver... )
if not mistaken left to increase right to decrease... it should work.. but im not sure if it work with 7v and above.
Input and output ceramic capacitors of 22μF are
recommended for MT3608 applications. For better
voltage filtering, ceramic capacitors with low ESR
are recommended. X5R and X7R types are suitable
because of their wider voltage and temperature
ranges.
yes these are faulty, be wary when using them. i have a 9v battery run through these to output 12v, works out fine but once unplugging and plugging it for multiple times it just started to spark and burn. another faulty one was when the battery ran out and was recharged, i attached the booster that was already set and working it caught on fire.i broke 5 of these too faulty.
I might seem stupid but i would ask what's your power supply,linear or switching? i've run into similar issues when i was supplying it using a switching power supply,the voltage seemed like it can't be adjusted. Apparently using a battery it worked like a charm! I can imagine using a capacitor in parallel on both input and output can fix many problems,mostly because any switching frequency on its input will mess up its own one,in which case it will be filtered out.
turns out, you need to turn the pot counter clockwise to increase the voltage.
turning it clockwise too much will destroy it for some reason.
Because the PWM goes to 0%, witch means the MOSFET shorts the inductor output to ground
Ive been working with this stuff for a while and there's no standard I know of that specifies which direction is up or down.
I've used dozens of these exact same modules and had no issues whatsoever with a single one. That said, they are very susceptible to blowing up if the output is shorted (even on very low o/p voltages). I found this out during destruction testing. Non have ever failed in service.
Les McQueen - When they fail it is most often the internal mosfet that shorting out. Although the modules are cheap enough to just replace, they can be repaired. In a recent video (see other comment) I show how to repair a failing module.
I looked at the internal circuit of the IC and I think because of the external pot being set too high, being a multi 30 turn pot before applying power you need to turn the pot clockwise till it clicks - clockwise lowers the voltage output and anty clockwise raises the output - I think What happens when the output is set to high, when power is applied the IC internal shunt resistor can not support the input surge and hence the smoke of death. The output voltage once set and turned off, Then when restarting needs to power the output via a stand by switch once the module is on then the output stand by switch can be switched On to apply the load. Also, note that whatever input voltage you apply will be the lowest output - so 2V in = Lowest 2V out - Highest = 27V. from a 2V input. -- So a 6V in = Lowest 6V out. - Highest 27V From a 6V input.
0) Get a fresh Module. ( Do NOT power it )
1) Do twenty counter clockwise turns on the pot
( This lowers the boost )
2) Connect 2 AA batteries (3v) on the input
( Start with a low voltage )
3) Now check the output. It should be higher than 3v. If it isn't, do 20 more counter clockwise turns on the pot.
Yes. it works (:
You think it works well with a 3V power supply with the module putting out 9V ??
0.5) Check that the potentiometer is wired up correctly. Two of the pins need to be bridged, or the voltage divider the chip uses to set the output just doesn't work.
NO! These pots only have about 25 turns total, so if you do 20 turns, there is no reason to do as many as 20 more. You also have it backwards, you do not want to lower the boost. That is what causes the problems.
You want to raise the boost above the input and then back off the pot a half dozen turns from that. Also, certain designs of the MT based boards, are turned clockwise to boost instead of counterclockwise. Best bet is measure with a multimeter to see what it's doing before applying power, and of course, give it under 6V input till adjusted, then try higher input voltage.
@@sazonswaggreviews4449 It will do that, if your 3V is constant, not a pair of 1.5V/1.2V batteries that drain too low during use, and of course, if your 3.nX voltage boost, does not exceed the current capacity of around 2A. That's 2A worth of boost. For example, if you have 3V in, and want 9V, 700mA out, that's 2.1A so check the temperature as it is probably too much to ask of this module.
stepping up voltage doesnt do any damage to the module,but when it comes to step down voltage,shits happen. Because when you try step down voltage this module setup trying to hold some of energy in itself in order to limit voltage, that s why it gets hot and circuit gets burn. I also burnt this exact module,while trying to step-down 12v -1.5amp tablet charger. it didn't burn immediately ,but after continuous usage, it get heated up and burnt after while. AND ALSO YOU HAVE TO TURN IT COUNTER-CLOCKWISE (FOR THE FIRST TIME) BEFORE USING
You just have to turn the potentionmetre more so that it works
I had the same problem :)
Yup bought some years ago, did the same thing. Pots are like 50 turn ones or maybe more
And counterclockwise is higher voltage, and you should never turn fully right. Because the PWM goes to 0%, witch means the MOSFET shorts the inductor output to ground.
