Forum - Recent Threads

Safety of plugin solar?

Workshopshed — Mon, 05 Jan 2026 07:36:14 GMT

Just seen a YouTube short from JerryRigEverything for a 4 panel solar system with roof top microinvertors.

The "interesting" feature is that it connects into a regular socket i.e. the power is provided through a plug.Jerry states this is only approved in Utah at the moment and is limited on the size of the system you can install.

For me this raises a couple of questions, firstly how it doesn't electrocute the person handling the plug? But how it doesn't back power the grid when there is a power outage?

I have a solar system at home and it is hard wired with a number of protective devices and isolators between the solar and mains.

Anyone know anything about these plugin systems?

ATX power supply control interface with a PIC18F1220

Stuart_S — Wed, 22 Oct 2025 01:01:22 GMT

Here's a little circuit I've developed to operate an ATX PSU with a single push-button instead of a toggle or slide switch.

The heart of the circuit is a few components including a diode and a transistor. I'm wanting to power the MCU from the standby 5 volt line, can anyone here advise me what current this pin is normally able to source and is it usually held stable or does it power down periodically?

For those who'd like to understand more about the circuit I'm providing a link to my Wix project page: ATX power supply remote start with single push button. | Blokeystuff The key to the circuit is how D1 is reverse biased in both the states (while the PB is open) via the higher resistances to Vcc and the lower resistance paths to ground to counter the reverse biasing when the PB is pressed -and at the same time when the uC gets a low input on uCin_btnPSU it then runs an internal timer when it's in the ON state to decide when to release PS_HOLD. The push button does both ON and OFF control in the same way it's done on an IBM AT-X on the front panel. Ignore the NTC for the moment, I was using that to help regulate the PWM signal into a salvaged HP desktop PSU.

Is the "imminent" availability of sodium ion batteries affecting your decision of when to switch to an EV or solar house?

dougw — Tue, 09 Sep 2025 20:32:13 GMT

Sodium ion batteries seem to be pretty close to being available and lots of investment is being made into the technology. They promise very fast charging time, 10s of thousands of charge cycles and lower cost than Lithium.

[Please visit the site to access the poll]

Leave a comment to elaborate on your thinking.

Looking for a hardware recommendation on USBC power analyzer

colporteur — Wed, 07 May 2025 23:38:27 GMT

I'm looking to the E14 Community for a recommendation on USB C Power Analyzer.

A small dongle that plugs into the USB C port a provides power analysis. A google search provides a number of hits. I'm not looking for the unit in the picuture but hoping someone in our community can provide a recommendation to make a purchase. Can you share a link where I might purchase such a beast.

Gate protesction against undisired switch on

finoconv — Fri, 11 Apr 2025 15:40:47 GMT

I'm designing a Marx generator with solid state switch. I'm using IGBTs.

Each IGBT has a 160uF capacitive load. At the moment I'm testing a generator that is powered with a 1kV and has 3 stages: the output voltage should be 3000V for a resistive load of 100Ohms.

Every stage of the marx is constituted by a half bridge with the gate isolated. Firstly I tested the same project with 6uF per stage and the system worked. When I try to discharge the Marx generator (with 160uF per cell) at the maximum voltage, I made the huge error to not connect any load. I heard a big burst and I noted that in one stage, the tracks of the half bridge were burned and the IGBT were faulted.

Apart from the huge mistake of not connecting the load, I'm thinking to protect the gate oh the high IGBT form the undesired switching on, due to floting voltage between the two IGBT. I saw in a marx generator (I am copying it in full) that there are a series of zener voltage and schottky diode between the gate and the emitter. Chat GPT says me that this should be a possible solution to avoid undesired switch on and protect the gate from overvoltage. It's not very clear for me how this could work.

Has anyone use it before? Any suggestion about the protection against undesired switching on for both IGBTs?

Question electrolytic capacitor

cloudff7 — Mon, 07 Apr 2025 12:40:13 GMT

Manufacturers of aluminum electrolytic capacitors in storage recommend energizing these components once every 1 or 2 years and the temperature range is 5-35C and humidity is below 75C. After this interval, under these maximum conditions, they recommend refurbishing.

