Power Supply Inside Out, or the Control Board for an Electronic Load
This is the fourth article in a series about building a rather specialized device: a powerful electronic load. I’d been planning to publish it for a long time, but somehow never got around to finishing the device.
First, what is an electronic load?
A programmable electronic load is designed to act as a load when testing, setting up, and adjusting power supplies, amplifiers, and other electronic equipment.
As it happened, while I was working on this device, the board went out of stock on AliExpress, so here’s a link to a snapshot of the order page. But so readers won’t be disappointed, I found a place where it’s still available. It’s also on eBay, but the price there is not encouraging at all.
You can find this board online using the keywords M8 V6 electronic load. Note that it must be the V6, since the V7 is a completely different device by the same author.
There will also be some additional materials that may even help you try building one yourself.
While looking for the product link, I even stumbled across a mini-review of this board.
The board was packaged properly, but since it’s no longer available from the seller I bought it from, I won’t include photos of the packaging.
The board looks like a single unit with the display. The encoder, power transistor, and power and input wires are already connected.
So this is a ready-to-use solution straight out of the box: all you need to do is add a power supply and mount the transistor on a heatsink.
Board specifications:
Maximum input voltage — 50 V
Maximum load current — 5 A (15 A after modification)
Current setting resolution — 1 mA
Cutoff voltage setting resolution — 10 mV
Supply voltage — 12–16 V (limited by the input capacitor)
Current consumption with the display — about 100 mA
Fan control, but it’s very basic.
Battery capacity measurement
Low-resistance resistor measurement.
Three-wire power circuit connection (can be modified for four-wire operation).
Audible alert when the input voltage drops below the set level.
I’ll go into more detail in the review itself.
The board came with:
A PSMN020-150W MOSFET. I was even surprised that the heat-shrink tubing was color-matched to the wires.
A 20-position encoder with a built-in push button. The knob was included too.
And now, photos of the board. Some will be available for download in full resolution. If you need any other photos, let me know and I’ll provide them.
The board is about the same size as a standard 2004 display: 98.0 x 60.0 mm.
This is far from the latest version of the board, since there’s also a version with an additional shunt signal amplifier already laid out. Photos are included in the supplementary materials.
The circuit is roughly as shown. I say “roughly” because there are many versions of the device, but this schematic matches the actual board as closely as possible, right down to the component designators (the attached files include a component list).
The board lists the designer of the schematic and PCB layout, as well as their Taobao page.
The board uses one of the most popular microcontrollers, the Atmega8.
The flux around the microcontroller hasn’t been cleaned off, although the rest of the board is clean.
The board also has pads for connecting a programmer (which came in handy for me), as well as pads for connecting to a computer.
Overall, the board is quite well thought out, with plenty of pads for connecting various add-on modules.
8 MHz crystal.
Schematic of the processor section, computer interface, display connection, and the rest of the supporting circuitry.
I was really pleased that it can connect to a computer, and surprised by how many different menu options the software has. But I was very disappointed to find that it’s entirely in Chinese, with no English version at all :((((
What’s more, the menu items aren’t even displayed as Chinese characters—they show up as question marks.
This is what the software should look like when it’s running.
The circuit for controlling the MOSFET, amplifying the current-sensor signal, and setting the current with an integrating circuit.
MOSFET control is implemented using this circuit:
The Atmega’s PWM output feeds an integrating circuit with a 1/20 divider, followed by a comparator amplifier (which compares the control voltage with the voltage across the shunt).
That’s where the 250 mV control voltage for the power module from my review comes from.
The voltage across the shunt is then amplified 10 or 100 times (selected by the Atmega) by a second amplifier (a dual op-amp) and fed to the Atmega’s ADC input.
Schematic of the current-sensor amplifier and MOSFET control circuit.
The resistance measurement circuit is implemented using a four-wire method, by the way.
I haven’t looked into it in detail, since I haven’t thought about expanding the device’s functionality in this area yet, but it works—I tested it.
As far as I understand, it can also measure capacitor ESR.
The Rc circuit, in case anyone’s interested :)
The board supports several types of displays. It’s a small touch, but a nice one—credit to the designer. The display contrast is adjustable.
There are far fewer components on the back, but most of them are fairly large. Incidentally, the original design also has the crystal on this side. That makes the board a little flatter, since the side with the smaller components faces outward.
A group photo of the board's smaller details.
1. The current-sense shunt resistors, apparently even branded ones, are rated at 0.1 Ω and connected in parallel.
2. The fan temperature-control circuit. The design is primitive; there's something odd in place of the thermistor, and the transistor was faulty. The circuit for this section is shown below. In reality, the resistor in series with the thermistor is missing. It started working after I replaced the transistor, but not very well, so I'll replace it.
The power connector is also visible on the left.
