OffGrid Switchboard Guidance

gregoinc

Member
Joined
Jan 7, 2018
Messages
274
Have begun investigating the size of switchboard I would use for my off-grid system i.e. to contain circuit breakers, RCD's, etc. One question came up that I hope someone with experience in Australian electrical standardscan help me answer...

If I were to have a switchboard similar tobelow that was wall mounted with DIN rail, could I have AC and DCcircuit breakers in the one switchboard?


image_hshmhn.jpg


I figured the answer would be that AC and DC circuit breakers must be mounted in separate switchboards or enclosures. Just wanted to be sure as the answer will affect the size of each switchboard/enclosure.

Thanks in advance for any guidance/feedback.
 
Also, I wouldn't want to keep battery wires in the same box as Mains power wires. There's too much chance of something going wrong. Plus, an AC power wire coming close to a DC power wire could cause noise in the DC side of things. Remember, that AC wire has a changing magnetic field around it and that could cause the other wire to pick up on it. You don't want any kind of alternating current in the DC side.
If you think distance will help, just look at how far a decent wireless charger works. Some can be as far away as 2Ft (.7m) away and still be charging. So the two power sources really should be in their own boxes.
 
Thanks for the replies gentlemen. I figured it would need to be separate, so I'll start with sourcing separate junction boxes.

Designing the wall layout for all the equipment that makes up the off-grid systemis damn hard... :sbut I'll keep plugging away.

Thanks again for the advice, having the help in thisforum is fantastic.
 
Thanks for the feedback.

In relation to earthing, does there need to be seperate DC and AC earthing? Or can it be a common earth?

And can the PV earth cable run in the same conduit as the PV positive and negative cables?
 
Earthing for the AC DC sides should be separate. You don't ever want AC running on the DC for any reason (by this I mean in the case of a short where the Line on the AC comes in contact with the Earth/Ground side, then AC would travel back along the Earth/Ground wire and onto the DC side if they were bonded). The potential of frying something on the other side is too high.

You can run earthing along the DC power transmission lines as the Earth wire is the same as the Neg wire once it gets to the box. At the panels they may not be, tho; not sure.
 
Korishan said:
Earthing for the AC DC sides should be separate. You don't ever want AC running on the DC for any reason (by this I mean in the case of a short where the Line on the AC comes in contact with the Earth/Ground side, then AC would travel back along the Earth/Ground wire and onto the DC side if they were bonded). The potential of frying something on the other side is too high.

Thanks, makes perfectly good sense. So, crazy question... would you have two earth stakes in the ground? One for AC, and one for DC?

Korishan said:
You can run earthing along the DC power transmission lines as the Earth wire is the same as the Neg wire once it gets to the box. At the panels they may not be, tho; not sure.

Figured as much. I had planned on having a DC earth connection on my DC negative busbar. But still unclear on how that will all hang together? Any circuit diagrams greatly appreciated :)
 
Yes, there'd be 2 stakes in the ground, preferably several feet from each other. Not sure what the code is in your area. Here in Fl I think they have to be about 6 inches from each other.

Usually, the way I understand it, is that Pos/Neg from the solar panels are ran to your charge controller. The panel frames are connected to the ground wire and each panel is subsequently grounded (if installed at ground level) or they all merge into one ground lead that goes down the house to the rod. Then the ground wire that runs with the Pos/Neg connects to the Ground/Earth of the Charge Controller. Internally, the controller has bonded the Ground and Negative connections.
Then the Ground wire from the inverter is then connected to another rod in the ground.

I could be off on the specific details, but I think that's how it generally is wired up.
 
Just want to add that if you are using telecom voltage like I will be (-48V) then it is the positive that is grounded not negative. It cuts down on corrosion drastically, and if youre talking about people who know how to do DC voltage properly Id argue the telephone industry has been the main users of it for a very long time. All of the datacenter DC specifications are based on telephone systems.
 
Yes in the telco industry they use positive ground for corrosion reduction - but that's intended to help protect kms of underground buried cables, which isn't so common in Solar systems ;-)

I'd suggest grounding the -ve is the "expected" thing ie someone wouldn't typically expect a +ve battery terminal to be ground & contact with the -ve terminate to be dangerous.

If you had two separate ground systems you could have voltage between them - "equal potential bonding" is the recommended practice.
The ground system should be one common system using one or more stakes all "bonded together" electrically.

Grounding of systems also depends on the AC side to DC side isolation inside devices such as inverters, ie there are some non-isolated devices & they've got to be done differently.

The solar panels frames & mounting rails should be grounded as well if over "ELV" voltages ie >120VDC.

The exception is lightning grounding which should be separate.
 
Redpacket said:
Yes in the telco industry they use positive ground for corrosion reduction - but that's intended to help protect kms of underground buried cables, which isn't so common in Solar systems ;-)

I'd suggest grounding the -ve is the "expected" thing ie someone wouldn't typically expect a +ve battery terminal to be ground & contact with the -ve terminate to be dangerous.

If you had two separate ground systems you could have voltage between them - "equal potential bonding" is the recommended practice.
The ground system should be one common system using one or more stakes all "bonded together" electrically.

Grounding of systems also depends on the AC side to DC side isolation inside devices such as inverters, ie there are some non-isolated devices & they've got to be done differently.

The solar panels frames & mounting rails should be grounded as well if over "ELV" voltages ie >120VDC.

The exception is lightning grounding which should be separate.

I've done a lot of reading and arrived pretty much where you have stated above.

Not saying the references to telco systems isn't important, but I am yet to find reference materials from a solar manufacturer or installer that promotes bonding earth to +ve battery. Most appear to take the approach of above, including the DC/AC earthing split going via the inverter.