Do these recommendations only work on new aluminum electrolytic capacitors with little or no use because their chemistry is new and unchanged, or does this recommendation also work on old aluminum electrolytic capacitors that are used and assembled in electronic circuits and manufactured in the early 1990s?

Recommend a bench power supply

stanto — Wed, 19 Feb 2025 13:58:25 GMT

Hey all,

I am interested in a bench power supply (I think that's right?), I am often working with electronics, such as repairing tile tags, game consoles, phones, computer kit, working with microcontrollers and I would benefit from having a power supply that I can use for this.

I had intended to modify a pc atx power supply, but I simply haven't gotten around to it and I'm often in the position where I want a tool, not a project.

Bonus if it can do usb-pd.

Any recommendations? I'm not a professional or run my own business, so it'll have to be something available to a consumer in England.

Thanks.

LDO Linear regulator for lead-acid to 12v?

Workshopshed — Thu, 13 Feb 2025 13:45:22 GMT

I'm planning to power some low current devices from a sealed lead acid battery that require a 12v supply. I was thinking of using an LDO linear regulator to provide the 12v. And to have some kind of voltage monitor to predict battery life.

I don't have the exact discharge curves for the one I'm using but it does seem that once the voltage drops below 12v we should stop and recharge them anyway.

Is this a reasonable assumption or should I be looking at a buck-boost dc-dc instead?

Is there any clean energy challenge where I can win solar panel as a prize

manojroy123 — Thu, 24 Oct 2024 08:07:48 GMT

Is there any clean energy challenge around internet where I can win solar panel as a prize.

Troubleshooting a PCB

uxphil — Sat, 19 Oct 2024 06:42:03 GMT

Hi, hoping to get a little help here with a basic PCB trouble shooting question

I have a kids toy here, which runs of 3aa batteries

it’s a roof of a train station that has two buttons and should make a noise when either is pressed

I get some noise, but not at all the right ones

i took it all apart and found this PCB.

i suspect one of the elements it busted.

question, how to I check whether the following are working

- capacitors

- resistors

- diodes

I’ve tested the speaker and it works

any other tips on how to troubleshoot this with an multimeter?

What order should I test in? And which settings?

thanks for any help!

Phil

RF Noise - Do you notice the RF noise floor has greatly increased with the influx of wall warts?

Cquk — Tue, 15 Oct 2024 17:54:46 GMT

I would be interested to see if any one else has noticed the issue. Maybe if you work on RF devices / have a spectrum analyiser etc.

Even some bench PSU's seem to produce more EMI than i would expect.

I am working on adding Ferrite's to all my cables but where does it stop?

What would you like the upcoming Fluke webinar on PV and Solar (Nov 13th) to focus on?

JoRatcliffe — Fri, 11 Oct 2024 14:09:04 GMT

Hi everyone, just cross-posting from the Members Area, I thought it made sense to share here as well. I really appreciated all the feedback on my last event-related poll discussing the Arduino Opta webinar which took place yesterday, so I wanted to get everyone's thoughts on the next webinar we have coming up:

Tools and Techniques for Commissioning and Maintaining Photovoltaic Systems which is being hosted by Fluke on November 13th next month.

We have an agenda for the event:

What tests are conducted during the lifecycle of a solar installation.
Ground faults and how to use voltage to locate a ground fault in a string of solar modules.
Category and voltage ratings and how they apply to selecting the proper test tool.
A live Q&A section where you can put all your questions to our host from Fluke.

The Q&A section proved very popular yesterday so I wanted to ask you, if you're planning on attending would you like us to reserve time for a longer Q&A session at the end? (ie. 15-20 minutes) Or would you rather the session focus on the core content?

[Please visit the site to access the poll]

Plus, of course, let me know if there's anything you'd like the session to include or have addressed, share your thoughts and I will pass them on 🙂 Have a good Friday!

Can someone provide me with teardown of chassis mount resistor tjt300 10R from TE

manojroy123 — Sat, 28 Sep 2024 11:04:46 GMT

Can someone provide me with teardown of chassis mount resistor tjt300 10R from TE. What kind of strip is used inside it.