3. External connection terminals: the input is at the top, with three terminals below for connecting a resistance meter.
4. Capacitors for the resistance meter. You can see a non-polarized electrolytic capacitor (green), but the board has pads for two polarized capacitors in series.
Fan control circuit.
The power supply section uses a popular KIS3R33 DC-DC converter, with a slight modification to provide an output voltage of 8.4 V.
Since the board draws very little current, the converter doesn't get hot during operation.
The op-amps are powered from this voltage; the rest of the board runs on 5 V, supplied by a 78L05 regulator powered from 8.4 V.
The circuit for this section—or rather, the converter's supporting components.
There are 1000 µF electrolytic capacitors before and after the DC-DC converter. They are clearly used parts, since one even has some minor visible damage.
The display is a standard 4-line, 20-character model with a bright backlight—very bright, in fact.
I think it could easily be replaced with any equivalent display.
There's firmware for a 2-line, 20-character display, but I don't think it would be very convenient to use.
When powered on, the display shows the device name, firmware version, and the author's email address for a few seconds—or so I assume, though I'm not sure.
I tried emailing them, but got no reply :(
But things didn't go entirely smoothly. I wrote to the author because I managed to break the board literally the day after it arrived. I later figured out why it happened, since I ran into the same problem again due to an oversight, though it looked a little different the second time.
The problem is that the processor's Reset input isn't pulled up to the supply voltage on the board. Under certain conditions, this can cause the firmware to get corrupted.
So, this is what I ended up with—
As you can imagine, I was very upset :(((
I had to find some way to solve the problem. The seller was unlikely to help, and I had already confirmed receipt of the order.
After a long search, I found a Chinese forum where the developer of this device hangs out.
But things weren't that simple there, either.
First, I had to register, and the forum was entirely in Chinese. To make matters worse, it turned out that you needed an invitation to register.
Further searching led me to a page where you could get an invitation. And, just to remind you, it was all in Chinese.
In the end, I managed to register, and this is what my profile on the site looks like now —
The problems didn't end there. It turned out that, as a newcomer, I couldn't create threads. I could reply, but not in every thread; I couldn't post links or send private messages.
Basically, one restriction after another. What a nightmare :((((
It was made even worse by how sloooooow the forum was. It felt like I'd gone back to a 14,400-baud modem.
Still, I managed to post photos of the problem and explain what I needed. They sent me firmware for my type of display without any trouble, which was great :))))
I know very few people in China speak English, but I was hoping that radio amateurs would know at least some, even if they had to use a translator. Surely they read datasheets in languages other than Chinese, too. I got lucky.
After that, I started figuring out how to get 15 A from the board. It took a while for them to understand what I meant. Luckily, I got in touch with a user called 100MHz (there are lots of his projects online), who eventually explained what I was doing wrong.
Though he didn't always write in English.
I was really surprised by how well they treated a newcomer from another country, and by the generally friendly atmosphere. Though replies often took quite a while.
Just out of curiosity, I posted the description of my radio component tester build. In response, I got a couple of private messages asking me to sell them kits of parts, since buying everything separately isn't very convenient. The first time, Chinese people asked me to sell them something instead of buying from them :)
But the device's developer (GANDF) never got back to me, even though I later sent him a private message and posted in his own thread. :(((
Next are the device's operating modes, which I'll hide under a spoiler.
Next, I'll move on to putting this board to use.
I'll start by listing everything I used (except for the enclosure and wires), with links to the corresponding reviews.
1. The board being reviewed.
2. Power module — review
3. Power supply — review
4. Current sensor amplifier board — review
A brief aside on connecting the current sensor amplifier board.
Since I chose a fairly small enclosure, I ran into a problem: the display wouldn't fit properly, so I had to shorten it a little :)
For the connections, I used colored wires from some kind of video cable. For the display's power wires, I used thinner wire with a cross-sectional area of 0.22 mm².
I took a photo of the wires connected to the encoder, just in case it comes in handy.
After extending the wires to the display and encoder, I decided to test it and show what the device looks like in operation.
The load is disconnected, the input is 5 V, and the threshold voltage is 1 V.
Voltages of 5 V, 10 V, 20 V, and 25 V. The voltmeter reads a little low, so I'll have to calibrate it. The load current is 1 A in every case.
There wasn't much to invent here, so I went straight to assembly.
First, I cut the necessary holes in the front panel, placed the components in the enclosure, and got an idea of how it would all look.
There was even some room left inside.
At first, I thought about keeping the control board and display together, but they wouldn’t fit that way, and I didn’t want the fan sticking out the back, so I had to split the boards.
Ideally, this compartment should also have held four batteries to power the electronics during a power outage (testing batteries can take quite a while).
I soldered the fan wires together and routed them out of the power module through some extra insulation.
