Related Tutorials
No related posts.
AmplifiersInterfaceNon-Isolated DC/DCVoltage ReferencesAudioLinear Regulator (LDO)Power ManagementWebenchBattery ManagementLogicPower Management IC (PMIC)TI DesignsClock and TimingMOSFET and IGBT Gate DriversPower ModulesTI StoreData ConvertersMotor DriversSensorsMyTI Registration
In an earlier blog post I did a teardown analysis of the XL4015 DC-DC Buck Converter module. I recently received an improved version of the same module. I purchased it from an online seller in China. I tried it immediately in a new experiment.
This time the module has onboard constant voltage (CV) and constant current (CC) control feature which is useful for general purpose power supply and battery charger applications. This is the quick specification chart of the module posted by the seller.
This is the photograph of my XL4015 Step-Down DC Module with CV/CC Control.
As you can see, apart from the input and output screw terminal connectors, there are two multiturn trimpots to adjust the output voltage and output current. Further, there are three LED indicators – the first one near the input connector is the constant-current (cc) indicator, while next two LEDs are intended mainly for battery charging applications (battery charging and battery full indicators). However, when used as a general-purpose power supply, the battery charging indicator LED will work as a ‘load-on’ annunciator.
Slowly turning the voltage adjustment (V-ADJ) trimpot clockwise will raise the output voltage gradually, and a counterclockwise rotation will lower it. Likewise, clockwise turning of the current adjustment (I-ADJ) trimpot increases the current limit while counterclockwise action decreases the current limit. It’s advised to adjust the voltage level at first, and then the current level (this fine-tuning procedure will be discussed in detail later).
What follows is the thorough inspection of the module. I included this session to share information about the background electronics I think might be useful to people. Let the teardown begin!
Needless to say, at the heart of the module is a XL4015 chip from XLSEMI (www.xlsemi.com) which is a 180 KHz fixed frequency PWM step-down (buck) DC/DC converter, capable of driving a 5A load with high efficiency, low ripple and excellent line and load regulation. According to its datasheet (Rev 1.5) the PWM control circuit can adjust the duty ratio linearly from 0 to 100%. An over current protection function is built inside so that when a short circuit occurs, the operation frequency will be reduced to 48KHz. Below you can see the function block diagram of XL4015 (XL4015E1).
The core of the design merely follows the typical application example (see below). An exception is that the recommended 1uF bypass capacitor (C1) is not used in this module, provision is already provided in the PCB, though.
The rest of the electronics include one 78L05 fixed voltage regulator, one TL431 precision shunt regulator, and a LM358 operational amplifier. I haven’t seen the 78L05 chip in my module at first as it’s situated underneath the flat toroid-core inductor.
As you might guess correctly, the three ‘surplus’ chips are included in this revised XL4015 module for current control function. If you turned over the module, then you find a 0.05Ω (R050) current sensing resistor at the bottom of the PCB.
Next is the adopted schematic of the XL4015 CV/CC module. This is a very basic schematic as there is a myriad of befuddling replications and revisions – No surprise, simply a common Chinese business practice!
The overall design of the module is indeed a pretty neat and clever piece of work. Frankly, I was quite impressed by the unperplexed design idea of the LM358 based current control segment as this edition also includes a TL431 shunt regulator. In a previous version, I found that there was no TL431 chip onboard but the same regulated 5VDC (outputted by one LM317 chip) is used for the reference voltage as well (see next schematic snip). In principle, TL431/TS431 provides a reference that is more stable than the 78L05!
Anyhow it’s pointless to use this module unless you know how to run it properly. Let me show the (believed to be right) way suggested by my trusty overseas vendor. Just presume that you want to get 1A current at 9VDC output. Power up the module and connect a digital multimeter (dialed to proper dc voltage range) across the output connector, and adjust the V-ADJ trimpot to get 9.0VDC readout. Then, put the multimeter in proper current range, and adjust the I-ADJ trimpot to get 1.0A readout. Finally, remove the digital multimeter, wire the load, and run it. It’s worthy to notice that tuning of one trimpot might affect the setting of the other one slightly – not a big issue, let it be!
For the quick test, I used the module to drive one 10W white LED, through a 2A/12VDC lab power supply. Cool and simple!
Everything has a tolerance. See the oscillogram taken while my DSO was probed at Pin 3 of XL4015. The switching frequency (Fosc) is about 190kHz. According to the datasheet, this lies in 144-216kHz range (180kHz typical). The output ripple is 18mV. Not bad at all!