Does anyone knows. How a duracell powercheck feature works ?

manojroy123 — Fri, 27 Sep 2024 14:36:48 GMT

Does anyone knows. How a duracell powercheck feature works. I need complete detail about it. like what is the current consumption, resistance of the strip etc.

For RFID/NFC Antenna Vin: If there is a 5V supply apply this voltage otherwise use 3V3

BigG — Mon, 01 Jul 2024 20:11:01 GMT

I am back looking at RFID/NFC modules and I am grappling with a basic circuit design challenge.

As some of you might know the NXP PN5180, the PN7150 (NRND) and the PN7160 controller chips allow you to apply a 5V input to antenna, which boosts performance. But it will also accept a 3V3 supply.

So I am trying to come up with a circuit that will apply either voltage to Vin_antenna but 5V will always take preference. As in, if 5V is detected apply this voltage, otherwise apply 3V3. This means that 5V input can be optional.

I had always had in my head that this circuit would work, but I clearly hadn't studied the circuit closely enough until very recently, when I realised that this circuit applies a VBUS charge to VBAT, if VBUS is detected. Not quite what I wanted. But then maybe I could simply insert another diode to ensure one way traffic on the VBAT or 3V3 line.

(source: https://cdn-learn.adafruit.com/assets/assets/000/130/918/original/adafruit_products_schem.png?1719573412)

Then, I discovered the tiny LM66100DCKR integrated diode IC. I think this potentially could work but I was not sure how well it would work with the 2 different voltages. I just don't fully understand the ins and outs of this IC. I just want to ensure that the 5V bus took priority over 3V3.

https://www.ti.com/lit/ds/symlink/lm66100.pdf

(see section 9 of PDF).

So, any thoughts and suggestions greatly appreciated. Thanks.

DC-AC power amplification

Maxbiggair — Wed, 29 May 2024 18:32:57 GMT

There is a kind of converter I need help on. It is a buck-boost converter but the inductor is replaced by a steel toroid core transformer with a core power of 2025W max, it has a primary input and two secondary output 250 turns 1:1:1 each.

The transistor is changed to a MOSFET and is driven by a 555 timer pwm circuit. Powered by a 12V, 75AH Battery

Flyback voltage is tapped out from the primary coil and the high current is tapped from the first secondary coil through the use of diode and capacitors.

The high voltage and high current is combined and fed to the inverter converting it to 50Hz AC. The output of the inverter is connected to the first secondary coil. Which is the same coil that the high current capacitor is drawing current from.

The second secondary coil will be the output whereby a band pass filter will be used to extract the 50Hz frequency from the transformer.

The goal is to provide good efficiency while reducing the cost.

negative 48Vdc system to positive 48Vdc

Lemuel7 — Wed, 06 Mar 2024 07:41:20 GMT

Hi, I have a situation where the existing 48Vdc system is a positive grounded (-48 or negative hotwire), now I need to replace it with a new 48Vdc charger that is earth ground isolated and its output is already negative volts (positive to ground=0V & negative to ground=0V and positive to negative potential is negative 53V)

At the DC combiner panel, positive busbar was grounded and that is where we will connect in order to parallel the new system to avoid loss of dc when removing the old system.

I understand that both system outputs have different polarity hence I cannot just simply parallel them (positive grounded system does not cause polarity to change) . How do I safely remove the old system without interrupting the loads and not to loss the 48Vdc?

Visual Signs of Capacitor Failure - Bulging

Jan Cumps — Sat, 24 Feb 2024 15:19:48 GMT

I received a new TV box from the cable provider They suggested to drop the old one at an electronics recycle collector. That didn't happen 🙂 - I opened it up. I've put the internal hard drive for sale on a 2nd hand site. For the other components, I'm checking what (if anything) is worth keeping.

What caught the attention, is that one of the electrolytical caps is beyond end-of-life. It's definitely bulging: It's the first time that I see one where the symptom is so obvious. I had others that failed or leaked. But never such a visible sign of the internal pressure effects.

Can you spot the culprit?

Title inspiration: capacitor labs.