I decided to mount the power module in the case using small pieces of fiberglass board. To keep them from sliding around during assembly, I stuck them down with double-sided tape.
But back to the control board.
It also went through a few changes and improvements.
First, I soldered in pins for connecting a programmer and a computer. The programming pins came in handy literally the next day.
The amplifier board fit right in as if it had always been there, though I did have to bend the trimmer resistor slightly. The board’s negative contact lined up perfectly with one of the ground holes on the main board, and I attached the board itself with double-sided tape.
The positive supply is also taken from one of the existing contacts.
Here’s what the amplifier board’s contacts do.
Signal — the input comes from one of the shunts on the power module. The positive lead connects to the junction of the shunt and the MOSFET lead; the negative lead connects to the common ground on the same board.
The output goes to the other side of the main board.
I had to make a few changes to the main board to connect all this stuff :)
1. I lifted pin 2 of the op-amp and connected it to pin 1. In this configuration, the amplifier works as a buffer.
2. I desoldered the capacitor between pins 1 and 2 and soldered it to one contact so I wouldn’t lose it, in case I need to reverse the modification.
3. The yellow wire (output) from the current-sensor amplifier board connects to resistor R12 (there’s a more detailed photo below).
There’s a diode at the board input to protect it against reverse polarity. Since I wanted the highest possible voltage for the fans, I added a jumper from the board input to the fan-control circuit’s supply. I’d cut the trace beforehand, but the cut is hidden under the jumper, so you can’t see it.
Though really, I could have left the trace intact: the connector is keyed, and the wires are soldered into the power supply.
While I was at it, I replaced the electrolytic capacitor at the load input. There wasn’t any particular reason; I just didn’t like the original one. I also soldered a 100 nF ceramic capacitor in parallel with it on the underside of the board; there was room for it.
And here's the overall picture of the modifications.
The modifications didn't come without a hitch. During the first power-up, I found that the current reading wasn't working—instead, it always showed 0.999 A.
After investigating, I found that the current-sensor amplifier board was outputting 4.35 V, which was then amplified tenfold. But since the op-amp supply voltage for the 10× (100×) amplifier was 8.4 V, the voltage at the Atmega input ended up well above 5 V.
At first I thought the current-sensor op-amp was faulty, and that I'd also burned out the ADC input (the voltage at the ADC input was coming from the Atmega—in other words, it was being driven out). But I got lucky.
I bought a couple of new processors and amplifiers. I replaced the amplifier first, but that didn't help. It turned out there was a small error in my circuit (I'll correct it in the relevant review). Once I fixed the error, the amplifier worked as it should. On a friend's advice, I decided not to replace the Atmega right away, but to try reflashing it first—and that did the trick.
Apparently, the spike scrambled the configuration of that input, turning it into an output.
This happened in part because the Reset input was left floating, so afterward I pulled it up to 5 V.
There aren't actually as many connections as it might seem. I sketched a block diagram to give a rough idea of what connects where.
Almost all the connections are in place; the front panel is still left to do. It's best to connect it last.
I connected the input terminals using a small piece of copper-clad fiberglass board, with the wires from the power module and the sense wires from the control board connected to it.
At first I wanted to mount the display with screws, but then I swallowed my perfectionism and simply glued it in place with hot glue.
The control board and power supply fit as if they'd always been there. I secured them with small insulated brackets, and the boards even ended up at the same height :)
As you can see in the photo, there aren't many connections, and there's plenty of free space for air to circulate.
By the way, I dealt with ventilation by adding more holes, drilling some in the bottom and back as well. Air coming in from the back cools the outside of the power module and the boards mounted on it.
It would of course be nice to secure the wires somehow, but they're holding up quite well thanks to their stiffness. I also considered putting a grille on this fan, but when I held it in place, I realized it would probably get in the way—especially since something could short out if the display board landed on it (during debugging, for example). So I left it as is and drilled some holes in the bottom instead :)
View from the side where the control boards and power supply are installed.
The device is already powered on in the photo.
I did install a protective grille on the back, though. I think it definitely won't hurt there.
The power cable connects via a standard connector salvaged from a disassembled computer power supply. A connector for connecting to a computer is planned below it.
The final step: assembly.
The case model is Z2A; here's a link to the description
Since I somehow managed to lose the case feet and their screws somewhere at home, I had to find replacements. I used the first screws I found that were the right size, and self-adhesive feet that had been sitting in my stash just for occasions like this.
I'll also show how they assemble a device like this in China, but I'll hide that under another spoiler :)
And there's also a big thread with several ways to use it, some of them quite attractive.
So, here's what I ended up with.
As you can see, I didn't go to all the trouble of using a small enclosure for nothing. The goal was not just to make a working device, but to give it a matching design with the power supply.
