/       /       /    Meanwell AD-55A, a power supply with an uninterruptible power supply function, or how to turn a UPS into a UPS
Support the project on Patreon


Meanwell AD-55A, a power supply with an uninterruptible power supply function, or how to turn a UPS into a UPS

Some time ago, one of my readers asked me to help modify an uninterruptible power supply from Meanwell.
This company's power supplies are very good, as I've shown more than once, but this UPS managed to surprise me—and I wouldn't say pleasantly. More on that in the review.

As usual, a reminder: to stay up to date with new posts on the blog, I recommend following me on Instagram, where I'll post notifications about all new posts and maybe write about something interesting from time to time: link to my account

The request was to increase the charging current and add an auto-start function when the battery is connected.
The thing is, this UPS is rated for a charging current of just 230 mA, which under current conditions means charging takes almost forever. Also, if the UPS shuts down because the battery is discharged, you have to briefly connect it to mains power to get it running again after replacing the battery with a charged one.
I agreed with the first point, but I'm skeptical about the second, though that's mostly a subjective opinion. Right now, users are advised to install a button to start these UPS units, which is obviously not very convenient.
There was also a third request: to try to increase the output power.

This review overlaps with quite a few of my other reviews, so if you haven't read them, I recommend checking them out.
1. 12 V, 3 A power supply and a little tweaking
2. VariCore 32700 LiFePO4 batteries and a few thoughts on replacing lead-acid batteries in a UPS
3. MeanWell PSC-160A-C, a 160 W power supply with a UPS function
4. MeanWell PSC-160A-C: fixing its shortcomings, modifications, and an addendum to the review
5. Another set of Vapcell LiFePO4 32700 batteries with a claimed capacity of 6200 mAh
6. An interesting, but slightly odd, battery protection board

The unit has the familiar design many users will recognize: a metal case with a perforated cover.


This UPS series comes in several versions, differing in voltage (A or B suffix), power (AD-55 and AD-155 models), and whether they have a second 5 V output (AD and ADD series).
In my case, the UPS was marked AD-55A, meaning 13.8 V, 55 W, and no second 5 V output.


At one end is the usual terminal block, with connections for the mains cable, 13.8 V output, and battery, plus one unused terminal used in the ADD series. On the right, there's also a spot for a three-pin connector (presumably), a trimmer, and an operation indicator LED.

This power supply is clearly simpler than the newer PSC-160A series, which delivers three times the power in almost the same volume and also has relay outputs that signal mains failure and battery low voltage. In fact, though, there are many more differences between the AD/ADD and PSC series.


Once again, I found a bent corner on the case. I get the feeling this corner gets bent during shipping in about 30% of cases—or am I just that lucky?


I immediately unscrewed the case and started inspecting the board. I actually came across these power supplies quite a few times about 10 years ago, but those were 27 V, 155 W models. Back then, though, I just bought them, installed them, and forgot about them—and I shouldn't have.


The layout is quite spacious, with lots of empty space.


Mains filter, thermistor, capacitors, transformer—nothing to complain about here. The only thing missing, perhaps, is a varistor; otherwise, it's all there. No complaints about the components themselves either: branded capacitors and fairly neat assembly. Mean Well is Mean Well.
The PWM controller is the well-known UC3842, which makes repairing power supplies like this much easier: the controller is not only inexpensive, but also available at almost any electronics store. I was pleased to see the 150 µF input capacitor. That's enough for the power supply to operate across the full input voltage range, and for our purposes it's more than enough.


The output diode assembly is mounted on a heatsink, but for some reason they used an X12SL, which is just a fast diode, rather than the usual Schottky.
The output has four 470 µF, 25 V capacitors: three before the inductor and one after it.
There's also a 10 A, 45 V diode assembly on a separate heatsink, with a relay next to it that switches the battery circuit.
The battery circuit has a 15 A fuse.


The owner of the power supply had already tried to improve the cooling a little by adding thermal paste between the case and the transistor heatsink. But apparently his uncle works at a thermal paste factory, because there was... a lot of paste :) Still, it was a useful idea.


Since there's no 5 V channel on the board, there are lots of empty component footprints. Also, it's already clear that it was implemented using a DC-DC converter, but even so, there seem to be far too many empty spots.