I've played a bit with these & had them die when trying to go below the input voltage. There is another model that integrates a micro-USB connnector for 5vdc input. I've had them powered by 4v up to 11.6v (12.6v 3S with 2 diodes). They work great for boosting but also appear to be able to output a lower voltage than the input, although I only tested 11.6v input down to 9v output. They are also marked MT3608 but are shorter & have wires visible on the inductors.
I got the same version of this DC/DC converter as you and it works just fine.
Also I think the layout is better on the version you're talking about (the small FB resistor is much closer to the IC and the output capacitor is closer to the output pins).
In my converter the output capacitor was dead shortcircuited so I replaced it by a 1uF capacitor with the same package (1210) and now works just fine.
This little converter is perfect to run fans or stuff that need more voltage from a USB port.
Hi guys, could you please suggest some similar module, which will be able to invert 12v to 19.5-20 and give 2-4A - need this to charge the laptop from 12v battery and immiting UPS 220v inventor (which consumes the same as my laptop).
Thanks
You can use XL6019 or XL6009 boost converter for under 4A power. Or you can use (05KJ) 150W 10A boost converter, that's the most common boost module people use to charge laptops.
@@keith_ferdinanduz thanks a lot! I bought an adapter for car with the settings feet my laptop need! But, anyway took 09-1411 8.0A conventor
Could the input and output be on the wrong side?
Could be, but it isn't - it has the wrong ic installed
one Buck convert after another buck convert can cause problems.
Edit/ I used couple of those modules in past years and none of them did catch on fire, I use them to often boost batteries 4.5v to 12v or higher.
Of the 5 units I have ordered, only 1 worked. 2 units burned and 2 had faulty pots, in which, output voltage varied from 3v to 6.6v depending on the input voltage. For the 1 unit which worked, I have set the output to 12v which remained constant no matter what input voltage I put. Used an assortment of input sources including a bare li-ion batt at 3.6v, powerbank at 5v 2A, a 5.1v 1A iphone charger, 5.1v 2A ipad charger and an 11.1v 3s li-ion battery pack. All these inputs had a constant 12v output.
For your initial input voltage, only set it to 5v.....then adjust the pot to a higher output by turning anti-clockwise. These units are often set at 9v or lower, so if you input 12v it cannot handle it and will burn out. Current limiting on the input supply could also save them. After the initial use and set, they can be set to the voltage needed and they will be fine. Put simply, if the input is higher than the set output, they burn out...cannot handle it. Believe me, i burned several before figuring it out.
Obviously the DMM is not the correct load , put a resistor ~ 10k, bulb or inductive load , then try .
You are prematurely writing the MT3608's obituary! Let me start with some background. How these DC to DC converters work is as follows. Basically the input is used to charge up the inductor. The inductor is the put in series with the input voltage and supplied as the output voltage. Since voltages add in series the output voltage is higher than the input voltage. This switching is done by the IC, in our case the MT3608.This is the DC-DC Boost converter.
You also get a DC-DC Buck converter where the inductor is charged by the input and only the inductor, but only the inductor is then applied to the output, giving you a lower voltage. This happens at very rapid switching (in our case 1.2MHz) so that you get a relatively reliable output. How the voltage is set is with the use of a feedback resistor, in our case the 10 turn pot. This feedback is used by the IC to charge the inductor to the correct voltage.
You cannot use a Boost converter as a Buck converter, since the input voltage are inductor must be in a different configuration. You do get modules that can do both Boost and Buck, but they will usually have two different inductors. It can also be done with MOSFETs, but an extra inductor is probably more economical.
Why the modules burn out is that they are not Buck-Boost modules, but only boost modules, therefore the output must be greater or equal to the input voltage. Not quite sure why you were able to set them to a lower voltage than the input, but they probably did not add a safety in the event that someone tries to lower the voltage.
I am not sure what you meant by your last statement that all your modules were defective , since I saw at least one of the modules working perfectly and you even powered an LED strip with it.
