Benchtop Power Supply Question

[jiroy]

That's the anti-electro i talked about previously :

RESIDUAL CURRENT CIRCUIT BREAKER (RCCB)

Device installed in the distribution board upstream of the electrical circuits, this detects current leaks due to insulation faults and prevents any risk of electrocution by immediately cutting the power to the circuit concerned. A switch handle is used to reset it after the fault has been eliminated.

[stj]

an RCD, good but not foolproof.
it works by balancing an arm between 2 coils - if one side has more current than the other then the arm moves and it trips.
i think the newer ones overcomplicate things using a 3rd coil and an opamp to trip a solenoid

if you connect a transformer to the line the RCD cant see what happens on the secondary side.
in short - you can kill yourself if you get careless with an isolation transformer!

[redwire]

I added a GFCI outlet to my workbench power and it's tripped once, paid for itself.
I wasn't using an isolation transformer (what is everyone rich?) and had line/neutral flipped on the alligator clips doing some live stuff. A small spark and that's all.
In North America these are electronic, use a CT and trip at >4-6mA imbalance and activate the latching relay. Not the hairdryer crap mechanical ones.
You can also find GFCI's in the trash, take them apart and modify. Although they frequently are blown inside, you could make a custom one.

Modern complicated circuit TI AFE3010 but I like the old orginal LM1851 DIP-8.

Old GFCI's get hit with mains transients, MOV's fade out.
I had one GFCI trip but the relay jammed and power stayed on and the coil+SCR burned up badly. It latches power to the coil which is fine for a short time to trip.

[sam_sam_sam]

Originally Posted by redwire

You can also find GFCI's in the trash, take them apart and you could make a custom one.

How would you do this can you please give an example of a circuit for one because I would like to make one for my work bench

[stj]

Big Clive has reverse engineered a couple on his YT channel

[jiroy]

Originally Posted by stj

an RCD, good but not foolproof.
it works by balancing an arm between 2 coils - if one side has more current than the other then the arm moves and it trips.
i think the newer ones overcomplicate things using a 3rd coil and an opamp to trip a solenoid

if you connect a transformer to the line the RCD cant see what happens on the secondary side.
in short - you can kill yourself if you get careless with an isolation transformer!

Good points ; Electricity and Electronics aren't something to take lightly , and I always wondered if someone put some reminders each thread or so . Beginners and hobyists must always be aware of safety and security measures . If we add a case of spilled water or naked feet , it could be a recipe of disasters .
Even a variac or an isolated transformer won't protect you from a big capacitor's lethal discharge . Even a 19 volts secondary circuitry can hold a potential danger .
That's said , I had this idea since the table is in the room's center and future probable cables will pass on ground , so the RCCB is placed now in the lead casing of the room's circuitry .

[Per Hansson]

Originally Posted by sam_sam_sam

How would you do this can you please give an example of a circuit for one because I would like to make one for my work bench

What redwire was implying is you can modify them to be more sensitive.
Not a good idea to be tinkering with such devices and possibly introducing errors.
I would not recommend to build one yourself, they are cheap enough brand new and are designed to save your life if you fuck up too badly.

[redwire]

The GFCI basic circuit allows you to change sensitivity but the hassle is that latching relay in one. I could not find one to buy separately.
Also, the CT's are really good 1000 turn can be used for measuring low AC currents.

I just took a Leviton GFCI stuffed it in a metal box with line cord and that is an inline GFCI I added to the workbench. I have no DIN-rail stuff.
It's a good reminder I need to add a fuse to it as well. I don't need 15A surprises and even a 5A fuse is good for whatever I am repairing.
I might look at an arc fault circuit interrupter (AFCI) but so far these are all only available as breakers, not outlets.

[stj]

i have a din rail box containing an rcd, a breaker and a chinese volt/current meter with a socket on the outside

[eccerr0r]

I really should look into getting one of those cc current/voltmeters, been using my P3 Kill-A-Watt for that purpose but I only have one of them and would like to keep it for measuring real watts (versus VA which is more interesting for debug in general)...

Alas I just use an isolation transformer, though I should just use an RCD/GFCI near my work area. Alas as of now, I think every time I had a GFCI trip, it was a nuisance trip

As for the isolation transformer, I get no indication when it saves my life...

[stj]

it doesnt,
an iso transformer can kill you as easy as mains
they are to isolate the device from ground for your scope.

[stj]

i have this:
https://www.aliexpress.com/item/1005002687372957.html
the oldskool LED version.

although the lcd one is a tempting upgrade

[redwire]

I have an LED energy meter DIN panel mount like https://www.aliexpress.com/item/10000138101252.html
But I don't like it because it has such slow averaging, it takes 3-4 seconds to ramp up to its power reading. I guess that's OK for fluctuating loads. Otherwise it works really well.

Ali DIN-rail circuit breakers looked a bit hokey - they seem to need 3X rated to trip fast, never mind they have no safety approvals. They seem suited to motors etc. I would want a fast trip on a workbench, so I think a glass fuse is best.
edit: In North America we don't use DIN-rail for the service/consumer unit panel, and those breakers look really slow.

Others I look at and have used:
Tyco W58 series (thermal) but 0.5-3.4 sec fastest response at 1,000% rated current.
Tyco W28 series (thermal) but 0.12-0.3 sec fastest response at 1,000% rated current.

[stj]

the chinese breakers are probably "B" grade.
there are 3 or 4 speed-grades - B is often used like you say for motors and other high-inrush stuff.
you need "B" for a lot of the battery-tab spot welders!

btw, a fuse or breaker should stay intact at 100% current indefinatly.

[redwire]

Did you ever trip one on your workbench, like a B? That's 3-5x rated before they trip so a 3A breaker trips at 9-15A according to the charts.

I looked at mini circuit breakers MCB DIN-rail and they all seem to be about protecting wiring. So they are looking a bit slow and extreme in the magnetic trip part of the curve.

[eccerr0r]

time to go with electronic fuses (a triac with current sense that cuts off drive when current is exceeded by the tiniest of margins, even for a moment?)

[redwire]

I once worked for a company that bought a bankrupt company that had made products with electronics in the pipeline monitoring industry.

While we were boxing things up, workbenches an absolute pigsty, I got blasted by mains line cord someone had left plugged in with bare wires sticking up, in the pile. Guessing a disgruntled ex-employee did it. So I got my hand basically copper plated from the vapors.

So I kind of like a workbench with less than full mains current for whatever I am working on or repairing.
A 5-10A fuse is looking pretty good right now because they can pop fast and it's an uncommon happening for me.
These B circuit breakers I have not worked with, but they seem a bit slow.
An electronic fuse could work but fixing SMPS etc there is inrush current so I would expect it to nuisance trip.

[stj]

fuses are slow - read some datasheets

[eccerr0r]

unless you get those $ARM$LEG ultrafast fuses like they use in DMM's
I think I'd prefer the nuisance trip over having to pay another $ARM$LEG for another fuse, alas right now if I expect a short, it's a DBT that takes the place of a fuse...

[stj]

they are only $ARM$LEG if you want 1KV version,
i buy the 600v ones for about 2-3 $

think about it, are you *realy* gonna be reading current above 600v?