Poll - Do You Design, Maintain or Build PV Systems?

rscasny — Wed, 14 Feb 2024 18:09:55 GMT

[Please visit the site to access the poll]

Teardown Photos: Corsair RMe Series RM750e ATX Power Supply

scottiebabe — Sat, 27 Jan 2024 15:41:50 GMT

According to https://cultists.network/140/psu-tier-list/ the RM750e is a

Tier A • High-end PSU

Is it the best PSU ever? I don't know... You can decide that for yourself.

I do like the general layout and design to envelope. They made a great use of available volume with daughter cards everywhere!

The first EMI filter stage is nestled underneath the input IEC connector on a littler carrier card.

Cybenetics has a component breakdown list: https://www.cybenetics.com/d/cybenetics_9Nq.pdf

Maintain power long enough for SBC to power down

obones — Fri, 26 Jan 2024 10:48:50 GMT

Hello,

I have connected my 3D printer to an Orange Pi Lite running Octoprint as it is much more user friendly when it comes to managing my prints.
This works just fine but required an additional power source and so I opened the printer and connected the 24V PSU output to a step down converter that provides the 5V for the Orange Pi.
So far, so good, everything is powering up and running properly.

What I'm having an issue with is the power off sequence because when I turn off the printer, the OrangePi is immediately deprived of its power and does not have the time to shutdown properly.
This may lead to some corruption on the file system and most importantly OctoPrint thinks it has crashed and thus starts in "Safe mode" upon the next boot.

First and foremost, I would need a way to detect the sudden disappearance of power and inform the OrangePi. Looking around for existing resources, I came up with this arrangement using parts that I have sitting on the shelf here:

As soon as the 24V is gone, Q1 stops conducting and the Orange Pi sees a logic 1 on one of its GPIO pins.
But I'm not too confident with this because when I usually such an arrangement, the base voltage is usually equal to or smaller than the collector voltage. Here the base voltage is MUCH higher and I'm afraid there would be a situation where the 24V would end up straight into the GPIO pin and frying the entire OrangePi.
So I came up with this design, once again using parts that I already have:

It feels wrong to have the same ground on both sides of the optocoupler but it gives me a sense of safety in that the +24V will have a harder time going to the GPIO pins in case of catastrophic failure.

What do you people think of those two options?

And while the above would be easily sorted out, I'm also struggling with the second part of the requirements which would be to maintain the +5V power for a sufficiently long time so that the OrangePi could run a script that properly turns it off.

At first, I took the biggest capacitor that I have here : 3300uF

But this failed miserably, there wasn't much delay in between +24V disappearing and step down converter shutdown.

I should have measured the current it consumes because with 250mA it would have been obvious this capacitor is too small.

So using a capacitor discharge curve calculator, aiming for 4.7V after 20s it showed me that I would need about 20F capacity!

That would cost me anywhere between 10 to 20€ which made me look for other alternatives based on what I have lying around here.

First idea was to use a phone PowerBank that would have its charge port connected to the step down converter and its output to the OrangePi. It's not guaranteed that its output is active while charging, but even then I have a feeling that it would quickly kill the battery to source so much current for so long.

Second idea was to use a 500mA LiPo that I have along with a 3€ controller board like this one:

While I know its output is active while charging, it still has the same potential issue of shortening the battery life.

This is why I figured I would need a way to OR the step down converter output with the battery step up converter output.

Easiest way is to use Schottky diodes on both power sources, so this requires 2W diodes and a careful setup of output voltages so that the step down one is higher than the "battery" one to ensure the step down one is used preferentially.

Reading about ORing power sources, I stumbled on Ideal Diodes chips like the LM66100 and LM66200, the latter being a very good candidate as it even shows exactly my use case in its datasheet. Getting my hands on one and being able to solder it is something that I have not even started to consider, though.

In the end, I'm left with those choices:

20F super capacitor
Inline powerbank / LiPo battery with management board
Side line powerbank with Schottky diodes
Side line powerbank with Ideal diodes

Which option would you recommend?
Do you have any other suggestion?
Are there any potential issues that I did not think of?

