EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: beepbob on September 20, 2020, 01:00:04 pm
-
Hi guys,
I have a 12V battery system which is connected to an inverter to generate 230V AC.
The whole system is not grounded and has no connection to earth on neither side.
What could happen if there is a fault and the phase of the isolated AC accidentally connects to battery plus or some low voltage circuitry powered by the 12V side?
After my understanding, the 230V AC is isolated from the primary side through the inverter's transformer so there is no (or only a very small) current that can flow.
I understand that the potential is floating. Can this damage circuitry and lead to a malfunction of the batteries with potential fire though?
-
what phase you're talking about?
Does your connections involve mains?
Please draw schematic of your wiring, so it will be more easy to explain what you're talking about.
And what inverter you're talking about exactly? Could you provide schema of the inverter output?
-
Thanks for your answer and sorry for the lack of information.
I've drawn a simplistic diagram to illustrate my setup, I hope that helps.
The setup includes mains voltage (aka. 230V AC) but is not grounded. So the generated AC is like from an isolated transformer.
I'm talking about the isolated output phase of the inverter (L). What happens if that shorts to the control circuitry or to the + of the batteries.
(https://i.imgur.com/MoYbn0B.jpg)
-
At a glance it depends on what is inside inverter. If it has isolation between input and output, such failure should not affect inverter.
But according to your schema, there is also control circuit which is connected to both - 12 V and to 220 V. So it may dangerous for your control circuit. How it will react for 220V between it's input and output? Can it survive in such case?
For example:
(https://www.eevblog.com/forum/beginners/what-can-happen-incase-of-a-fault/?action=dlattach;attach=1070538;image)
-
The DC input seems to be isolated from the AC output.
Also, the control interface in the load is powered by a SELV voltage generated directly from the load itself.
I understand that such a failure wouldn't damage the inverter. However, it is not clear for me how my control circuitry would react if theres a short as you have drawn it. The used components clearly can't handle normal mains voltage.
But if this fault happens, are there even "real" 230V applied to the circuitry? I'm having a hard time how it will electrically behave if I connect the 12V potential with the isolated 230V potential.
-
It all depends on what's in the inverter and what kind of failure mode you experience.
I would at least put a fuse on battery side to protect the battery, and on 230V side. Is load something that is fixed or is this going to a socket for general use? How can it have 230V feed going back to it? Accidental connection? Also, is there going to be a charging circuit? I see you are grounding 12V side. Why? It would make more sense to ground high voltage side but that, too, depends on your application. Floating power itself doesn't make the circuit dangerous.
You might want to study the inverter carefully. I'd expect control feedback to have some kind of optical isolation somewhere.
-
The battery is fused, the 230V side relies on the inverter's internal overcurrent protection. If current is limited, the inverter will shut down after a few seconds and restart after 30 seconds. This cycle repeats for three times and after that the inverter has to be restarted manually.
The inverter is equipped with a Schuko socket but the load is fixed and will rarely change. The batteries are charged over a solar charge controller which is not shown in the diagram.
Sorry for the inaccurate drawing - the battery is not connected to earth. The ground symbol just refers to the 12V common ground.
I'm not sure what you mean with "230V feed going back to it". If you mean how the 230V connections which @radiolistener drew resulted, that's only in a fault scenario, e.g. through damaged cables.
The manual of the inverter only states that DC input is isolated from AC output and a few safety standards which are met. Unfortunately, there is no info about the control interface. I expect it to be either potential free or tied to the DC side.
-
I would not trust inverter's overload protection 100%. I'd still put a fuse or a circuit breaker.
I think you could mitigate a lot of this input to output short type scenario by carefully dressing wires and routing them separately. It would be important to keep wires physically separate, so there is no chance of one side overheating would cause a fault like that.
I really don't think there's much more you can do.
-
I'd expect control feedback to have some kind of optical isolation somewhere.
Yes, it will be good idea to have optical isolation between control circuit and high voltage part.
Another way is to power control circuit from high voltage side with some efficient switching mode power supply, so you can decouple it from 12 V side. In such case 12 V will be used to feed inverter only, so if inverter has isolation, such failure should not affect it.
-
The 12V batteries are the supply .If a catastrophic failure occurs in the inverter or anywhere in the circuit the voltage will drop instantly to 12 V since there is no external 230V supply . If a short occurs then the short becomes the load and will heat up and melt which would likely break connection before the batteries heat up enough to cause an explosion. If by chance your inverter goes into a run away situation ,anything that is powered by the 12V supply will likely burn up immediately again breaking connection to the battery.
This of course is assuming there is no external 230V supply from mains .
-
By the way....
I wouldn't even try to limit what can happen in a worst case. If history is a guide, It will fail in the most unusual and unexpected way given long enough time. It will likely do so in more severe way, if human is involved. I wonder what would happen if someone drops a roll of aluminum foil just so that it won't blow a fuse but cause enough of heat?
Google search Therac-25 and you'll see what I mean.
This is why I recommended fusing everything everywhere.