I had to shift the power button and encoder slightly because mounting posts were in the way. I ended up cutting them out of the power supply, since I noticed the problem when it was already too late. Here, I decided to keep the posts and move the controls instead.
Of course, you can stack them on top of each other too :)
The backs are also very similar, but the device being reviewed doesn't have a USB port yet, and it does have extra ventilation holes.
And of course, I did a little testing of the finished result.
It’s not quite as smooth as I’d hoped, of course. There’s still some tweaking to do, though the board offers plenty of options for that. I’ll get to it after making a few more modifications.
For now, here’s a load test with the power supply.
Since the power supply can deliver a maximum of 5 A and the load’s maximum input voltage is 50 V, I tested it at 50 V and 5 A.
I also tested it with a battery, since I wanted to try it at the maximum current.
But I ran into a minor disappointment, though it’s easy to fix.
At 14 A, only the ammeter needs adjusting.
But I couldn’t get 15 A. The likely culprit is that the current-sense shunts have slightly too much resistance. I made them from 10 resistors of 1.33 Ω each in parallel, trying to get the signal close to the 250 mV limit. I’ll probably have to solder one more resistor in parallel with each shunt, increasing the upper range by 10% and bringing it to 15 A.
By the way, while testing with the battery, I noticed that the internal resistance measurement seems to work too. If I’ve understood the purpose of the readings at the bottom left of the display correctly, the value rose slightly as the battery discharged.
Well, that's about it for now. As always, below you'll find a link to additional files to download. In the meantime, I'll try to list the pros and cons of this board.
Summary.
Pros
The board is practically a complete solution; all you need to add is a power supply and a heatsink.
Plenty of room for expansion
Can be connected to a computer
Can be used as an ESR meter (at least for measuring low-resistance resistors)
You can now build a similar device yourself.
Can be controlled using either the encoder or the keyboard (at least with some firmware versions).
Well-thought-out circuit board design.
Cons
Some shortcomings in the circuit design, though fortunately they're easy and inexpensive to fix.
The software is only available in Chinese.
The fan control circuit came with a faulty transistor.
My opinion: This is a very interesting board. With it, you can build a device that may not be as essential as a power supply, but is just as useful in certain situations.
I'd really like to reverse-engineer the communication protocol, since PC software would be useful and practical.
There are some complaints about a few shortcomings of the board, but even as it is, it works perfectly well.
I really wanted to increase the maximum current to 30 A (the seller's page claimed this was possible), but I couldn't get the firmware. And since I'm not a programmer, I couldn't modify the software myself.
I was also interested in getting in touch with Chinese radio amateurs. As it turned out, that was a positive experience.
Of course, this is a very basic electronic load. Professional instruments can do a lot more, though they also cost a lot more.
For example: constant-voltage mode (CV), constant-current mode (CC), constant-resistance mode (CR), constant-power mode (CW), and combined modes: CC+CV, CR+CW.
But I think that with enough motivation and skill, these modes could be added to this device too. The hardware can handle them; it's a matter of software.
I wouldn't be surprised if the PC software could extend the functionality in this direction.
I hope you found this review interesting. I've accumulated a lot of material, but tried to include only what's essential. If anyone is interested in this topic, I can help you design your own version or modify the finished board.
Plans for further modifications (some of these may be published).
1. Add a connector for an external high-power resistor to handle some of the dissipated power. It will be connected in series with the positive output of the power module, before the voltage-sensing point.
2. Redesign the fan control circuit. At the same time, I'll add indicators for several heatsink temperature thresholds and automatic shutdown in case of overheating (in addition to the hardware protection).
3. Install a module for connecting to a computer. I haven't decided yet whether the connection will be wired or wireless.
4. Try to translate the software into a language I can understand, though unfortunately that's probably beyond my abilities :(
5. I may also bring out terminals for measuring low-resistance resistors.
6. Convert the circuit to a three- or four-wire connection.
First, what is an electronic load?
A programmable electronic load is designed to act as a load when testing, setting up, and adjusting power supplies, amplifiers, and other electronic equipment.
As it happened, while I was working on this device, the board went out of stock on AliExpress, so here’s a link to a snapshot of the order page. But so readers won’t be disappointed, I found a place where it’s still available. It’s also on eBay, but the price there is not encouraging at all.
You can find this board online using the keywords M8 V6 electronic load. Note that it must be the V6, since the V7 is a completely different device by the same author.
There will also be some additional materials that may even help you try building one yourself.
While looking for the product link, I even stumbled across a mini-review of this board.
The board was packaged properly, but since it’s no longer available from the seller I bought it from, I won’t include photos of the packaging.
The board looks like a single unit with the display. The encoder, power transistor, and power and input wires are already connected.