Let's switch it on, poke around with a multimeter, and see what we get.
The power supply is initially set to 13.8 V output. The voltage at the battery connection is 13.4 V, which surprised me a little, since with the circuit designs I'm used to, the voltages should be the same.
The output voltage could be adjusted from 10.9 to 14.3 V. But since I measured at the battery terminals, the main output would actually be 11.3–14.7 V.


For the next tests, let's connect the power supply to the load, without a battery for now.


With an input of 230 V and an output of about 14.3 V (I set the charging voltage to the standard 13.80 V), I got a load current of 4.8 A and nearly 70 W. At an output of around 12 V, it delivered 65 W, but the current was 5.4 A.


At an input voltage of 115 V, the power was even higher: 70 W at 5 A.


The charger's load characteristic, though, turned out to be rather strange. Instead of constant-current mode, I saw a simple linear relationship between voltage and load current, like in primitive circuits with a current-limiting resistor.


With a battery connected, the power supply delivered up to 10 A to the output (the load's maximum), but above 4.8 A the output voltage started to fluctuate, indicating that the built-in power supply's overload protection was kicking in. In other words, the output was running from the battery, but the power supply kept starting up and immediately going into protection. If it had simple current limiting instead of overload protection, it would help the battery; as it is, all we see is the voltage fluctuating.


When running on battery alone, we see just a linear relationship between voltage and load current.


When switching back and forth between mains and battery on the fly, you can see that the output voltage is regulated while running from mains. Naturally, there are no battery/mains switching events, since this is essentially an online UPS: the battery is always connected to the load through an isolation diode.


Next came a long-duration load test. I tested it in three stages at 1.5, 3, and 4.5 A, even though the rated maximum current is 3.5 A.
First, I checked how the output voltage varied with load current and temperature.
1. Cold power supply, no load
2. Cold power supply, 1.5 A
3. Hot power supply, 4.5 A
4. Hot power supply, immediately after removing the load.

Personally, I think this is simply perfect: even under maximum load, the difference is just 12 mV due to load current and 4 mV due to temperature.


Temperature after loading at 1.5, 3.0, and 4.5 A. The first test lasted 20 minutes; you can estimate the duration of the others from the clock in the thermal images.


Anyway, I played around with it for a bit, partially discharged the battery, then left the UPS connected to mains and went about my business. About 6 hours later, I decided to check out of curiosity whether it had charged. I'd discharged it by about 800–1000 mAh, and the stated charging current was 230 mA, so it should have been charged by then.
But things turned out to be more interesting: the battery was at only about 13.4 V, even though I'd set it to 13.8 V, and the charging current was... about 60 mA!

Naturally, at that charging current, it'll take... almost forever to charge... And if you consider that there are places where the power is on for only 4 hours between outages, it will never fully charge.


Everything fell into place when I saw the schematic for this section. Thanks to the person who drew it up. Of course, I could have done it myself, but having it ready-made is always a plus.

I never expected this kind of dirty trick from Mean Well. Charging current flows through two 3 Ω resistors and a 1N4007 diode, so with an output voltage of, say, 14.2 V, the charging current will be close to zero when the battery voltage is around 13.7 V. Brutal.


I can't help thinking of a video by someone who showed off his UPS circuit. I mention it because his design is even worse, despite being based on a similar principle. He also shows a newer Mean Well UPS, the PSC-35, and I suspect they got everything right in that one. The channel is interesting in its own way; it often features teardowns of various inverters, power stations, and so on. But the author is very strange. For example, he banned me for leaving comments, so be careful: any criticism may not be taken very well.


By the way, here's another video, and a question while we're at it. The author claims his USB tester burned out from the QC output because it was “old and not rated for voltages above 5 V.” But as far as I know—and what's more, the author showed this tester in a review and seemed to say that the voltage regulator had burned out—it should clearly be able to handle at least 20 V at the input. Has anyone come across testers like the one shown that are rated for only 5 V?


By this point, it was clear that there was no way to fix this with simple means, so I had to go back to my review from almost nine years ago and repeat what I did then.
Basically, I need to convert this UPS into a regular power supply, then turn it back into a UPS.

The board I made back then was designed for an output current of just 1 A and a charging current of up to 0.4 A, so I had to modify it. And since this board was designed for a specific device, I also had to reroute it.


