Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Safely monitor 300v 10A current

jrichemont

New Member
I have added a mains conditioner to the ring that powers my PC and AMP. I'd like to log (since the fuse has cut out a few times prior (hence this project). Measuring L-N safely is easy; juse use a little torroidial transformer to drp 250vRMS to 6v at 10ma and fiddle with resistors to get the voltagew right. Fully isolated.
Measuring the current being taken by the load safely is not so easy. I can put a low reistance hight current shunt from Lout to the load and this will give me a few mv/ma but still at 250vRMS which is a bit too close to turning my Arduino to a film of solicon oxide. Does anyone have any idea how I can measure current (16A circuit) at 250vRMS withoutuning resistor bridges (right voltage if all works but not isolated ftom HT. I don't think a transformer will cut it because either the tiny voltages/currents across the shunt or I could use one ratedat 16A 250v but that will drop the output to the load significantly; it's supposed to be the load after all, not part of it.

Any bright ideas welcome. Or I just stick to measuring the voltage - but I'd awfully like to know current use too. The idea is to use an Arduino or similar to log these 10x/s even if the power cuts out (microelectronics on battery) so when I restore things I can look back at just what went wrong. Spike on the input? Running the load too close to limits? Who knows but with 24 hours of good data I should be able to find the issue; I do need as much data as I can collect though. Without a visit from the electric utility, preferably.

Many thanks;
Jeremy
 
Analog Optocoupler ....


Some form of cal will be needed.

Or a digital coupler solution with a processor on source side, possibly it gens a pwm
that can be filtered on observer side to recover DC value proportional to pulse width,
the pulse width proportional to micro RMS or peak measurement.


Regards, Dana.
 
Last edited:
Dear Dana, I', afraid your experetise exceeds mine and I have not really understood your proposal. I have tried to draw what I envisiage (I hope the link to my crappy photo works or please ask for it another way)./ I understand everything needed - the toroidial is fully isolated so either secondary may be connected to zero vots and the other used to ADC input. I halved the voltage just on general principal and besides one resistor can be made variable giving me exactly the voltage I want.
Actually, T! is supposed to go on the other side of the conditioner - I want to measure the improvement - but there is an argument for collecting both sides to compare. My wallet may not permit this though. Still, the 'good' side is better then the bad side.
Because the secondaries of T1 are isolated either may be connected to VDD and the other to the ADC input. The voltagle may be fiddled with and since the current is so low I don't think I need concern myself. An ADC should not really draw current anyway but there's plently more resistors if needed to rewduce it.

This bit is clear bto me. Next wew have the mysterious component X which is supposed to take the mini voltagle (1ohm at 300v at 4kv worst is still not going to produce much voltage differential. S1 may get a little warm though. Map havew to rethink the 1ohm part. My problem, anyway.

What I can't get my head around is both ends of s1 will be at 300v (I gave up writing 250cRMS. The night is old). And I do not want 300v anywhere NEAR my digital loggger. I don't think another transformer wouild work - 1mv at 1ma will just vanish unless I make the primary the shunt in which case I have just scuppered the whole point; there will be no power left for my poor load!

I need something that I can tie one end on to VDD without, er, exothermic disassembly and the other must have a sensible voltage differential of 3-5v. Aslo without the 300v part.

I can, of course use a bunch of resistors to step the voltage down ie 300v -> 100k > vout -> VDD and the same at the other side of s1 but my maths is too rusty to do more that intuit that thgis is a) dangerous and b) not likely to be both safe and produce much more than 0v at 0ma by way of output.

I need your superiour intelligence badly!

PXL_20250122_040858668.RAW-01.COVER.jpg


ps: The system (minus monitoring and shunt etc) is running now seeimingly without problems so I assume my 4kva was a wild overestimate. Of course, the PC states it's drawing 30W (CPU) and all drives are low current SSD/NVMes, only the graphics card is hungry and reports a TDP of 320W but I am typing this, not playing graphics intensive games but even so I think we can estimate 500W all in. I do have a 14Tb NASS but that's also all ssd and I guess uses around 10w.

Thanks for your time;
Jeremy
 
To monitor current you use a current transformer, in your case simply passing the live wire through a toroidal type is probably all you need.

There are numerous Arduino examples, a simple google search finds lot's - here's just one of them:


However, it seems a fairly trivial problem to sort out regardless, and I don't see how monitoring with an Arduino will help?.