So I would like to know based off what do you declare you modules faulty.
Also note that in order to reach the rated values of the modules you might have to add heat sinks. Since operation might not require one, they might not provide it, since it is extra cost.
Please also beware that you will not always be able to run the modules at the rated values. They do often oversell the operational conditions!
Also for future reference, the Chinese mentality differs greatly from the traditional Western mindset. If you are going to order cheap items from China I would just like to keep the following in mind.
If you order a larger quantity of items from Chin, they do not check that all the stock is in perfectly working order. Quality assurance is expensive ad they will rather want to provide you with a cheaper product.
I have never had large batches of defective components, but be aware that you might receive defective components.
Hey, that's some good information, the one that did work, i couldn't adjust, so none of the boost converters were usable as they were advertised at. I did however get another batch and they did work better, but had an extra link on the pot in the back, I'll have to make a revisit video.
if the pot had a link in the back it means that one side was shorted.so perhaps the link was missing from the pots of the first batch u got.
I've had several of these modules, the reason they burn up is if you go too low with the pot which operates counter to what you think, so turning the pot clockwise reduces the voltage output. These modules also use a 25 turn pot, at no stage did the op turn the pot more than 5 turns in any direction (before swapping direction). They are temperamental but do work.
That's not an explanation for why they burnt out. Turning the pot down does (or should do) nothing bad, the IC will simply switch off the internal MOSFET. I suspect the power source in this video (also a switcher) had an inductive output impedance, so the MT3608 module needed a decent bypass capacitor at its input.
Hi, are you using them now ?..I have no problems with them but they are a bit twitchy...so I use the chip itself on projects and take the inductor and the diode of the modules and it still works out as a very cheap boost solution in a small form factor.
Turn pot counterclockwise until voltage goes up. I had same problem getting voltage increase going clockwise many turns but then turned back counterclockwise many times until I actually got 5.0v out from around 4.2V in (got 5V out using one 18650 easily 2.5V - 4.2V in) . Also, as I purchased these to use with two 18650s for input voltage 6.0 to 8.4 to give 12V had no problem getting 12. I went back to 6V and actually got up to 27.7V max with no heating problem and ran 6v in at 18v out also with no heating problem. I didn't have the burn out problem going clockwise but didn't push the limit with high voltage. I have no idea why you did but I'm going going to test this procedure. I believe these should perform well. Note that as voltage in becomes greater than voltage out setting, the voltage out will follow the voltage in but be just a little lower. Thus set out for 5V and then put 6V in will get around 5.8V out. I tried two units and thought they didn't work and then looked up datasheet and found that NEED TO TURN POT COUNTERCLOCKWISE TO GET VOLTAGE INCREASE. Google is your friend.
you need to keep 1.5 volt difference between i/p and o/p...o/p should be higher by at least 1.5 v
Why can't I increment the voltage with a potentiometer? What are reasons could be?
Your module is broken?
I have 5 of these on order.. based on other videos i have seen.. i believe that these can handle up to 10W on the output. In your vid you probably want to restrict the current on your little PSU. Also the tech spec for the MT3608 shows that the max you can get out of it is 800mA at 12V with 5V input. The best efficiency of it is at 200-300mA at 5V and 12V out. Its not really designed to power high V or drain devices. There is no cooling on the chip at all.. You can drive a 3X3W LED at 300mA at 12V in a parallel configuration. This would be max that this chip would handle.. ie 12V at 850-900mA, even then i would stick a tiny 6mm by 6mm aluminium heatsing on it.
While you are correct in the sense that the IC designers didn't think it was a good idea to exceed 800mA, it is not necessarily safe to assume that ICs can't operate outside of specs on a datasheet graph as that was never spec'd as a max. You just assumed it and yet it is possibly true since there is evidence that these heat up before 2A.
However IDK where you are getting the 3x3W LED at 300mA from. 3x3W LED would have approx 9.9Vf and 900mA, and this chip can do that, or at 300mA you'd be talking about a little under 1W each LED unless you mean they are special triple die LEDs but those tend to have the dies in series. Did you mean 3 x 3W triple die LEDs with each having a forward voltage of near 10V, so the module is pushing about 300mA at 30Vf sum of all leds in series? If so, yeah, that's about right for a 12V input to this boost driver.