The circuit is essentially unchanged. The number of current-sense resistors (R3–R7) has been increased to 5. Of course, you could replace them with a single lower-value resistor. The charging current is calculated using the rule that 1 Ω = 600 mA, so the charging current is given by I = 0.6/R, or 0.6/I = R.
Most components can have fairly wide tolerances. R1 and R2 can be anywhere from 47 to 470 Ω, though around 100–200 Ω is better.
I used a BC547 transistor; any similar one will do, even a basic KT315 or KT3102.
R8–R10 form a voltage divider, selected so that the board disconnects the battery when its voltage drops to 10 V. You can raise the cutoff threshold a little by using a 4.7 kΩ resistor for R7; that may even be more convenient.
C1, C3, and C4 can be 0.1–0.22 µF, but 0.22 µF is better, at least for C1.
C2 can be 47–220 µF. It protects against false shutdowns caused by brief voltage dips.
For diode assembly D1, choose one with the lowest possible forward voltage drop; the other diodes can be whatever is convenient.


The board was laid out for through-hole components, except for capacitor C4, which I decided to add later. I considered making it for SMD components, but the space savings would have been small, so I left it as is.


The rest is familiar to every electronics hobbyist: fiberglass board, paper, printer, iron, ferric chloride, drilling, cleaning, tinning—and the PCBs are ready. I made three at once because I have other power supplies to modify as well.


To modify it, remove the board from the case. Yes, I laid out the board before completely disassembling the power supply; I already knew what I would find inside.


There's thermal paste here too—both gray and white. I suspect the white paste was there originally, and the gray was added later.
Anyway, I wiped the paste off the heatsinks, but still kept getting it on myself somehow :)


The underside of the board; the components are mounted on one side only.


But what we'll be interested in later is the secondary (cold) side of the power supply.



Since large and small components are mixed together on the top of the board, let's start by removing the larger ones:
1. Relay
2. The diode assembly on the heatsink.
3. R50 resistors, 2 W, 3 Ω, 2 pcs.
4. Resistor R27, 2 W, 330 Ω. In principle, I could have left it in place, but personally I think it just heats the air and draws extra current from the battery.


Now desolder everything in the highlighted areas. You don't actually have to remove everything; I just didn't see any point in leaving those components there.


There are two small 4148 diodes and a ZD42 zener diode on the board. Set the zener aside right away so you don't mix it up with the diodes.


The board should now look like this.


We'll need these salvaged components:
1. Relay
2. Diode array
3. A pair of 4148 diodes.

You can use the original transistor, but it has a different pinout, even though it's otherwise suitable. I didn't think of that when I laid out the board.
Add the other components to the pile, too. They don't have to be new, but they do need to be in working order. For example, I salvaged the TL431 from some power supply.


Assemble the little board; there's nothing complicated here either. I could mark where each component goes, of course, but there are so few that the board itself makes it all obvious.
I originally wanted to design the board so the diode array would fit in the original spot, but the power supply's own diode array got in the way, so I had to bend the leads as shown in the photo.


Now for the actual wiring.
The board shows where to connect the new board. For easier installation, you also need to solder in a jumper. Use wire that's not too thin, since it will carry all the current from the battery.

That gives us 5 connection points. The most awkward one is CC. The wire is thin, but it has to be soldered directly to the lead of resistor R24 that's closest to the edge of the board, or to the TL431 lead that's also closest to the edge. The holes there are large, though, so I simply inserted a trimmed resistor lead, heated it up, soldered it from the back, and then soldered the wire on top.


You also need to cut one trace on the board. Since we're modifying a UPS rather than a regular power supply, the job is a little easier—at the very least, it already has a terminal block with the right number of contacts.


Then solder on the wires and route them as shown in the photo. All the wires except CC should have a fairly large cross-section: at least 0.5 mm², preferably 0.75 mm².
Solder a tinned piece of wire to the left of the relay; you'll also need it later to secure the board.

The diode array is insulated from the heatsink. I don't think anything dangerous would happen if you left it uninsulated, since its case is connected to the battery negative, but it's still better to fit an insulator.


The result is this little contraption.


Here you can see why the tinned piece of wire is needed: besides connecting the board to the P+ point, it also holds the circuit board in place.
In other words, the board is secured at two points: by the diode array and by this lead. You can't just leave it screwed to the heatsink; an impact would simply break it off.
There's no need to drill any holes in the heatsink. The diode array screws into the existing hole for the 5 V channel's diode array.

Speaking of the diode array.
Originally, it was mounted on a small heatsink, even though almost the same current flows through it as through the power supply's main diode array, which was screwed to the aluminum case. A strange design choice.