So this is a ready-to-use solution straight out of the box: all you need to do is add a power supply and mount the transistor on a heatsink.
Board specifications:
Maximum input voltage — 50 V
Maximum load current — 5 A (15 A after modification)
Current setting resolution — 1 mA
Cutoff voltage setting resolution — 10 mV
Supply voltage — 12–16 V (limited by the input capacitor)
Current consumption with the display — about 100 mA
Fan control, but it’s very basic.
Battery capacity measurement
Low-resistance resistor measurement.
Three-wire power circuit connection (can be modified for four-wire operation).
Audible alert when the input voltage drops below the set level.
I’ll go into more detail in the review itself.
The board came with:
A PSMN020-150W MOSFET. I was even surprised that the heat-shrink tubing was color-matched to the wires.
A 20-position encoder with a built-in push button. The knob was included too.
And now, photos of the board. Some will be available for download in full resolution. If you need any other photos, let me know and I’ll provide them.
The board is about the same size as a standard 2004 display: 98.0 x 60.0 mm.
This is far from the latest version of the board, since there’s also a version with an additional shunt signal amplifier already laid out. Photos are included in the supplementary materials.
The circuit is roughly as shown. I say “roughly” because there are many versions of the device, but this schematic matches the actual board as closely as possible, right down to the component designators (the attached files include a component list).
The board lists the designer of the schematic and PCB layout, as well as their Taobao page.
The board uses one of the most popular microcontrollers, the Atmega8.
The flux around the microcontroller hasn’t been cleaned off, although the rest of the board is clean.
The board also has pads for connecting a programmer (which came in handy for me), as well as pads for connecting to a computer.
Overall, the board is quite well thought out, with plenty of pads for connecting various add-on modules.
8 MHz crystal.
Schematic of the processor section, computer interface, display connection, and the rest of the supporting circuitry.
I was really pleased that it can connect to a computer, and surprised by how many different menu options the software has. But I was very disappointed to find that it’s entirely in Chinese, with no English version at all :((((
What’s more, the menu items aren’t even displayed as Chinese characters—they show up as question marks.
This is what the software should look like when it’s running.
The circuit for controlling the MOSFET, amplifying the current-sensor signal, and setting the current with an integrating circuit.
MOSFET control is implemented using this circuit:
The Atmega’s PWM output feeds an integrating circuit with a 1/20 divider, followed by a comparator amplifier (which compares the control voltage with the voltage across the shunt).
That’s where the 250 mV control voltage for the power module from my review comes from.
The voltage across the shunt is then amplified 10 or 100 times (selected by the Atmega) by a second amplifier (a dual op-amp) and fed to the Atmega’s ADC input.
Schematic of the current-sensor amplifier and MOSFET control circuit.
The resistance measurement circuit is implemented using a four-wire method, by the way.
I haven’t looked into it in detail, since I haven’t thought about expanding the device’s functionality in this area yet, but it works—I tested it.
As far as I understand, it can also measure capacitor ESR.
The Rc circuit, in case anyone’s interested :)
The board supports several types of displays. It’s a small touch, but a nice one—credit to the designer. The display contrast is adjustable.
There are far fewer components on the back, but most of them are fairly large. Incidentally, the original design also has the crystal on this side. That makes the board a little flatter, since the side with the smaller components faces outward.
A group photo of the board's smaller details.
1. The current-sense shunt resistors, apparently even branded ones, are rated at 0.1 Ω and connected in parallel.
2. The fan temperature-control circuit. The design is primitive; there's something odd in place of the thermistor, and the transistor was faulty. The circuit for this section is shown below. In reality, the resistor in series with the thermistor is missing. It started working after I replaced the transistor, but not very well, so I'll replace it.
The power connector is also visible on the left.
3. External connection terminals: the input is at the top, with three terminals below for connecting a resistance meter.
4. Capacitors for the resistance meter. You can see a non-polarized electrolytic capacitor (green), but the board has pads for two polarized capacitors in series.
Fan control circuit.
The power supply section uses a popular KIS3R33 DC-DC converter, with a slight modification to provide an output voltage of 8.4 V.
Since the board draws very little current, the converter doesn't get hot during operation.
The op-amps are powered from this voltage; the rest of the board runs on 5 V, supplied by a 78L05 regulator powered from 8.4 V.
The circuit for this section—or rather, the converter's supporting components.
There are 1000 µF electrolytic capacitors before and after the DC-DC converter. They are clearly used parts, since one even has some minor visible damage.
The display is a standard 4-line, 20-character model with a bright backlight—very bright, in fact.
I think it could easily be replaced with any equivalent display.
There's firmware for a 2-line, 20-character display, but I don't think it would be very convenient to use.
When powered on, the display shows the device name, firmware version, and the author's email address for a few seconds—or so I assume, though I'm not sure.