First off what's blowing? - is it an MCB? - these trip for almost no reason at all, particularly B curve types (the usual ones). If you replace the B curve type on that ring with a C curve type, it's likely that will solve your issues.

Your power requirements are low, and highly over estimated as well, for example your amplifier and PC will be taking no where near their maximum possible powers.

But, what on earth are you doing with 500W of lighting?, what type of lighting?, and why is that connected to the ring main?, and what is 500W Lab?.

Where I used to work, when the shop was completely refurbished, and MCB's fitted, neither the lighting ring nor the main power ring were able cope without blowing the MCB's at power up - in the end we went back to proper fuses. You've perhaps seen on television where they turn the lights on in an industrial area, and they turn on separately in sequence, this is to prevent blowing the MCB or fuse, by replacing one large surge with a number of small ones.

Where I work now, we have a small battery spot welder - and in our previous unit that used to intermittently trip the MCB on the ring, we cured it with a C curve MCB. In our new units, they have really old MCB's which aren't so overly sensitive.
 
There are also current transducers that work very well, and are easy to use:-
https://uk.rs-online.com/web/p/current-transformers/4995356
You can pass the live or neutral through the hole in the current transformer without needing to remove the insulation, and no further isolation is needed. On that model you can use the wires provided but you don't have to.

Those will also monitor DC currents.
 
When you say "the fuse has cut out" do you mean the fuse has blown (30 amp fuse), an over current circuit breaker (MCB) has tripped out or an earth leakage breaker breaker (RCD) has tripped out. What is the nature of the load and it's rated current consumption ? Can you provide information on the mains conditioner so we can understand how you expect it to stop the fuse cutting out ?
Les.
 
Last edited:
My apologies, I was in error thinking you wanted to measure DC loads. The subsequent posters got it
all right,use a current transformer, that way isolation is maintained, and ground reference for A/D on
measurement side can be user choice, such as simple ground in an Arduino or other embedded type
solution.

Note when you get the signal from load side in secondary side of current sensor, probably a
lot of line and power supply noise and complex waveform. So that can be handled by filtering,
and/or software filtering. Additionally the signal, if secondary side of transformer is grounded,
is AC so has to be level shifted into ADC if ADC operating single supply, which Arduino is. One
way of doing this -


Another way just R's can be used to level shift, see attached docs. Calculator included
for determining R's. Here we shift an AC ground referenced input into a positive
only range.

1737550995614.png



Regards, Dana.
 

Attachments

  • ADinput negative inputs.zip
    644.7 KB · Views: 22
Last edited:
H i all and thank you for your time with this idiot. I will try to answer all your questions. Firstltr I'm a programmer that knows something about electronics but have never worked in that field so pardon the mistakes.

First, the fuse is a trip type. I have no idea what sort but here's a picturte:

PXL_20250122_151449672.RAW-01.COVER.jpg


Hope that helps. It has only tripped twice and I'd like to oknow why. Plus a mains consitioner seems like a gopod idea since I had one lying around.

dana; that was my first thgought but if R3 dies I am left with HT goind into the Arduino; complete isolation seemed a better idea if it could be done and the small transformer I found seemed to fit the bill.

Nigel; I'm afraid that's what happens when youi writye stuff at 04:30. Giberrish is emitted. I have, in fact, 4x Beamz 36 RGBW LEDS, unit is rated at a max of 4.2A so 16.8A all in. At 2150v that gives 4.2Kw but I suspect when they say "power is 4.2A" they mean "4.2W"; so the total; is about 16W. The 'Lab' consists of a dual 31v 5A bench supply; I don't know what efficiency it has but going by the outputs at max I's be using 310W. Lewt's call it 400W. Aslo a peogogrammable signal generator (negligable I expect' and two digital4ch storage scopes. I don't know their power either but don't suppose it is much - say 100W for both. That gives 516W for lighting and lab if all turned to max.

The amp is a Yamaha DSP AX750 of a certain vintage but it still is ratewd at 100W/channel and has 7 of them. There is also a 200W powered subwoofer so sound maxes at 900W although I's probably blow up my speakers if I tried to hit that max. Shall we agree on about 100W for sound? That gives a total of 600W of power use. Again, that's surely an overestimate.

diver, sadly I am unablke to find details on the conditioner. From memory it contains a transformer, several capactitors and anothert transformer to LOAD. My guess is a low voltage step down, a bunch of stull to get rid of spikes and a step-upo output sstage. If I can find the spec I will supply it.

