We can often find useful or interesting things in old items. Some items can become invaluable under the right circumstances!
In this post, I am talking about an older adjustable step-up dc power supply module that most online resellers usually refer to as the “150W DC-DC Boost Converter Adjustable Power Supply Module”.
A belated trial – Why?
In a recent agro-electronics project, I needed a dc-dc boost converter to step up 12V to about 24V at a few amperes to energize a few 24V/100Ω solenoid valves (see below).
A large range of modern Chinese boost converter modules are available from various online sources. This module is centered on the ubiquitous UC3843A IC, an outdated chip, but it seems all requisite documentation and related tutorials are still available. Even though the module isn’t new, its price (and availability) is perfect for me, so I thought I’d take a chance.
Module’s electronics – Breaking apart
This is a big module, but some say its engineering is great! I am not sure what effects that might have on the circuit, we will check that in a moment. First, the schematic itself.
This is the original schematic (not drawn by me, found by googling). Note that the actual builds might differ slightly – there are some variations on my module as well.
Below is the module’s schematic I traced out. I deliberately omitted most of the component values, and I have some ‘legal’ reasons for that.
As you can see, this is a straightforward adaptation of an application design example generously provided in the datasheet of UC3843A. That means the circuit has a power MOSFET (HY1707) for switching and a variable resistor in the feedback loop to set the output voltage. The design also includes a BJT (S9013) which feeds some bias to the current sense feedback loop. As pointed out in the datasheet, this improves the stability of the circuitry at duty cycles higher than 50%.
Further, an independent 10V power rail for UC3843 is met out by a fixed-voltage linear regulator chip (78L10). This is a pretty great addition but there are a couple of things to be noted!
This is the bottom view of the module:
We can see that there is a pair of solder pads for 0Ω resistors which select whether the voltage regulator is feed from the input supply or the output supply. The first pad was occupied so that UC3843A gets its meal from the input supply. Below is a closeup of the relevant portion of the PCB showing those funny resistor pads.
First test – Bits and bobs!
I went ahead and made a quick functional test of the module. I got better insight and noticed some quirks as well. I listed some of those quirks below.
Minimum Input Voltage: The recommended minimum operating voltage for the UC3843A IC is 8.4V. Two undervoltage lockout comparators have been incorporated to guarantee that the IC is fully functional before the output stage is enabled. The positive power supply terminal (Vcc) and the reference output (Vref) are each monitored by separate comparators. Each has built−in hysteresis to prevent erratic output behavior as their respective thresholds are crossed. In short, the UC3843A is tailored for lower voltage applications having UVLO thresholds (Vcc comparator) of 8.4 V (upper) and 7.6 V (lower). The Vref comparator upper and lower thresholds are 3.6V/3.4 V. So, it will not start up until its supply voltage (Vcc) reaches 8.4V!
Although the module already has a 10V onboard regulator (9V regulator chip is used in some modules) which feeds the IC with a steady voltage, the onboard regulator itself calls for an input voltage of 12.5V minimum, and that is a bit of a problem for certain applications where the available input voltage is only 5V or 9V. Note that the typical Dropout Voltage of 78L10 is 1.7V at 25°C.
Maximum Output Power: Simply forget about the adorned 150W claim. We can safely run the module up to 20W with natural cooling and up to 50W with forced air cooling. I already measured different loads connected to the output, and can tell, that more than 50W is not possible without a beefy cooling mechanism. I will not get into the fine details, so do your own experiments to figure this out.
Now to a few words about the quick function test carried out by me!
I powered the module from a 12V/7Ah SLA battery. Initially I measured the current consumption in the idle state, and it came out about 20mA. Next, I set the output voltage to 24V thru the onboard multiturn trimpot and wired a 1A simulated load across the module’s output connection screw terminals. The simulated load is in fact a 50Ω/100W rheostat set to circa 24Ω resistance!
At that time, I managed to achieve an efficiency close to 80%. However, it’s clearly observed that above 2A the efficiency drops drastically in my case.
I did not measure the ripple because I could not find where my special scope probe (short probe with a ground spring) was hidden in the pile (I was in a hurry too, admitted)!
But see, you can measure the ripple with an oscilloscope by connecting one of its probes to the output and feed it into one of the analog channel inputs. Then change DC coupling (default) to AC coupling mode, use “Autoset” if necessary, and you can see/measure the AC ripple if it is high enough to display (https://www.powerelectronicsnews.com/power-supply- design-notes-how-to-measure-ripple).
Moreover, here are some of the most common tests for DC/DC converters (Thanks https://www.chromausa.com/):
- Input turn-on, input turn-off voltage levels and timing test
- Output line regulation
- Output load regulation
- Output Trim Settings
- Output ripple and noise voltage
- Output over-current protection
- Output over-voltage protection
- Output operating temperature and over-temperature protection
- Efficiency of the boost converter
I evaluated a classic design without too much difficulty. The UC3843A module has inbuilt current limiter, so it can be used for driving high-power LEDs after doing some design modifications. I am learning about this, so will post updates in near future.
My agro-electronics application (mentioned ab initio) happened to have a 12VDC/5A supply available, which would be perfect for powering the module just to drive a bunch of 24VDC solenoid valves. The module seemed comfortable delivering enough voltage and current for the proposed project. I know that this module has no protection features – ie, there is no protection against reverse power supply input, overload and over temperature.
Lastly, the module is still on my workbench as I need to refine its soft start circuit. I want to add a remote shutdown feature as well. I hope to do these soon. Meanwhile, any hints you can give me in how to start the fiddling will be appreciated!