Many thanks for your help

What is the negative 12 volt supply used for in an ATX motherboard?

scottiebabe — Thu, 25 Jan 2024 16:25:28 GMT

Technology... Never-ending compatibility problems....

Here is the 24-pin ATX connector of my new power supply

Two of the pins are missing.

Pin 20 was formally -5V and has been omitted from ATX power supplies for a very long time.

No negative 12 volt supply for me! Somehow I missed that looking at the product spec, and no one seem to mention it either.

Here is the ATX V3.0 power specification

https://www.cybenetics.com/attachs/52.pdf

The negative 12V supply is optional... great!

What is the negative 12V supply used for?

I am just going off the b350 chipset motherboard schematics available here: https://github.com/YeaTheMen/am4-motherboard/tree/main/Other%20Board%20Schematics

In many cases nothing.

In one case an RS232 line driver/receiver

In another case, an amplifier/buffer for a second audio line out for the front panel

Something to watch out for. In the two examples above, having the -12V rail tied to ground wouldn't be catastrophic, though the op-amp and rs232 would not function as intended...

Convert SPST (on/off only 2 contacts) switch to short a pair of conductors for 3-5 seconds when “switched” on and off

Dblaise — Sun, 07 Jan 2024 19:15:16 GMT

12 volt DC supply power for circuit/device.

on state to press button”short” circuit

off state to press button “short” circuit

each time for 3 to 5 seconds

Thanks in advance!

Dave

Switch Mode Power Supplies: Power Electronics course at Coursera starts again

Jan Cumps — Fri, 01 Dec 2023 21:34:31 GMT

The Boulder university's Power Electronics course starts again today:

www.coursera.org/.../power-electronics

If you select Enroll for free, then: Audit the Course, you can follow along. You don't get university credits or grades. But you 'll learn a lot.

It's not easy.

Courses:

Introduction to Power Electronics
Converter Circuits
Converter Control
Magnetics for Power Electronic Converters

Expected time needed: 5 days per week, at least 1/2 hour.

LED Worklight and teardown

robogary — Sat, 04 Nov 2023 19:09:04 GMT

My workshop areas have eight 4 foot double tube fluorescent worklights in the man-cave. I installed the 4 foot floursecent fixtures 35 years ago.

As the tubes fail, they get consolidated, and one by one empty fluorescent fixtures get replaced with LED worklights. 5 of 8 fluorescents are still in service.

One of the LED worklights failed after a couple years of service. The failure is weird, the worklight had turned into a strobe light.

community.element14.com/.../Worklight_5F00_strobing.AVI

Maybe the light can be repaired ?

I dunno, but the opportunity to troubleshoot, reverse engineer, and maybe even scavenge parts is too good an opportunity to pass up.

The endcaps are removed from the light fixture. The simplicity is smart. An extruded aluminum tube body, diffuser slides into it, and an LED strip is mounted on the AL extrusion.

The power supply leads were desoldered and the LED strip slide right out.

The LED strip is on a thin fiberglas pcb with 78 COB LED packages soldered in place.The LED strip is labeled with 94v.

Its easy to follow the traces to see the LED strip has 3 parallel circuits of 26 COB LEDS in series.

Since the LED fixture was strobing, the LEDs seem OK, but the power supply would collapse after being able to start.

The power supply was tucked inside the aluminum extrusion, encased in an insulated tube.

The power supply was first inspected for failed components. The parts are really packed in there.

Voila ! Nothing smoked. That black thing looks like its had an overtemp near the body. Its labeled L1, ah...an inductor.

After desoldering and removing, a closer look seems that the inductor had bitty wire leads soldered onto the stiffer component leads.

Ohm check shows its not dead shorted, and LCR meter sez its 1.86 mH. Darn, that seems reasonable.

The inductor probably isn't bad so it will have to be reinstalled onto the power supply card.

Next step is to do reverse engineering on the power supply card to identify components.

Its for sure a 120VAC 60 Hz to ~ 80Vdc (?) switching power supply, and appears to have 2 stages. I'll need to surf for info on switching power supply basics.

If anyone who has alot of design or service experience on switching power supplies, it would be appreciated to share some insights and hints :-)