Concerning the TTL shutdown feature of XL4015, logic level shutdown function can be used in typical system application with a small signal diode or standard LED. When the TTL high voltage is above 3.3V(referenced to ground, lower than VIN), the converter will shut down, and when the TTL Low voltage below 0.8V(referenced to ground) will turn on the converter. As might be expected this feature is exploited smartly in the XL4015 non-isolated constant voltage and current module covered here.
During my experiments I accidentally came across another similar (5A) XL4015 CV/CC module design with a slightly modified circuitry. See the two highlighted resistors in the below schematic.
To conclude, without the user manual and exact schematic of the module we can only guess. However, I’m rather pleased with this module. It’s probably going to be best suited to hobby projects that demand a constant voltage or constant current where a bit of drift isn’t going to case troubles. Online storefronts are awash with these minuscule modules costing peanuts, and when they play as promised they’re incredibly useful to have around. I am a little concerned with the impressive claim that this module can also be used as a lithium-ion battery charger as I’m afraid if the input voltage is cut, the connected battery at the output discharges back through the electronics and might destroy it.
I tested the module under various input/output voltages and load conditions and evaluated its efficiencies. In the end, I opted to include the evaluation report in an upcoming blog post. Yes, I already tried one of these modules in a small solar-powered beacon light project. I haven’t had any problems so far. Now I’m documenting that do-it-yourself project, so you can see the article here within a couple of weeks.
One annoying thing is that many sellers fallaciously described these XL4015 CV/CC buck converter modules as SEPIC converter modules. Look, in a single ended primary inductance converter (SEPIC), the output voltage can be higher or lower than the input voltage. Actually, a SEPIC converter module is a buck-boost dc-dc converter module based on SEPIC design topology which is a type of dc-dc conversion method allowing the electrical output potential (voltage) at the converter’s output to be greater than, less than, or equal to that at its input voltage. Below you can see the photograph of a quite popular (and cheap) SEPIC converter module.
…
Related Web Links
…
Looking for the latest from TI?
AmplifiersNon-Isolated DC/DCAudioPower ManagementBattery ManagementPower Management IC (PMIC)Clock and TimingPower ModulesData ConvertersSensorsInterfaceVoltage ReferencesLinear Regulator (LDO)WebenchLogicTI DesignsMOSFET and IGBT Gate DriversTI StoreMotor DriversMyTI Registration
@Amir: I hope the note posted by dickb_100 in response to your question will be helpful to you.
RE cap values
The large 470uf caps are marked all other small caps can be 0.1 uf or 0.22 uf ceramic X7R they are not critical. The cap between pins 1-2 of the LM358 is I believe 0.047 uf BUT it’s too small to prevent oscillation when in CI mode. I’ve been working to prevent oscillation at low frequencies in CI mode it’s difficult given the existing layout, but if your doing a new layout please email me as I have some changes to propose.
If your goal is to prevent oscillations in CI mode and do don’t care about very low current operation then do this. increase cap from LM358 pin 1-2 to 0.22uF AND add parallel 2.2 meg resistor ( piggback on exsisting cap location. AND add series rsistor from I limit pot wiper to pin 2 of LM358 of 20 K. This will prevent oscillation however it has a side effect of raising the minimum value that can be set for CI mode operation.
Additional unrelated comments
The values shown allow nominal output voltage to be ( in theory ) set as high as
46 V which is not possible since output voltage needs to be less than input and max input voltage is 36 V. This has effect of making output voltage very difficult to set via the 10 K pot. easiest way to correct this is to change value of 270 ohm resistor higher. Again email if you need more info.
If this module is mounted in an enclosure max continuous current is much much less than 5A indeed I’d start to worry about overheating above about 1. 0 A.
If your using the module to charge batteries I recommend a fuse in output lead as the module will be destroyed by connecting to a battery with reverse polarity.
hi.I am Amir from Iran
I want to build this module from a diagram block،But the values of the capacitors are not shown on the circuit diagram
Where can I get the amount of capacitors?
Thank you very much for your guidance
I continue to make progress on the current limit circuit. My focus is to change or modify as little as possible to continue to use this board WITHOUT having to make an auxiliary PCB for the current limiter circuit. ( you’ll see my earlier comment I thought that was going to be a necessity ) I now have stable performance with both resistive and LED loads in current limit mode from fullscale current down to less than 3 ma. Here is the amazing part current regulation is near perfect or as good as my instruments can measure. Changing from 12 ohm load to 24 ohm or to LED module results in less than 0.2ma change in output current at 544 ma setting!!!