I tried emailing them, but got no reply :(
But things didn't go entirely smoothly. I wrote to the author because I managed to break the board literally the day after it arrived. I later figured out why it happened, since I ran into the same problem again due to an oversight, though it looked a little different the second time.
The problem is that the processor's Reset input isn't pulled up to the supply voltage on the board. Under certain conditions, this can cause the firmware to get corrupted.
So, this is what I ended up with—
As you can imagine, I was very upset :(((
I had to find some way to solve the problem. The seller was unlikely to help, and I had already confirmed receipt of the order.
After a long search, I found a Chinese forum where the developer of this device hangs out.
But things weren't that simple there, either.
First, I had to register, and the forum was entirely in Chinese. To make matters worse, it turned out that you needed an invitation to register.
Further searching led me to a page where you could get an invitation. And, just to remind you, it was all in Chinese.
In the end, I managed to register, and this is what my profile on the site looks like now —
The problems didn't end there. It turned out that, as a newcomer, I couldn't create threads. I could reply, but not in every thread; I couldn't post links or send private messages.
Basically, one restriction after another. What a nightmare :((((
It was made even worse by how sloooooow the forum was. It felt like I'd gone back to a 14,400-baud modem.
Still, I managed to post photos of the problem and explain what I needed. They sent me firmware for my type of display without any trouble, which was great :))))
I know very few people in China speak English, but I was hoping that radio amateurs would know at least some, even if they had to use a translator. Surely they read datasheets in languages other than Chinese, too. I got lucky.
After that, I started figuring out how to get 15 A from the board. It took a while for them to understand what I meant. Luckily, I got in touch with a user called 100MHz (there are lots of his projects online), who eventually explained what I was doing wrong.
Though he didn't always write in English.
I was really surprised by how well they treated a newcomer from another country, and by the generally friendly atmosphere. Though replies often took quite a while.
Just out of curiosity, I posted the description of my radio component tester build. In response, I got a couple of private messages asking me to sell them kits of parts, since buying everything separately isn't very convenient. The first time, Chinese people asked me to sell them something instead of buying from them :)
But the device's developer (GANDF) never got back to me, even though I later sent him a private message and posted in his own thread. :(((
Next are the device's operating modes, which I'll hide under a spoiler.
Next, I'll move on to putting this board to use.
I'll start by listing everything I used (except for the enclosure and wires), with links to the corresponding reviews.
1. The board being reviewed.
2. Power module — review
3. Power supply — review
4. Current sensor amplifier board — review
A brief aside on connecting the current sensor amplifier board.
Since I chose a fairly small enclosure, I ran into a problem: the display wouldn't fit properly, so I had to shorten it a little :)
For the connections, I used colored wires from some kind of video cable. For the display's power wires, I used thinner wire with a cross-sectional area of 0.22 mm².
I took a photo of the wires connected to the encoder, just in case it comes in handy.
After extending the wires to the display and encoder, I decided to test it and show what the device looks like in operation.
The load is disconnected, the input is 5 V, and the threshold voltage is 1 V.
Voltages of 5 V, 10 V, 20 V, and 25 V. The voltmeter reads a little low, so I'll have to calibrate it. The load current is 1 A in every case.
There wasn't much to invent here, so I went straight to assembly.
First, I cut the necessary holes in the front panel, placed the components in the enclosure, and got an idea of how it would all look.
There was even some room left inside.
At first, I thought about keeping the control board and display together, but they wouldn’t fit that way, and I didn’t want the fan sticking out the back, so I had to split the boards.
Ideally, this compartment should also have held four batteries to power the electronics during a power outage (testing batteries can take quite a while).
I soldered the fan wires together and routed them out of the power module through some extra insulation.
I decided to mount the power module in the case using small pieces of fiberglass board. To keep them from sliding around during assembly, I stuck them down with double-sided tape.
But back to the control board.
It also went through a few changes and improvements.
First, I soldered in pins for connecting a programmer and a computer. The programming pins came in handy literally the next day.
The amplifier board fit right in as if it had always been there, though I did have to bend the trimmer resistor slightly. The board’s negative contact lined up perfectly with one of the ground holes on the main board, and I attached the board itself with double-sided tape.
The positive supply is also taken from one of the existing contacts.
Here’s what the amplifier board’s contacts do.
Signal — the input comes from one of the shunts on the power module. The positive lead connects to the junction of the shunt and the MOSFET lead; the negative lead connects to the common ground on the same board.
The output goes to the other side of the main board.
I had to make a few changes to the main board to connect all this stuff :)
1. I lifted pin 2 of the op-amp and connected it to pin 1. In this configuration, the amplifier works as a buffer.
2. I desoldered the capacitor between pins 1 and 2 and soldered it to one contact so I wouldn’t lose it, in case I need to reverse the modification.
