/       /    The IC and other components of a high-power LED driver.
Support the project on Patreon


The IC and other components of a high-power LED driver.

Go to store
I've published several reviews of LEDs, so it's time to write about what you can use to power them.
This review covers three components (links and prices included), all of which are needed for one purpose: building an LED driver.

First, apologies for the cover photo. It stubbornly keeps scaling itself the wrong way, and I couldn't fix it. The seller's page has a better version.

Everyone knows that LEDs are powered by current, preferably regulated, so their brightness doesn't change when the voltage changes. That's what a driver is for: essentially, it's a current regulator.
You can limit the current with simple ICs like the LM317 or with purpose-built current regulators (there's a review of one such component on Muska), but they usually dissipate quite a lot of heat because their efficiency is low. And high efficiency is one of the main advantages of LEDs.
Switching current regulators are more interesting. They're more complex, but much more efficient, especially when the supply voltage differs significantly from the voltage across the LED.
Sure, many people will say it's easier to buy a driver like this from China and save yourself the trouble. I agree.
But it's always more satisfying to make something yourself. That's exactly what I decided to do when I ordered the driver components.
Maybe I'm reinventing the wheel. But this review covers components that are useful for many other projects too, and the information about what's being sold and what we actually get may be useful to many people.

I'll start with the IC itself. This is the PT4115, well known to LED enthusiasts. Datasheet — www.micro-bridge.com/data/CRpowtech/PT4115E.pdf
The IC has a brightness control pin. As far as I understand, the input can be controlled either with PWM or by changing the voltage. It's quite high-impedance: when I touched this pin, the LED started flickering at 100 Hz.

A lot of 10 costs $2.
After I ordered the ICs, the seller messaged me to say the package would be sent without tracking and asked if that was okay. I figured $2 wasn't enough money to worry about, so I said yes.
Some time later, I found an envelope in my mailbox.

Inside was a bag with the ICs I needed.

IC and other components for a high-power LED driver.IC and other components for a high-power LED driver.

I tested one IC using a point-to-point setup, then messaged the seller to say everything was fine, confirmed receipt, and waited for the other parts.

Next came the inductors — link
A lot of 10 costs $1.66.

These were delivered to my door (as was the next order, by the way).
They came in a cardboard box, though that seems like overkill to me.
By the way, these inductors cost significantly more here, and I didn't buy them just for this project.

The inductors themselves: 68 µH, rated for 1.6 or 1.8 A (the seller doesn't specify, so that's an estimate), dimensions 12 × 12 × 7 mm.

IC and other components for a high-power LED driver.IC and other components for a high-power LED driver.

The inductance measurement was within the margin of error.

IC and other components for a high-power LED driver.IC and other components for a high-power LED driver.

As with the first order, I confirmed receipt and left a positive review.

And finally, the Schottky diodes arrived. They're useful to have around, so I ordered a hundred.
I wanted more, but didn't want to risk it — link
A lot of 100 costs $2.26. They cost more here too.

IC and other components for a high-power LED driver.IC and other components for a high-power LED driver.

The diodes are marked SS34, but they're smaller and match SS24 diodes in both dimensions and specifications. www.onsemi.ru.com/pub_link/Collateral/SS24-D.PDF
I measured the forward voltage drop at 1 A, and it was acceptable.

IC and other components for a high-power LED driver.IC and other components for a high-power LED driver.

That wraps up this AliExpress haul.
In principle, I could have ended the review here, but buying the parts and not trying them out would have been wrong. So naturally, I decided to take the project to some kind of logical conclusion.

While I was at the market, I also picked up some 1206 SMD resistors rated at 1 Ω for the current sensor.
At first I thought I’d buy low-value resistors like the ones in the IC datasheet, but they’re considerably more expensive, and if I want to set different current levels, I’d need to buy several values. In short, that’s inconvenient, and I use 1 Ω resistors from time to time anyway.
It turned out that one of these resistors corresponds to a current of about 0.1 A, two in parallel to 0.2 A, and so on. SMD resistors and capacitors are easy to solder on top of each other, so it’s easy to dial in the required current.
I already had capacitors for the power input filter and some scrap PCB material, so nothing else was needed.