The doc said, turn counter clockwise 20~30 turns before power it up!!
During factory test, many set to top multiply. Thus, for higher input voltage, the output setting many be too much.
To increase voltage is COUNTERCLOCK wise!!!
I blew up 7 of the 10 bought .My application was to go from from 12V to 24V.What I found is that the board offers no filtering to stop any non-resistive feedback .The unit run a resistive load indefinately with ease , however the moment I placed it on the non-resistive load they lasted a few seconds.Filtering may be the only means to reduce any feedback , as these boards are absolute bare bones.
Maybe there has been a revision but I recently have bought 10 and small no resistive loads have been fine, I also shorted it through current clamp to see what would happen and it easily held over 2.5 amp peak, aka temporary max current. She was noisy as hell on the high current load but other then that it’s been great as a small power unit for things like multimeters and data cable testers I was sick of buying batteries for.
My MT3608 does nothing ... in 4,1V and out my multimeter is sending nothing. I turned the screw up and down ... still nothing. These boards are crap.
These modules are shipped with the pot adjusted for low output. Before using for the first time, turn pot anticlockwise a good number of turns, and apply just 5 or 6 volts to input. Set output voltage, then modules should be ok to go with higher input voltage, provided input is always less than output. Dont exceed 2 amps on input.
I had used some similar modules. The modules needs a load on the output in order to stop feedback voltage.
Bullshit, I've got few of them, and I can setup output voltage before connect to the load. BTW my modules is not catching fire
@@X-mass you sound like you read the F Manual.. I did too.. no smoke no fire and no bullshit!
this module has 2amp limit. ur input was about to hit 3amp wen it burns
I tested the input capacitor on mine and they short to ground.
I smoked one too. For me, once I got my magnifier out I could see that it was caused by the crocodile clips of my multimeter. The one connected to VOUT+ was a little too far north and broke through the laquer causing a short. On the others I have, I went straight to soldering in's and out's and no issues since. Hope this helps.
lol, yeah I was in disbelieve that he thought using those alligator clips was a reasonable way to hook that up, and after the very first one blew, the last thing I would have done is kept using alligator clips instead of soldering wires on.
Where in the instructions does it state to slap alligator clips on the edge of a PCB? That is never the right way to connect anything.
It was so absurd that he did it, that it made me skeptical of the entire rest of the video as I'm thinking "if he can screw up something that simple, I can't put anything past him".
I do not know what happen to my mt3608. So far I burn out 3 of them. Are they faulty product ? Even they are very cheap from Aliexpress few years ago but I do expect they work.
*TL;DR* These are not faulty. As mentioned by @VauxhallViva1975, here is how to use them:
*1.* Even though it can handle a higher input once you have got the voltage output set, if your V in is greater than 6V, you WILL fry it when you try to adjust it. Only Adjust the potentiometer if V in is 6V or less.
*2.* Turn potentiometer screw anti-clockwise to increase voltage. Clockwise will decrease the voltage.
*3.* V in MUST be less than V out.
Except they are faulty - they were built without a link on the back of the potentiometer between the right and center pin. I have a bunch of these now with the link and they work perfectly
I fried about 3 of mine before I read the above info. It solved it perfectly.
check to see if the pot has 2 legs connected together
This SX1308 (b6286 MT3608) chip's a few problems. First, when the chip fails, it shorts Vin and ground. Second, for some boards, you need to unplug the load before switching on Vin. If not, Vout will always be Vin.
I have used quite a few of them and they work pretty fine as long as you stay within the load limits and not operate in its extremes. The multi turn pot starts to show changes only after more than 5-6 turns.
I used it for 12volt input and 24 volt out put for a current of 350ma on load side about 8.5 watt and they work perfectly.
Yes . I have 4 like that but a bigger one . Same problem as you have exactly . I was have to do a reverse engineering my model board . My first problem was the diod was short out and damage the ic
My second was the resistance 10k was not work so it have these problem . In order to make them work again I have to re change all part on them except the conductor and VR
...I have those but it was a step down..there is another module with coils with description 1 dcv to 3.3dcv/ 3.3dcv to 5 dcv..for Peltier thermoelectric generator booster.. trying to lit a 5mm LED just by body heat..