3. The yellow wire (output) from the current-sensor amplifier board connects to resistor R12 (there’s a more detailed photo below).
There’s a diode at the board input to protect it against reverse polarity. Since I wanted the highest possible voltage for the fans, I added a jumper from the board input to the fan-control circuit’s supply. I’d cut the trace beforehand, but the cut is hidden under the jumper, so you can’t see it.
Though really, I could have left the trace intact: the connector is keyed, and the wires are soldered into the power supply.
While I was at it, I replaced the electrolytic capacitor at the load input. There wasn’t any particular reason; I just didn’t like the original one. I also soldered a 100 nF ceramic capacitor in parallel with it on the underside of the board; there was room for it.
And here's the overall picture of the modifications.
The modifications didn't come without a hitch. During the first power-up, I found that the current reading wasn't working—instead, it always showed 0.999 A.
After investigating, I found that the current-sensor amplifier board was outputting 4.35 V, which was then amplified tenfold. But since the op-amp supply voltage for the 10× (100×) amplifier was 8.4 V, the voltage at the Atmega input ended up well above 5 V.
At first I thought the current-sensor op-amp was faulty, and that I'd also burned out the ADC input (the voltage at the ADC input was coming from the Atmega—in other words, it was being driven out). But I got lucky.
I bought a couple of new processors and amplifiers. I replaced the amplifier first, but that didn't help. It turned out there was a small error in my circuit (I'll correct it in the relevant review). Once I fixed the error, the amplifier worked as it should. On a friend's advice, I decided not to replace the Atmega right away, but to try reflashing it first—and that did the trick.
Apparently, the spike scrambled the configuration of that input, turning it into an output.
This happened in part because the Reset input was left floating, so afterward I pulled it up to 5 V.
There aren't actually as many connections as it might seem. I sketched a block diagram to give a rough idea of what connects where.
Almost all the connections are in place; the front panel is still left to do. It's best to connect it last.
I connected the input terminals using a small piece of copper-clad fiberglass board, with the wires from the power module and the sense wires from the control board connected to it.
At first I wanted to mount the display with screws, but then I swallowed my perfectionism and simply glued it in place with hot glue.
The control board and power supply fit as if they'd always been there. I secured them with small insulated brackets, and the boards even ended up at the same height :)
As you can see in the photo, there aren't many connections, and there's plenty of free space for air to circulate.
By the way, I dealt with ventilation by adding more holes, drilling some in the bottom and back as well. Air coming in from the back cools the outside of the power module and the boards mounted on it.
It would of course be nice to secure the wires somehow, but they're holding up quite well thanks to their stiffness. I also considered putting a grille on this fan, but when I held it in place, I realized it would probably get in the way—especially since something could short out if the display board landed on it (during debugging, for example). So I left it as is and drilled some holes in the bottom instead :)
View from the side where the control boards and power supply are installed.
The device is already powered on in the photo.
I did install a protective grille on the back, though. I think it definitely won't hurt there.
The power cable connects via a standard connector salvaged from a disassembled computer power supply. A connector for connecting to a computer is planned below it.
The final step: assembly.
The case model is Z2A; here's a link to the description
Since I somehow managed to lose the case feet and their screws somewhere at home, I had to find replacements. I used the first screws I found that were the right size, and self-adhesive feet that had been sitting in my stash just for occasions like this.
I'll also show how they assemble a device like this in China, but I'll hide that under another spoiler :)
And there's also a big thread with several ways to use it, some of them quite attractive.
So, here's what I ended up with.
As you can see, I didn't go to all the trouble of using a small enclosure for nothing. The goal was not just to make a working device, but to give it a matching design with the power supply.
I had to shift the power button and encoder slightly because mounting posts were in the way. I ended up cutting them out of the power supply, since I noticed the problem when it was already too late. Here, I decided to keep the posts and move the controls instead.
Of course, you can stack them on top of each other too :)
The backs are also very similar, but the device being reviewed doesn't have a USB port yet, and it does have extra ventilation holes.
And of course, I did a little testing of the finished result.
It’s not quite as smooth as I’d hoped, of course. There’s still some tweaking to do, though the board offers plenty of options for that. I’ll get to it after making a few more modifications.
For now, here’s a load test with the power supply.
Since the power supply can deliver a maximum of 5 A and the load’s maximum input voltage is 50 V, I tested it at 50 V and 5 A.
I also tested it with a battery, since I wanted to try it at the maximum current.
But I ran into a minor disappointment, though it’s easy to fix.
At 14 A, only the ammeter needs adjusting.
But I couldn’t get 15 A. The likely culprit is that the current-sense shunts have slightly too much resistance. I made them from 10 resistors of 1.33 Ω each in parallel, trying to get the signal close to the 250 mV limit. I’ll probably have to solder one more resistor in parallel with each shunt, increasing the upper range by 10% and bringing it to 15 A.