Without reassembling the unit, connect the battery and multimeter, then test it.
First, set the charge termination voltage to 13.80 V. Do this with no battery connected. You can connect to either the main output or the battery output.
Connect the battery and check the charging current: 1.8 A. The first modification is complete.


Connect a load to the battery output and check how the charging current depends on the battery voltage.
Everything looks good here too: the current stays constant from the start until the battery reaches 13.80 V. Then, for obvious reasons, it starts to drop. That's no longer a problem with the power supply, although there's still room for improvement here too :)))


The second task (or maybe the third) is to increase the output power. Warning: risk of electric shock!

The power supply handled loads up to 4.5 A despite being rated for 3.5 A, and it was clear there was still some headroom. This is partly because the company makes good power supplies, and partly because this one is designed to handle the full 100–240 V input range, so our 192–252 V gives it more headroom.

There is a current-sense resistor on the board in the high-voltage transistor circuit. Its value is 0.39 Ω. Figuring I could increase the power by 20–30%, I added a 1.2 Ω resistor in parallel with the original one.


But either I misunderstood something, or the power supply didn't understand what I wanted from it, because it delivered over 100 W, even though I expected it to be closer to 80–90 W. Only later did I realize that I'd calculated the maximum power as 60 W, even though the output was actually 70 W.


Anyway, I took another 1.2 Ω resistor and connected the two so their combined resistance was 2.4 Ω, then soldered them back in.


That's better: output current up to 6.3 A at 13.8 V, for about 85 W.


Keep in mind, though, that the battery charging current adds to the load current. So if the battery is charging at the same time, the power supply will go into protection at around 4.5–4.6 A. In other words, the protection trips at 6.4 A; the charging current is 1.8 A, so protection kicks in at 6.4 − 1.8 = 4.6 A.
If you load the UPS with a fully charged battery, it will switch over to battery power at 6.4 A, since there is no charging current.


When running on battery alone, there is almost no change. The voltage sagged slightly at maximum current, but I think that's because the wires are longer now, or the battery hadn't finished charging.


Voltage drops. Yes, they're there, but to be fair, even from a circuit-design standpoint they can't be any greater here than in the original version, because the current follows exactly the same path: terminal, fuse, relay contacts, diode assembly, terminal. The only change is the addition of wires between the boards. Also, previously all the losses were on the positive side; now they're split between the positive and negative sides.
At 5 A, the measurements are:
1. Negative-to-negative path: 0.47 V
2. Positive-to-positive path: 0.21 V.
3. The drop from the positive terminal to the contact after the fuse is 0.04 V.
4. Across the relay contacts: 0.076 V.

The total drop at 5 A is 0.68 V. Most of it is across the diode assembly—I didn't measure it, but it's about 0.45 V. Another 0.1 V or so is lost across the fuse and relay contacts, with the rest across the wires and board traces.
The PSC series UPS review wasn't entirely smooth sailing either. I showed there that to reduce the voltage drop, I had to reinforce the traces, replace the fuse with an SMD one, and I even included measured voltage-drop values. But that design didn't have a diode assembly as the main “load.” It used a different circuit design. The same approach could be used here, but that would mean a different circuit, with a low-resistance shunt and an op-amp, which is of course more complicated than just a resistor, transistor, and diode. So the question is whether it's worth developing this idea further.


On the plus side, I found during testing that the balancer on the protection board was working. Two LEDs lit up during charging, and the balancer resistors got warm. After a while, the LEDs went out. When the voltage reached 13.80 V and the charger switched to CV mode, the LEDs came on again, but noticeably dimmer and flickering. Now I'm wondering why they didn't light up during my tests.


Now, about the output voltage.
Above, I showed that the original circuit had charging issues, but the output voltage didn't depend on the battery's state of charge. Now it does, and that's inherent in the circuit design. In other words, the charging control circuit will try to maintain the set current, which requires keeping the voltage higher than the battery voltage. The difference is about 0.6 V, which is exactly the turn-on voltage of the transistor connected to the current-sense shunt. That's why the calculation is based on 0.6 V. After switching to the CV phase, the voltage will drop until it reaches 13.80 V, the charge termination voltage.
This refers to the main output voltage; the battery voltage will never exceed the value you set.