Hermano debes girarlo hasta que empiece a subir por son potenciómetros de alta presision
did it already.added 82k resistor to 22k trimpot.this is for the version where the 3 rd leg of the precision trimpot IS connected and the voltage variation only occurs over about 20k. in the second variation the 3rd led of precision trimpot is not connected to anything and voltage variation occurs more evenly over a 100k normal trimpot
I got a toy that operates on 3V. I wanna attach a board with a bargraph and a single LED that is powered by 9V. Do you think this buck would suffice ??
As long as yours isn't defective it will be alright
If you're drawing 1.5 amps in the input, you're drawing a lot less on the output because your voltage is higher. 1.5*5 is 7.5 watts /16, your leds are drawing about .47A
Hi, Have you tried to reverse engineer to see if the circuit is as per the data sheet, the schottky diode is okay? the pot is the correct value or even connected on all three leads?
The IC in the right way..
Please use a switch to switch it ON, not a clip lead.
The best you can do here is to remove the potentiometer and solder a fixed resistor considering the value of R2 which is from 2 to 3 kohm, and calculating with the formula V= 0.6*(R1/R2) as said in the datasheet. This chip is not designed for modifying the value of resistors while it's working.
I have the same problem, i am using it to power 12v leds from 3.7v 18650 battery and its flickering (seems like module is trying to shutdown because of overheating), however i was able to adjust output voltage.
Hmm interesting i have bought the exact same ones one of them the same chip went up in flames due to an accidental short circuit and I've had another go up in smoke for no obvious reason but the others (so far) are working just fine and as intended
I am guessing people have already explained why this is happening. I got my modules for roughly the same price and they are alright.
By turning the pot clockwise, you decrease the boost until it is 1:1 (pass-through), if you do it sufficiently, you short the MOSFET and effectively this starts burning.
By turning it counter-clockwise, you increase the boost.
If I am not wrong, you can hear a very faint clicking from the pot as you keep turning it clockwise--- that's because it had already reached the maximum turns and was just jumping the wiper pin over the resistivity strip
There is actually one connection missing on these modules specifically
Without resistive padding on the output side of the pot, when the wiper reaches that end it puts output voltage directly on the Feedback pin. If your input is greater than 6v and its not regulatung, ie at startup, that instantly kills the chip. Poor design for the sake of a protection/limiting resistor! Doesn't help the fact the pots are cheap and nasty either.
I had problems and they wouldn't power my arduino nano', so I put a 220uf 35v capacitor across the output and they work. Watching the movie I think you were adjusting in the wrong direction. I think if you wind it to far in the wrong direction it looses the thread inside the pot and it won't wind back in the opposite direction. Turn counter clockwise to increase voltage I think.
MakkyBrown Brown did you put the cap at the positive or the negative output?cuz im gonna build a bluetooth speaker with it
One on the output works pretty well, but we have been putting a second capacitor on the input aswell just like on most other voltage regulator boards. Just use electrolytic capacitors 220uf 35v connected across the + - at the ends of the board.
the potentiometer also is confusing.. while rotating clockwise is for decreasing voltage, and counter-clockwise is for increasing voltage..
Put a capacitor on the output (at least 10µF, respecting the polarity if any). It burns because the current raised into the inductor doesnt have a path to ground when the IC opens its switch, and so goes through the IC and melts it.
I agree on the fact that this trick should have been told by the seller, but on the other hand please document yourself about the circuits you are using before connecting them and watch them burning, or if you want to live dangerously put a fuse or something !
As far as I can see, the IC is so small i dont expect the output power to be much higher than a watt.
alexandre meyer Mine after 30 counterclockwise turns makes clicking sound once every next turn. Connected a dozen in two years successfully but this one wont boost from 3.7V to 5V. Why is that?
I bought 5 of these (ebay seller from China - exactly the same layout but different branding) the first developed a hard short across the input after a few minutes, the others would start "crowbaring" when connected to any load, and would not reach a steady output voltage if the load was connected before powering up the device. I found that if I added a series resistor (1/2W 220Ω) to the output they would produce a steady voltage, but soon the resistor began to dissipate a lot of heat, even though the circuit was only drawing 30mA. I gave up completely and went back to the XL6009 based device.
I tested the module by inputting a 8.5 V and as same as my module got burnt the IC within a minute . I am hopeless about this module . Bogus . I didn't take risk of the burning of IC . So I took XL 6009 module for booster and it is working fine .