By the way, while testing with the battery, I noticed that the internal resistance measurement seems to work too. If I’ve understood the purpose of the readings at the bottom left of the display correctly, the value rose slightly as the battery discharged.
Well, that's about it for now. As always, below you'll find a link to additional files to download. In the meantime, I'll try to list the pros and cons of this board.
Summary.
Pros
The board is practically a complete solution; all you need to add is a power supply and a heatsink.
Plenty of room for expansion
Can be connected to a computer
Can be used as an ESR meter (at least for measuring low-resistance resistors)
You can now build a similar device yourself.
Can be controlled using either the encoder or the keyboard (at least with some firmware versions).
Well-thought-out circuit board design.
Cons
Some shortcomings in the circuit design, though fortunately they're easy and inexpensive to fix.
The software is only available in Chinese.
The fan control circuit came with a faulty transistor.
My opinion: This is a very interesting board. With it, you can build a device that may not be as essential as a power supply, but is just as useful in certain situations.
I'd really like to reverse-engineer the communication protocol, since PC software would be useful and practical.
There are some complaints about a few shortcomings of the board, but even as it is, it works perfectly well.
I really wanted to increase the maximum current to 30 A (the seller's page claimed this was possible), but I couldn't get the firmware. And since I'm not a programmer, I couldn't modify the software myself.
I was also interested in getting in touch with Chinese radio amateurs. As it turned out, that was a positive experience.
Of course, this is a very basic electronic load. Professional instruments can do a lot more, though they also cost a lot more.
For example: constant-voltage mode (CV), constant-current mode (CC), constant-resistance mode (CR), constant-power mode (CW), and combined modes: CC+CV, CR+CW.
But I think that with enough motivation and skill, these modes could be added to this device too. The hardware can handle them; it's a matter of software.
I wouldn't be surprised if the PC software could extend the functionality in this direction.
I hope you found this review interesting. I've accumulated a lot of material, but tried to include only what's essential. If anyone is interested in this topic, I can help you design your own version or modify the finished board.
Plans for further modifications (some of these may be published).
1. Add a connector for an external high-power resistor to handle some of the dissipated power. It will be connected in series with the positive output of the power module, before the voltage-sensing point.
2. Redesign the fan control circuit. At the same time, I'll add indicators for several heatsink temperature thresholds and automatic shutdown in case of overheating (in addition to the hardware protection).
3. Install a module for connecting to a computer. I haven't decided yet whether the connection will be wired or wireless.
4. Try to translate the software into a language I can understand, though unfortunately that's probably beyond my abilities :(
5. I may also bring out terminals for measuring low-resistance resistors.
6. Convert the circuit to a three- or four-wire connection.
This page was found when searching for:
схема kblc240d, схема печатной платы контроллера сварки с постоянным сопротивлением, шім-контролер заряду акб 50v 50а, как выглядит сгоревший диод на платье, 1, плата управления питаниемчто такое плата управления питанием, блок питани плата управления нагрузкой, блок питания с энкодером своими руками, как на плате обозначается питание, обозначения на электронных платах, регулятор оборотов kblc 240d, trsl 60w c[tvf, подключение таходатчика к kblc-240d, ремонт и наладка платы kblc 240d 4771e китайского токарного станка, fta03 eu ghbywbgbfkmyfz c tvf, блок питания пароувлажнителя схема, kblc 240d datasheet, плата управления какие характеристики, блок управления на микроконтроллере для блока питания, схема блока питания лазерного станка, gandf нагрузка, схема бесперебойника 5 kwt, eltek circuit diagram, kblc 240d, kblc 240d схема подключения
схема kblc240d, схема печатной платы контроллера сварки с постоянным сопротивлением, шім-контролер заряду акб 50v 50а, как выглядит сгоревший диод на платье, 1, плата управления питаниемчто такое плата управления питанием, блок питани плата управления нагрузкой, блок питания с энкодером своими руками, как на плате обозначается питание, обозначения на электронных платах, регулятор оборотов kblc 240d, trsl 60w c[tvf, подключение таходатчика к kblc-240d, ремонт и наладка платы kblc 240d 4771e китайского токарного станка, fta03 eu ghbywbgbfkmyfz c tvf, блок питания пароувлажнителя схема, kblc 240d datasheet, плата управления какие характеристики, блок управления на микроконтроллере для блока питания, схема блока питания лазерного станка, gandf нагрузка, схема бесперебойника 5 kwt, eltek circuit diagram, kblc 240d, kblc 240d схема подключения
The figures from this review are in the measurement database: Electronic loads
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!
Comments10
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 forumOnly 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
Comments are published in their original language.