So, I set about inventing my own wheel driver. I quickly threw together a little board in Sprint Layout; the circuit was straight from the datasheet, so there was nothing to design from scratch.
I picked out a piece of PCB material large enough to make five boards at once (I’m planning to convert five halogen lights to LEDs).


The finished board came out like this. It’s larger than the ones sold by Chinese vendors, but it has a more powerful inductor and two diodes in parallel, which means lower losses and greater reliability. Size was no concern for me anyway.

IC and other components for the high-power LED driver.IC and other components for the high-power LED driver.

After that, naturally, I wanted to test it (how could I not?).
I tested it with these LEDs

Along the way, I found that the IC regulates the current properly, but even when the input voltage is increased by a factor of 1.5, the output current still changes slightly.
I suspect this may be due to a large measurement error caused by the pulsed current (I measured the output current in series with the LED).
Of course, I could have measured the current using a resistor and an oscilloscope, but I thought that was overkill: it was easy to see the transition from linear operation to current limiting, and then the transition to PWM current regulation.

The shunt value was 1/6 = 0.166 Ω.

With these input parameters, the output current was 0.7 A.

IC and other components for the high-power LED driver.IC and other components for the high-power LED driver.

With these settings, the output current was 0.65 A

IC and other components for the high-power LED driver.IC and other components for the high-power LED driver.

Before reaching the threshold voltage at which PWM regulation kicks in, I got a maximum current of—

The IC and other components of the high-power LED driver.The IC and other components of the high-power LED driver.

As I gradually increased the supply voltage, the input current initially rose smoothly. Once the circuit entered regulation mode, further increases in voltage caused the current to gradually fall, indicating that PWM regulation was working.
By the way, when I increased the supply voltage very slowly, I could see the transition: the LED brightness first increased smoothly, then suddenly dropped by about 10% at the transition. After that, it stayed the same as I continued increasing the input voltage.
Apparently, this is how the IC handles switching on PWM regulation.
At 600 mA, it barely gets warm. I don't have a non-contact thermometer, and contact measurement would introduce a large error.
I tried putting 1 A through the output. It did get warmer, of course, but not by much—and only the IC heated up. Overall, I'm pleased with it.

You might ask why I didn't just buy a ready-made one from AliExpress?
- The parts will come in handy for other projects.
- I wanted to get my hands dirty a little.
- The total cost of all the components came to about $1 per board.
- I decided to test the components rather than a finished device, since they're used for more than just drivers.
- The result is a more reliable device than the ones offered by Chinese stores.

I really hope this review will be useful.
Go to store
This page was found when searching for:
светодиодный драйвер на основе jw1969, компоненты драйвера led, jw1691 pdf, u92153, led driver olk3538bs, jw1969e схема включения, jw1130h, микросхема 3538s, купить olk3557 купить, схема включения blm9d2022-08am, olk3538s драйвер, bp2858 datasheet pdf, sr9967d pdf, sds3557bs, микросхема bp2858 даташит, bp2858 datasheet с ик управлением, datasheet u1003 купить, sds3537s datasheet, драйвер k3537b, драйвер k3537, olk1003b характеристики, sds3537bs даташит, olk3557bs схема включения, "olk3538bs", olk3103 datasheet

Questions and answers

What output current did the homemade PT4115 driver produce?

With a 0.166 Ω shunt, the author measured 0.7 A at the output. Another test configuration produced 0.65 A.

Which inductors were used in the PT4115 driver?

They have an inductance of 68 µH, measure 12×12×7 mm, and have an estimated current rating of 1.6 or 1.8 A.

Do the SS34-marked diodes match their marking?

Despite being marked SS34, their dimensions and characteristics match SS24 diodes. The author checked their voltage drop at 1 A and was satisfied with the result.

The figures from this review are in the measurement database: Other devices

Calculate manuallyAll calculators

Discounted products


You may be interested in


Discounted products


Support the project

All the content on the site and the advice are free, but we’d be grateful for your support of the project and the channel!

Discuss on the forumHave a question or some experience to share? Discuss the article with the author and readers: the forum is a convenient place to share schematics and photos.
Go to the forum

Only registered readers can comment. It takes a minute to sign up, and you can log in with the same account you use on the forum.

Log inRegister