👍
I don't know whether this would help or not. But, apperently there are some newer mt3608 with 330 inductor on aliexpress instead of 220 one on the older versions. I guess this might help in getting the module to last more.
@simpleelectronics I had the some problem with adjusting. But i learnt that chinese said that you must turn the potentiometer 20 times to counterclockwise. And it should be work,during turning. It will start regulate voltage after while
Have expiring the same with my 10 each until I checked the potentiometer with a ohmmeter and realized it has to go CCW for higher output (as many other write here) . I guess it was a well thought thing to let it start with the lowest output from factory. The problem with this is just it can't go lower than the input. 😕
You have to turn 20x,30x or more to see a reaction (but in the SAME DIRECTION) and not 5 times forward and 6 times backwards!
Until 1A should it work normaly (without warming)
... I installed more than 25 pieces each day and I had an error rate of 5% only
I am giving 11.67 V to VIN of mt3608 and some spark is coming and after rotating screw it gets burn…. Anyone help here i am doing something wrong?
Gentleman,
I got similar module (I say similar because layout of components are exactly as in the video, but there is no text in the back).
My modules may have different R1, R2 etc. may be, but same MT3608 chip.
I believe the problem is powering up this module with a switch mode power supply. Since the module itself is switch mode.
I am working in industrial automation sector for 20+ years and we always have problems with switch mode PS connected in parallel or powering each other (connected in series).The details are not important but it is the case for the %80 percent of switch mode PS.
I would recommend to use 2 pcs alkaline batteries as input for the first trials. For the first trials do not use lipo, sw mode PS or re-chargeable batteries since their discharge current is very high, in case you may burn yourself and module.
Before connecting batteries I checked my modules, according to MT3608 datasheet. My module R2 is 2.2 K and R1 is the blue trimpot. I set trimpot arround 30.6 K. According to the calculation in the datasheet it will give 9 V output with 3 V input.
After I set the trimpot, I connected input, and without any load I measured 9 V output.
When I connected load (actually JYE tech DSO150 oscilloscope with 9 V power input), there is no voltage drop it is stable at 9 V.
Here is my findings:
1. Voltage stays stable even if input voltage changes but not below ~2 V as it is the lowest voltage according to datasheet.
2. With or without any load setted output voltage does not change.
3. If batteries are weak or the load current is very high to the MT3608 (check specs), when load is not connected there is no problem. For example from input you can read 2.4 Volts and from output is 9 Volts.
However when you connect load, module output power drops 4-5 Volts and input still seems 2.4 Volts. Input does not drop under 2 Volts however module not functioning. You may hear a sharp sizzling sound when you connect load.
This is the time you should check either your load current or your batteries.
I read on some comments that there should be capacitors, actually there are capacitors on the module. The brownish components in each side of the MT3608 chip are SMD tantallum capacitors.
However if you have problems with feeding from another switch mode PS (Powerbank etc.) add capacitors, this may solve your problem.
Ok guys, I burned 2 and decided to move on to 6V, but I intended to change the trimpot for a regular potentiometer (100k) but now the output stays under 5V and gets hot really fast, any one tried this? Also it consumes 1A at the input!!!
The problem is with these units is the Potentiometer, I did replace it with an external 100K potentiometer and works perfectly.
XL4015 burns down your place - it claims to have short circuit protection, but that only applies when the current gradually rises through an inductor. When the external buck converter diode fails, it creates a hard short circuit and the XLSemi part destroys itself.
We were lucky to have had the board in a steel enclosure, but the place wreaked for weeks.
i had three mt3608 and all three wont boost, i have 2 2400mah 18650 in pararalle to a tp4056 with is attatched to a mt3608, any suggestions?
Check to see if the pot is linked on two legs in the back, I've had a new batch, and those are soldered together in the back and they work. go figure.
just for information sake, the batteries were low. i left it charging and when i came back the mt3608 started boosting the voltage. but! i had spun that pot like 10-15 turns counterclockwise as well before it started boosting. after the batteries had charged so in all honesty idk what was the actual problem but now its resolved thank you for all your help and the people who left useful comments and great vid
Okabe Rintaru hi I want to do exact same thing 18650 to tp4056 to mt3608 to get 5 volt for Arduino did you get it to work regards Dan. I was also considering parallel connections from the output of 5 X tp4056 (discharge protected version). To one single mt306 so that it would use all 5 18650s ?
