The idea is to have an 8-channel power amplifier in one box …
I still have to source the right trafo, but one large one would be both lighter and cheaper (new) than, say 4 small ones. An argument to use multiple would potentially be that peaks on individual channels might have less impact on the other channels. Still working on it …
Disclaimer: not at all an expert on this, this is just my thoughts based on what i’ve heard, please correct me if i’m wrong
Normal power supplies are CV, so you don’t have to worry too much about that, CC supplies are a bit more niche
If you want to be able to run the amps at full power, you’ll have to add up all their rated powers and find a psu that is rated for that, it could also be wise to go with one slightly larger as well, as power supplies tend to get a bit noisy when they’re close to the limit (the voltage they give out gets noisy that is) which reduces the audio quality
what are the input for the amps? You say they have built in rectifiers and everything? Can they be aupplied by mains directly? What is the max input voltage?
There will be loss in the process so you should go a little above. You also need to account for the efficiency curve of your power supply: is it best efficient at 80% load? 90% load? Can it handle 120% momentarily in case of a spike?
CV power supplies are the standard: constant voltage. It outputs say 12V, and trips when overcurrent. A CC supply would limit current to say, 20A. It does so by dynamically adjusting the voltage output to match that target. That’s a lot less common and usually used for battery charging or testing/troubleshooting. So, I guess, don’t plug it on a battery charger.
It should come with specs as to what input it can take. Follow the recommendations. If it says DC give it DC unless you’re absolutely sure of the circuit in there. The presence of a rectifier and caps doesn’t tell you much given it’s an amplifier, it could be part of the amp circuit for the MOSFETs and not its power supply.
Well, it says to use a toroidal transformer between 12V to 32V output, and looking at pictures of other people who bought it, it seems to work on AC directly, so I have high confidence that that’s what it is.
Don’t think I can solder on plastic, plus it would create new thickness and I don’t think that would help some sandwiched membranes, but I’m going for double sided copper tape, that should help I hope.
Do you have a multimeter? Or other way of gauging the resistance of the new traces?
I don’t know the specific product you’ve used, and my experience with conductive paints and glues is almost nonexistent. But what I remember is that it was neither useful as a glue or a conductor. So I suspect that the resistance of the trace is too great to be used for traces going into the 100s of mm.
If my suspicion is correct, then maybe you can fix it by using the paint to attach something with little resistance in parallel to the paint traces. Maybe stripping a multicore wire and using a single strand of copper, lay it down on the trace and paint over it? Or cutting the traces out of tinfoil and gluing them down to the existing traces with some of the paint?
Copper or silver-based should be lower resistance. These conductive paints tend not to be very conductive, the carbon stuff is essentially making a thin-film carbon composition resistor. Good for repairing rear window defroster heating elements, not so great as a 0-ohm trace in a keyboard. For short (<1cm) wires it’s usually not too bad, but with the amount of damage I’m not sure you’ll be able to repair the thing.
It looks like it might be from a Model M-style keyboard. Unicomp sells those.
Unfortunately the shipping fees would kill me since I’m in Europe, I really wish I could buy a new membrane but it seems like I need to do the repair myself.
Would copper tape suffice as a low resistance trace?
Worth trying. It’s already broken, you can’t really make it much worse. It’d probably work, and worst case you’re back where you started & paying for expensive shipping.
I don’t have any improvements to offer but I want to mention that I saw a crazy video that used a circuit breaker as a spot welding device. The concept is pretty much right and it is as cheap as it can gets, but it is electroboom kind of crazy and the guy even hurt himself on camera while using it.
He burnt himself off a 12v battery. Increase the voltage by the amount you want to live dangerously at the benefit of getting better weld. It also suggests using two MCBs in series in case one of them does not trip for any reason.
I’m slightly confused by your question. Are you trying to power two thingies off a twofer adapter, or are you trying to multiplex something?
It sounds like you’re just trying to do power so:
See how much each strip draws (either via data sheet of amps per meter times 0.69, or measure it) and if that amount exceeds the capacity of the adapters you want to use, it’ll either not work or shorten the life of the gizmos (or if it’s really cheap crap or you’ve gone way the heck over the capacity of any component it might catch fire a bit).
Note the current on the datasheet wont be linear per length, but for 27” chonks if should be close enough.
Edit: I see now your LEDs are 3V. Personally I’d recommend replacing them with 12 or 24V, or at least 5V to run properly on the 5V thing you’re plugging them into. Why did you go with 3V product?
I am assuming the LEDs are white (blue with yellow phosphor), which always have a roughly 3V voltage drop, this is just a physical fact. Some chips like those in LED bulbs have several LEDs in series for a voltage drop of 6/9/12/15/18 V but this is not the case here.
The vast majority of 5V strips have no step-down switching power supply (aka constant current buck converter) to reduce 5 V from the power rails to 3 V, instead just driving the LEDs with a resistor in series – it drops 2 V and if it’s a 100Ω resistor (usually labeled “101”), it lets 20 mA to the LED as per Ohm’s law. In practice, multiple LEDs are often in parallel to one resistor to save cost, in which case the current divides among them.
The strip may be RGB, in which case the LED voltages are 🔴1.8 V 🟢2.4-3.0 V 🔵3 V and pretty much the same applies. There may be an external controller but it usually just uses PWM to pulse the 5V rail of each color rather than adjusting the current. Individually cntrollable LED strips have a chip driving each LED, and there is just one power rail and serial data line between them.
Wow awesome, thank you for the info and resources! I have to follow the links later but I really appreciate the reply.
I don’t think OP’s strips have pixel control since in my experience you wouldn’t have a separate USB power line that bypasses the controller like OP is describing, but now am curious and hope they follow up.