Diver300 + @Nighel I will look at your suggestions; it's a lot more complex than I expected but I'm hjere to learn and I did buuild a Z80 computer from tewlephone wire and solder back when I was a niupper so I shouldn't have lost all my skills. Mind you I've not ridden a bioke for akmost as long;l perhaps I'd just fall off!

@ anyone I forgot, thank you again for your helpful suggestions to a for all practical purposes is a newbie.

Jeremy
 
Why is the MCB only 16A?, I thought you said it was for a ring main?, which should be a 32A MCB.

A 16A MCB would probably just be feeding a spur.

Does it blow without the conditioner connected?, and why the conditioner anyway?.
 
I was just going to ask the same questions as Nigel. If it is not on a ring main then I assume the 16 amp MCB just feeds a single or double 13 amp socket I have just found a picture of the back of a Yamaha DSP AX750 and is show as 295 watts mains input which is about 1.2 amps. If you have 7 of them swtiched on at full output power that would be 8.4 amps.
When the MCB trips are ther any other sockets in your house that stop working ?
When the MCB tripped had you switche all the devices on at the same time? (Were the mains switches all switched on on the devices an you switched on at the 13 amp socket ?) I am thinking that the initial switch on surge caused the MCB to trip.
Les.
 
@dana; that was my first thgought but if R3 dies I am left with HT goind into the Arduino; complete isolation seemed a better idea if it could be done and the small transformer I found seemed to fit the bill.

You could always use a zener at ADC input in case R3 fails, or a pair of diodes -

1737567875862.png
 
Yes, I thought of a zenor too. Perhaps I'll go there. As for the breaker, I did not install the power to the flat and it's my opinion that the monkey that did made a royal cock up.
The high power circuit for the kitchen is on the upstairs ring main. That main is high current meaning it is a) dangerous and b) I have to plug the oven, microwave etc into what should be for uplighters etc.
It does not surprise me at all that the breaker labeled "downstairs ring" is on a 16a breaker.
Can I simply replace it?

Part II) I have no idea why it cut out (and the others didn't so no general problem. It is new to me. I installed the conditioner because I was tired of seeing it on my desk and I thought maybe it would help the audio and PC stability.
I now want to know if it made any difference hence this project.

I'm not sure about the battery charger idea; isn't it their job to produce constant current / voltage regardless of input? (to a limit of course but I suspect measuring the output of that would be highly non-linear.

I'm more keen on replacing the load shunt with a toroidial but to get enough power left to drive the rest of the kit would enough juice to run it would need to be quite beefy - or an I wrong?
 
Yes, I thought of a zenor too. Perhaps I'll go there. As for the breaker, I did not install the power to the flat and it's my opinion that the monkey that did made a royal cock up.
The high power circuit for the kitchen is on the upstairs ring main. That main is high current meaning it is a) dangerous and b) I have to plug the oven, microwave etc into what should be for uplighters etc.
It does not surprise me at all that the breaker labeled "downstairs ring" is on a 16a breaker.
Can I simply replace it?

You need to get your home wired properly - a 16A circuit isn't enough for a ring main, and lights should be on a completely different (and low power) circuit - unless you mean portable lamps, and not fixed ones?.

The UK electrical system is the best (and safest) in the world, so there no 'dangerous high current circuits', all UK plugs are individually fused.
 
I would NOT advise on replacing the 16 amp MCB with a higher rating one. The ring may not be wired with wire of a high enough current rating. A ring main in the UK is normally wired with 2.5mm cable. (The 2.5mm is the cross sectional area of the conductors. )
Les.
 
Can anyone suggest a suitable torrid to replace the shunt with? I apparently need 9a available after the torroid at 300vwhich I work out at 30 ohms or so, primary. I really am stuck here. I don't want the mother of all transformers in thr circuit but it must still let 10A (i rounded a bit) through at 250v RMS. What to I need to buy for this?

Pls help;
Jeremy
 
Suitable current transformers (CT) are suggested in the links in posts 4 & 5. There are also many types of current transformers available on Ebay.
Les.
 
The UK electrical system is the best (and safest) in the world...
With the giant, expensive and unsightly plugs/outlets used in the UK, it would be pretty embarrassing if the UK's wasn't the safest.
 

Latest threads

New Articles From Microcontroller Tips

Back
Top