Another problem in this common design is that the R1 should be fixed and near to 100K, because the RV too low exceed the FB maximum allowed input... 22uH Isn't also the best value for that, only recommended when you will connect light load, which isn't you case.
first of all, pin 4 =EN need feed small current nA and should not connect direct to V in.
second then pin 3 =FB with max 6v: need constant fixed resistor, not variable adjust or can not adjust while going on operator; this will drop voltage that going generate problem.
third max ic=3 amp that mean 2v x 3=6 w, then the reliable for stable running is 3w, any over could give over heat and burn ic.
after all , this module is good for small loader like remoter control, led like 2 AAA battery to 3.3v, or 5v to 12v, 12v to 18v with fixed resistor; other way it not good for test or heavier load tha 7 watts.
you're absolutely wrong because 3A is the open switch current while the MOSFET resistance is 80 mOhm typ. and the power is only 19 mW.
I also ordered 2 and one blew up at 12VDC Even though they are rated at 2 v ~ 24 v
other users who buy and try to do the voltage folding without connecting any load to the output as much as I can see from the video footage its working but some of the modules may be defective or solder faulty.I bought 5 units, renewed the first module corrupt output solder did not work again..
If I knew that they were bad productions, I would go to higher modes..I would use 18650 dual batteries for the bluetooth speaker project to upgrade 7.4 volts to 12 volts and up..
I will not get a cheaper circuit :(((
um didnt that meter thing say 2.96A, these are not rated near that. What load are you pulling from this, just a meter, you probably need a min load or bleeder resistor?
Yeah that is because the buck supply (small blue LCD device) is current limited to 3A, the IC shorted internally and drew the buck supply's limit
Oh i just realized you were over voltaging it. Did you have anymore failures after keeping it under 6v?
@@tablatronix it has 2-24V input by datasheet.
Mine also burned when trying step up 12V to 20V.
Jan Vojíř feedback voltage is 6v max, so the pot needs to be adjusted before increasing the voltage
Same issue I used a 9.6v nimh battery at input and all 6 of my modules burned
It's 2020 now and those module still suck buckets. I have 5 of mine and two had faulty pot meters (replaced them and all worked fine) and one did nothing at all.
The pots are cheap and nasty, the go opencircuit when you adjust them. The feedback pin on the IC needs to see 0.6v at the correct ouput voltage. The best thing you can do with these is remove the pot, calculate thd voltage divider that you need for your output voltage and put a fixed resistor in place of the pot. The lower of that voltage divider is on the board as a 2.2k. The problem is the the pot wiper, when delivered is as one end ... the output voltage end!!! So, any thing more than 5v in will immediately put an overvoltage (>6v) on the FB pin and burn it out! As delivered, they are crap, but can be tamed if you know the issues!!!
i had the sameee problem in all i ordered soooo i solder a 10uf cap plug a 24v 20 watt led and it started adjusting no problem they probably need some kind of load to response
you have to turn the pot really slow counter clockwise for a good while until you find that sweet spot and it will function as advertised.if you change input voltage you have find that sweet spot again.
Yes, the wonders of cheap Chinese modules. I have one that takes 3.3v and 5v and boosts it over 20v, but does not adjust to 12v which is what I bought it for.
i replaced the100k 25 turn multi turn pot with a normal 100k normal 270 degree trimpot. the 25 turn trimpot is really stupid and you need a screwdriver and there is no knob for it anywhere
it is possible to replace the potenciometer on defective boards ? and if yes how ?
When receiving the module, before power on it,
please rotate the potentiometer with 15 laps in a clockwise direction.
Mine's didn't work either ; applying 9vdc yielded 2vdc at the output after turning the pot 25 times. The circuit seems to behaving completely differently from the MT3608 datasheet by AeroSemi (an unheard of Chinese chip company). I don't think I'll be using these of my
web cam and going on vacation, Basically, it's typical Chinese junk.
Did you turned 20 times couterclockwise on the last ? No, i saw you going only 1 - 2 turns in each direction.
what power supply you use ?
any link to buy it ?