The strips use 3V white LEDs, power to which is delivered via resistors or linear current regulators. Unless you see any inductors, there is no buck converter from 5 V to 3 V.
Why does this matter? Well, with resistors and linear regulators, 𝐼in (current in) pretty much equals 𝐼out (current out). So the efficiency 𝜂 = 𝑃in / 𝑃out = (𝐼 × 3 V) / (𝐼 × 5 V) = 60 %. Extra cable resistance will reduce the current, brightness and power, but still exactly 40% of power leaving the USB charger will be wasted before it gets to the LEDs.
However, I would advise against multiple, cheap USB connectors in the circuit: when moving the setup, their resistance changes somewhat and you would get blinking. The worst thing that could happen is a switch, such low voltage cannot spark over an oxide layer and eventually even small movement will blink the lights. I would get a good thick USB cable and solder it directly to one of the strips instead of whatever it came with, connecting the other with some thick wires.
So it does not matter, if you want better efficiency, use 12V (75%) or 24V strips (87%), or get just an LED array without resistors that needs a constant current driver (theoretically 100% but CC PSUs are slightly less efficient). Or make a constant current driver by fine-tuning the voltage of a PSU (by adjusting the feedback resistive divider) to 0.5-1 V above the LEDs’ voltage drop, then using an appropriate resistor to limit the current.
which current and voltage do the LED strips need (5V, 2A would be something you could support with USB 2).
what output does the charger provide as a maximum? Is it enough to power both strips (5V and two times the A).
If these match it should work. Another topic could be the cable itself. Theoretically it could start to burn, if you try to channel to much current through it, but with simple USB I doubt it. If it is getting hot after some time, scrap your setup. This would be a fire hazard.
Fire hazard? Not really. The charger will not provide more than 5 V, 3.1 A or whatever its rating is. Even if the strip fails short circuit (unlikely, most LEDs blow open and there is at least a resistor in series anyway), nothing will receive more than 15 W of power. A 1m cable can safely dissipate that.
Worst case scenario, an LED fails short and its series resistor will receive over 6x its intended power (with 5 V across it instead of 2 V, and its current will increase correspondingly). It is soldered on power-dissipating flexible PCB and will probably not blow open. It will get the area somewhat hot and potentially melt the plastic in the back of the monitor while the rest of the strip keeps glowing but more dimly. Hard to tell if it could get over flammable temperatures with moderate heatsinking and only a few watts of power.
To keep safe, I would deliberately add resistance before the LED strip, using something like a 1Ω 5W resistor (or a shitty long cable). That way, the voltage drops significantly in case of a short. Also, the LEDs will run at lower-than-intended current, which prolongs their life and decreases risk of failure.
Edit: Some microwaves have a 0.8Ω 25W resistor as part of inrush, at least in 230V regions. Feel free to use that, it will happily handle a semi-short circuit. Or you know, an automotive fuse.
The safest option is replacing the whole setup with an LED strip that has no resistors (bare LEDs) and a constant current driver.
The LED strips use SMD 3528 LEDs which need 5V and the wattage is listed at 11.52W/min. The amperage isn’t listed but for those LEDs, I’m seeing 5Ah online. The charger provides 40W
You’re confused. W/m means watts per meter, and the “5Ah” is probably actually 5 A, or the current you can push through the strip (limiting the length to below 2 m).
Watts in a resistive example like yours is Volts x Amps. I would have been able to much better answer this question a year ago so forgive me if I’m misremembering the specs but I’ll answer since nobody else has. Two things that suggest to me this might be a bad idea:
Charger is 40W, that’s probably usb PD (I don’t know anything about QC so maybe I’m wrong). PD supplies more than 15W (5V x 3A) by stepping up the voltage, not the amperage. While stepping up either would likely be bad or very bad for some part of your circuit, don’t worry about that though; without the powered device telling it to, PD won’t activate. It should max out at 15W… I think. It depends on the resistance on the CC lines and using a splitter could screw up the resistance that tells the power supply which USB version to support so it can go up to 3A (15W). Sorry, it’s been a while since I’ve worked with USB power. 2 strips of 11W will need more power than that. Basically my concern is you won’t get adequate power out of the charger for one reason or another.
Where are you getting the 11.52W/min number? Watts don’t have a time unit and that much precision sketches me out. Almost as if someone who isn’t adequately educated measured the power straight off a multimeter once and just wrote that on a product page. Is the LED strip from a reputable manufacturer?
He meant watts per meter, not minute – the strips can be cut and rejoined. As for the 5 Ah, no clue.
The circumference of a monitor is more than 1 m, so a charger of 3 A at least will be necessary for each. This is why I prefer higher-voltage strips where less current is required and higher resistance is tolerable. Anyway, the power is quite high and this could cause overheating problems.
Sam Zeloof (sam.zeloof.xyz) has been working in the area of DIY semiconductor fabrication, and has successfully fabricated working ICs (from 6 - 100 transistors). He’s done an incredible job at reaching ~1970’s technology levels in a home lab.
These examples are still very small-scale compared to even simple microprocessors, etc - but it’s fascinating to see this level of technology becoming more accessible.
If it’s a 240v transformer, the primary will have less turns than a 120v one. You may not have a high enough voltage out of your new secondary because of that.
If it’s just a matter of getting a larger voltage, then you already got 400V. No need for new windings, unless the insulation on the primary can’t handle 400V.
Just in case you really didn’t understand: it’s wave, not vawe. It’s a common spelling mistake, especially for people whose native language doesn’t have the w letter.
You might be interested in the YouTube channel ProjectsInFlight, which is currently trying to build a DIY solution for fabbing simple ICs in their garage & documenting the process on YouTube.
askelectronics
Hot
This magazine is from a federated server and may be incomplete. Browse more on the original instance.