I intend to install a single ground rod, primarily so I can earth the solar panels and mounting hardware. From there I'll use it for the DC battery charging earth, and I will continue my research soprior tothe time to install the Victron MultiGrid I will have a clear understanding of what will be required.

I am no expert... so very much open to other views, advice, examples, etc. Thanks again for all the feedback, it continues tobe a huge help.
 
The telecoms legacy of earthing the positive side goes all the way back to when telephone/telegraph wires were a single cable with a ground return loop. Only 1 wire was needed end to end and a spike at each end in the ground.

Earthing for domestic wiring has to do with the way the whole system is setup end to end as the generator at the power station is grounded on the neutral and subsequently any accidental live conductor contact with anything conductive will try to return to the generator (or local transformer) via the ground and via you if you are the go between fore the live to ground contact. The general wiring standards usually require most conductive items to be earthed (metal water pipes, etc) to prevent any fault condition (live wire contact) making them live and then posing an electrocution risk when you turn on the water tap for example.

With low voltage DC the ideal preference would be to leave it ungrounded because any touching of any individual conductor has no risk of electrocution because there is no ground loop. There is no return path via the earth to the battery if it is not grounded at all. Equipment used on construction sites use part of this principle and rely on isolation transformers to break the negative ground loop and they also reduce the voltage as well. That way if your drill cable is cut and you end up touching one of the wires you will not get a shock as there is no earth path back to the power station / local transformer.

If you have NPN MOSFET switching within your BMS and switches the pack off on low voltage your previously 0V -ve connections are now at 48V looking for a -ve return path because the BMS has disconnected the -ve line. If you then earth anything after the BMS the ground loop created will then bypass the BMS. The effect of using NPN with telecoms and +ve ground eliminates the ground loop issue.

House wiring always has any switching in the positive side, unlike the battery pack approach if the BMS is NPN based which disconnects the -ve terminal leaving all the wiring at a +ve potential.
 
Hey folks... thought I would throw this question in here first.

Am sourcing parts to build my DC busbars and having done some research I am not sure what bolts to use to attach cables to the copper busbars?

Some folks say stainless steel and others talk about something called 'yellow tinned' and others say the bolts should be brass?

Would be interested in the collective thinking from those of you that have experience/knowledge.
 
completelycharged said:
.... way back to when telephone/telegraph wires were a single cable with a ground return loop ....
Yeah we forget how far we've come!

completelycharged said:
With low voltage DC the ideal preference would be to leave it ungrounded because any touching of any individual conductor has no risk of electrocution because there is no ground loop.
Yes that's one way to do it.
completelycharged said:
If you have NPN MOSFET switching within your BMS and switches the pack off on low voltage your previously 0V -ve connections are now at 48V looking for a -ve return path because the BMS has disconnected the -ve line. If you then earth anything after the BMS the ground loop created will then bypass the BMS. The effect of using NPN with telecoms and +ve ground eliminates the ground loop issue.
For large systems, typical MOSFETs directly in +ve or -ve the probably wouldn't be good - breakers with shunt trips would be used due to the high currents & possible fault currents involved.
Yes it could probably be done with some of the heavy duty devices.

completelycharged said:
House wiring always has any switching in the positive side,
Agree - a technicality (for other readers): AC mains obviously doesn't have a positive & negative but a "live" or "hot" and a "neutral" or "return"


gregoinc said:
Hey folks... thought I would throw this question in here first.

Am sourcing parts to build my DC busbars and having done some research I am not sure what bolts to use to attach cables to the copper busbars?

Some folks say stainless steel and others talk about something called 'yellow tinned' and others say the bolts should be brass?

Would be interested in the collective thinking from those of you that have experience/knowledge.

Each has it's good & bad...

Stainless bolts probably apply more for lead acid batteries where terminal corrosion is more of an issue.
Corrosion from cell leakage in lithium packs means you've got more serious issues!
Stainless doesn't conduct quite as well as other metals - does this matter? probably not because it's the underside of the lug to battery terminal direct contact path taking the current.
They look nice :)

"Yellow tinned" is typically = "zinc dichromate" plating used on steel bolts. Not sure if this plating is conductive or not eg alumminium anodising is insulating, but this?

Brass is a good conductor & resists corrosion pretty well but physically it's not as strong & you could shear bolts if muscle applied!

My take is you should have a flat washer + a spring washer under the bolt head. Eg bolt > spring > flat > lug > busbar/battery terminal
Flat washer helps keep the lug flat & make widest contact area with busbar. Spring washer reduces risks of loosening from any thermal expansion changes, etc.

All the commercially made bus bars & switch gear I've used came with either zinc plated steel bolts (silvery looking) or zinc dichromate plated steel bolts (yellow looking)
 
Circling back to earth cabling. I am planning on running an earth cable from the ground spike to a DC cabinet containing an earth bar (the bar would be the central point for earthing).

If I am running an earth cable from the earth spike to the earth bar (by itself), does that earth cable need to be run inside conduit, like you would with PV cables.

Would appreciate your advice.
 
gregoinc said:
Circling back to earth cabling. I am planning on running an earth cable from the ground spike to a DC cabinet containing an earth bar (the bar would be the central point for earthing).

If I am running an earth cable from the earth spike to the earth bar (by itself), does that earth cable need to be run inside conduit, like you would with PV cables.

Would appreciate your advice.

From a safety perspective no, from a UV protection perspective probably yes.
My neighbor's mainsearth cable swithboard to stake is in flex conduit....
 
Redpacket said:
From a safety perspective no, from a UV protection perspective probably yes.
My neighbor's mainsearth cable swithboard to stake is in flex conduit....

Thanks for the tip. There will be UV compliant conduit installed from the external building cladding to the earth spike to protect the earth cable. Internally I had planned to run the double insulated earth cabling within the wall, held in place by appropriate cable retention.
 
Back
Top