Hello, this video will show you my quick look at the power supply: ruclips.net/video/Ve0msHFoyo0/видео.html just search for it on your favorite Chinese reselling site (ebay aliexpress etc...) and get the cheapest one!
Don't buy those power supplies, they are rubbish!!!
Its the reason why the booster wouldn't work in the first place - the power supply he was using produces noise, which interferes with the booster causing it to oscillate and then burn out when it tries to boost the supply voltage.
Just calculated for module 2Amp maximum for input ( actually is too much) and depends how many times increase voltage with same times divide current on output from 2Amp. But keep in spare Calculate 1.7A in inlet. If want from 5V to make 10V 2 times output current 1A . If want from 5V to make 20V 4times output current -0,4-0,45A Better find other modull :)
I used 3.7V Li-ion and it burns out too
Just had a similar module but one with an integrated micro usb connector. The thing was still in the package and bought it over a year ago. Finally found some nice use for it and before implementing I decided to adjust to the desired voltage. Mmm error, full shortage. Checked my probes, all ok. Then decided to check what got hot and limited the current in steps. At about 7amps, yup that high, the pcb felt a bit warmer but not that much. Not 1 component got hot. The only thing that would be causing this is a massive short in the pcb where the input was or the micro-usb connector. So I desoldered the micro-usb ... working as designed now. The IC itself might be good but implementation on the pcb or testing is a big no no.
They only rated for 1A on input too. Try the XL6009E1 as an alternative, has some fat caps on it as well.
Hey did I charge battery's like 4volt,6volt in this module????????
Buck converters are not robust. You should expect that. You need to put a load BTW. Don't keep them idle when using even only with a multi meter connected. And you should always start them with the minimum voltage input.
Input from 2v up to 9v
Out put from 3 up to 27.7v
Test my model on 1amp pass
3.3v / 5v
5v / 6v
5v / 8v
Test on 800ma
9v / 12v
Test on 600ma
9v / 16v
Test on 500ma
9v / 18v
Better use input not exceed 8v
And out not exceed 16v so you don't lose the power effective
And yes it's better to read the data sheet for the IC model for more information as I did .
It say 2amp max but it's only 1Amp
And in high volt out more than 12 v your current need to be lower than 1amp for sure
Hey i am willing to buy it and need some help i want to boost 12v to 25v can i ?
So yes you can put in 12v and get 25v out BUT be sure to check your amp output! this module maxes at 2 amps but if you were to run it constantly, I would probably not output more than 1 amp, which will still be around 2 amps in!
Yes out is less then 2amp and in is 2amp (battery). do i need anything else like capacitor or somthing ?
You should be good to go, I would order more than one though. (hint: usually 5 units cost as much as 2 unit usually)
Thanks for hint
Mine catched fire without load on 12V in. With 4V in worked.
Same for me with a XL6009E1 module.
Just turn the pot completely to the left before connecting to a supply. And while setting up only use 3V as input until your voltage is set. Remember: clockwise is down and counterclockwise is up! And never turn it to an output voltage lower than the input. Because the PWM goes to 0%, witch means the MOSFET shorts the inductor output to ground.
I have similar problems with this modules as well some of them are fake there's a version with micro usb input they are absolutely solid n work well
I think the lesson here is that, if you want to fool around with electronics, a decent bench power supply with fully adjustable current limit and output voltage is going to save you a lot of grief. I have been involved with electronics since the early 1960s and I still don't go powering something new from a lead-acid battery or a LiPo before it's been checked on a bench supply.
I also experienced the same problem, IC burned !!!!! But it is working well for 3.7 V to 12v up conversion .
I am late to the party, but I can tell you what went wrong. The teeth of your alligator clips shorted the ground plane through the blue coating.
Doesn't seem plausible.
@@misc_channel It is very possible. In fact, it has happened to me before. How about you take a look at the ground plane of these modules and the input/output locations. Then you will see what I am talking about.
@@Mr_Wh1 Do you have one of these? Because you will see that on the other side of the pads is just a continuation of the pad. These is no coating for the crocodile clip to breach.
@@misc_channel think about what you are saying. There are 4 connection points. How would they all be connected "continuously"?
@@Mr_Wh1 You misunderstand. The pads are double sided. There is no coating for the crocodile clip to breach on the opposite side.
I have a problem with over heating too.