You can get rid of this by reworking the circuit to use the low-resistance shunt version I mentioned above. This isn't a drawback of my version: the original UPS also needed a voltage almost 0.5 V higher than the battery voltage, but that was due to the voltage drop across the diode.


The third task is automatic startup. In my view, it's an unnecessary feature, but as the experience of the person who sent me this UPS showed, he does need it sometimes. As he said, he was advised to install a button, but that isn't very convenient, so I decided to automate the process specifically for him.
This is very easy to do: just add a capacitor, shown in red in the schematic, with the same capacitance as C2 or higher.


There was no provision for it on the circuit board, so I mounted it on the back, suspended by its leads. With a 220 µF capacitor, the board starts reliably when the connected battery voltage is around 11 V or higher.

By the way, auto-start can easily be added to the original circuit, too. To do this, connect a 47 µF capacitor (possibly a smaller one; I tested this value) in parallel with Zener diode ZD42, with its negative terminal connected to the transistor.


It occurred to me later that there might be a problem when using lithium batteries with a protection board. If the protection board trips, the relay won't start even when mains power is applied, and the battery won't start charging. One possible fix is to add a resistor-and-diode chain that can "activate" the protection board; after that, everything should work as usual.
Of course, many will ask: why not put a diode between the power supply's positive output and its output terminal—in other words, across the relay contacts—or between the positive supply rail and the junction of diodes VD2 and VD3?
You certainly can, but then reverse-polarity protection is lost. In my original version, the relay only turns on if the battery is connected with the correct polarity. In theory, it should turn off and lock out if the battery is connected while the unit is running, but I haven't tested that.
The resistor and diode won't draw much current or interfere with the protection, but they may start the battery's protection board if it trips.


This is the power supply I ended up with. Here's a list of all the changes compared with the original version:
1. The charging current is higher now, at around 1.75–1.8 A. It could be increased further, but that would make the current-sense resistors run hotter.
2. The charging current is regulated and doesn't depend on the battery voltage.
3. Output power has increased from 65 to 85 W.
4. The battery's diode assembly runs cooler thanks to being mounted on a heatsink.
5. The ballast resistor, which was always consuming power, has been removed. In my opinion, it isn't needed here.
6. The power supply now monitors the battery voltage, making it easier to set up for different battery types.

Possible additions:
1. Auto-start
2. Automatic activation of the protection board.
3. R9 could be changed to 4.7 or even 5.1 kΩ. This would only slightly reduce the battery runtime, but would make the protection board less likely to trip.

Issues I'd like to fix
1. Voltage overshoot during charging. The original circuit has it too, but there the power supply always outputs a voltage more than 0.5 V above the battery voltage; in my version, this only happens at the end of charging.
2. A large voltage drop across the current-sensing circuit during charging. It's not critical, just wasted heat.
3. A large voltage drop during discharge, which is also present in the original version.

All these issues can be fixed by changing the current-sensing circuit. That would also change the shunt value and eliminate the diode assembly. It should be possible to get the voltage drop down to around 0.1 V from the current 0.4–0.6 V—four to six times lower. But this would require adding an op-amp.


By the way, I just remembered that I have a Chinese UPS that also uses a simple charging circuit. Unlike this version, though, it has an extra winding, so a proper CV mode can be implemented. In other words, even that one has what I consider a better charging circuit.



Conclusions.
Honestly, Mean Well makes excellent power supplies, but did a poor job with this UPS. They built a great platform, then ruined it with a primitive charging circuit. This was all fixed in the PSC-160, but they set the charging current to 4 A, which is a good fit for lithium batteries but not so much for lead-acid ones. For those, you'd need a capacity of at least 25 Ah.
I don't know how the PSC-35 is designed, and I'd like to take a look inside. But judging by its specified charging current of 0.9 A, I suspect it's the same as the PSC-160.

The task is complete, and there's enough information here for anyone who wants to replicate the modifications. In the future, I plan to describe converting another 12 V power supply into a UPS, and I also have a Chinese low-voltage UPS on my bench.
And one question: is this a topic worth exploring further?