That’s far better than Voltage regulation when in voltage mode. Good thing the modules are very low cost as I’ve managed to destroy several with my attempts to change parts
@dickb_: Thanks for the come back, and for sharing your thoughts again.
I already noticed the “powered” mounting holes. I think it’s intentional as you can see certain modules with a DVM PCB at its top as a second layer, powered through 4 metal studs.
As you pointed right, its difficult to work with those nasty LF oscillation both physically and electrically. LM358 is not a very good candidate there.
Yes, I’m still in the loop, and following you to get updates on your experiments. Good luck!
I wanted to add to my earlier comments regarding this module and it’s ability to drive hi power leds.
1) the 4 mounting holes in corners of board contain the 4 different signals IN+ , IN-,
OUT + and OUT -. Do not use metal hardware to mount this module to a metal
panel !
2) The current limit circuit will oscillate at low frequency and result in flicker in your LED lights.
3) because the module uses 0603 SM components, and has very thin traces and pads its difficult to work on.
4) I have managed to control the oscillation in the current limit circuit by lowering the gain in first section of LM358 op-amp..BUT …it requires several component changes and results in a unfortunately high lower limit of LED current.
5) Stay tuned I’m working on a small aux PCB circuit that does not oscillate.
Will allow setting I limit current to zero, and uses an External panel mount pot to vary led current ( brightness ) from zero to full current with no flicker. I’m expecting to use the 4 corner mounting holes to piggyback this new I limit circuit and conduct those signal to the new I limit circuit.
6) I’ve also lowered the 180 KHZ output ripple to about 1/2 what it was in original circuit. This is not needed to drive LEDS but it may be interesting to people using this as a general purpose LAB type supply with panel mounted pots for voltage and current
First thank you for the schematics and tear down, it was very helpful.
I wanted to offer a few suggestions for people who are using these modules to drive LED modules.
If your driving LED modules what is important is the current limit circuit. But the existing circuit suffers from drift an poor setabality of the current limit. ( because of the low millivolt levels at the inputs to LM358 op-amps )
My suggestion replace the 0.05 ohm current sense resistor with a higher value. Depending on what current you want to use in LEDS
here are a few examples 0.100 ohm will limit max current to 2.7 A
0.150 ohm will limit max current to 1.8 A
0. 200 ohm will limit max current to 1.35 A
0.250 ohm will limit max current to 1.08 A
You should pic the highest resistor value consitant with the current for your LEDS.
If you want to be able to continuously vary the LED light level consider removing the 10K Iset pot and replacing with wires to a front panel type pot.
Another idea is if you have a front panel pot and you want to limit the max current that can be set at max CW rotation of the pot then use a front panel pot of 20K or 50K value but install the 10 K screwdriver adjust pot ( as 2 wire resistor ) in shunt with outside terminals of your front panel pot. Then using some sort of dummy load set front panel pot at max CW rotation and adjust the 10K screwdriver pot to set the max possible current. Result is you’ll be able to smoothly vary led brightness from very low levels to max. But its unlikely to work smoothly and without flicker if you don’t replace the 0.05 resistor with a higher value per my comments earlier.
@dickb_100yahoo-com: My pleasure! Thank you very much for your very insightful feedback and point of view. I appreciate your time and effort to comment 👍
Guys, have you looked with an oscilloscope at the output when the current limiting is triggered? My board gave me a big surprise.
@babilon78: Thank you for the screenshots!
Without current limiting: https://ibb.co/q0NR3xn
When current limiting is active:https://ibb.co/ZcFf3cJ
@babilon78: I’d like to see that ‘surprise’. Hope you’ll share your screen captures here. Thanks!
Hi TK.
Did you read my email I sent with details of the Dummy load. You never replied .
If you are still interested I would like a little feed back. on your thoughts Thanks.
I also modded this board it works really well . really cheap boards .well worth the few $$’s .
@mcrubish: Thanks for your patience. Yes, I got it & replied. Hope now you can see that, perhaps in your spam box. The project is very interesting, and I’d like to see more updates. Keep up the good work. All the best!
Hi the first link I gave was the full translation of the PDF so just read the english.
save messing just scroll down from the first link page . the download has the diagrams etc .
If you want I can give you the last prototype of my dummy load .
just give me were to send it .or upload it
@mcrubishL Noted, Thanks! I’m curious about your dummy load, so you may please email the snap(s) to – designlab.technode @ gmail.com