That's all for today. I hope this information was useful. I tried to keep it brief, but I also wanted to explain things in more detail along the way, so it turned out long—sorry.
This page was found when searching for:
,kjr gbnfybz фв-55ф ecnfyjdrf, ad55a схема подключения, блок питания ада 12 е 400 как сделать регулируемый, 2, meanwell ad-155 подключаем к акб на 12в, делаем источник бесперебойного питания из компьютерного блока питания, ad-55a для чего он нужен, 7, 3, плата усилителя для сабвуфера как подключить разъем для наушников если его нет на плате f50-00318-030 маркировка, источник питания ad-55a, https://www.kirich.blog/obzory/bloki-pitaniya/1130-meanwell-ad-55a-bp-s-funkciey-istochnika-bespereboynogo-pitaniya-ili-kak-iz-ibp-sdelat-ibp, плата для подачи энергии из одного из двух источников, ибп melan well ad отсутствует питание 13.8, mean well ad-55a, ad-55a какой аккумулятор подходит, схема подключения ad-55a, mean well ad-55d, сделать usb 5 вольт в ибп, ad-55a как подключить, 1

Questions and answers

What are the specifications of the Meanwell AD-55A?

It is a 13.8 V, 55 W UPS with no additional 5 V output. The author measured up to 70 W without a battery, but with a battery connected the output voltage began fluctuating above a 4.8 A load.

What is the Meanwell AD-55A charging current, and what is the issue?

Its stated charging current is 230 mA. The author describes the charging circuit as basic: voltage varies linearly with load current instead of operating in constant-current mode.

What did the author modify on the Meanwell AD-55A?

The request was to increase charging current, add automatic startup when a battery is connected, and check whether output power could be increased. The author says the requested task was completed; the supplied excerpt does not give the modification details.

The figures from this review are in the measurement database: Power supplies

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

Comments are published in their original language.

  1. 16 February 2023 22:53#1
    Тему развивать стоит.
    Если, конечно, у Вас на это время есть.
    И обстановка позволяет.
    Спасибо за обзор!
    +5
  2. 17 February 2023 00:21#2
    Супер. Дуже цікаво і на перший погляд дуже проста переробка :)
    0
  3. 17 February 2023 07:05#3
    Спасибо за интересную статью!
    Тема очень интересная и актуальная, ждём продолжения - и про 12В и про низковольтный!


    0
  4. 17 February 2023 15:27#4
    Ещё раз спасибо за очень полезную статью!
    Вопрос по psc-160a
    Можно ли его доработать, чтобы он стартовал только от батареи?
    Из коробки он стартует только от 230в и только потом можно на батарею переключиться. Типа в полях использовать в качестве пауэрбанка..
    0
    1. 17 February 2023 15:44#5
      Заставить думаю можно, но вопрос, а зачем "в полях" бесперебойник? По моему гораздо проще для такого иметь батарею, ни и можно зарядное отдельно, в сумме будет дешевле и легче.
      Не совсем понимаю смысл данного варианта использования.
      0
  5. 18 February 2023 05:07#6
    Да он может понадобиться в таком режиме раз в год. Не думая, взять и использовать,  в нем лишнее для поля только блок питания.

    0
  6. 19 March 2023 12:52#7
    А Ви досі травите хлорним залізом?

    Є ж кращій рецепт - лимонна кислота + сіль + перекис водню.
    Компоненти доступніші та дешевші, на одязі не залишає плям (принаймні свіжий розчин - він прозорий, відпрацьований синьо-зелений), травить так само.

    Щоправда, розчин не зберігається, готується перед застосуванням.
    +1
    1. 24 March 2023 21:57#8
      оглосите пропорции плиз.
      +1
  7. 25 March 2023 10:23#9
    Цитата: emdskar
    оглосите пропорции плиз.


    Лимонка та сіль приблизно 1:1 по об'єму (міряю ложками), залити перекисом до розчинення солі та кислоти.
    Перекис з аптеки (водний розчин).
    Насправді оптимальне з точки зору хімії співвідношення лимонки та  солі трохи інше (хоч і близьке до 1:1), але я його забув. Рецепт читав на одному російському форумі, але лізти туди шукати зараз просто огидно.

    Концентрація перекису менш важлива, бо перекис не реагує безпосередньо з сіллю та лимонкою, а діє на мідь, окислює мідь атомарним киснем, який виділяється на поверхні металу при розпаді перекису (а оксид міді вже реагує з іонами хлору та цитратним кислотним залишком).

    PS Я той ще хімік, можу помилятись в розумінні хімреакцій )
    Але рецепт робочій, перевірений. І спиртовий фломастер не розчиняє.

    0
  8. 6 May 2023 16:39#10
    Можете выложить файл платки
    +1

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