Fuse/Circuit breaker Sizing and placement.

[JoeHam]

Main failure mode for mosfets and diodes is short-circuit. You also want to protect against accidental short-circuit by operator error.

Are you saying the internal fuse won’t protect this situation?

 

[BiduleOhm]

It will, if there's any of course

But it'll not protect if you short the wires when attaching them for example (but you should not attach energized wires to begin with...).

 

[Dzl]

This whole discussion is based on the assumption that the inverter will destroy itself trying to put out more power than it can safely do.

Is there any evidence that this is true?

To the contrary, any inverter I have used has an internal fuse which is happy to sacrifice itself in this case.

A valid question, the assumption is based on the video (the link takes you to the relevant part of the video) I linked to from Pacific Yacht Systems / Jeff Cote, where he explicitly states that inverters will exceed their ratings if asked to.

I don't take their/his word as gospel or have any personal experience, but from what I know they are a very well regarded marine electrical company, that does work on >1000 boats a year (from system design, to installation, to maintenance/troubleshooting, and insurance claims inspection, have monthly columns in several electrical industry publications, host marine electrical trainings and seminars, and have 100's of hours of youtube content.

This is all a long winded way of saying, I have no personal knowledge of whether inverters are fused internally, but moderately high trust in the source of the information.

That said, it seems logical that inverter manufacturers (at least top tier ones) would fuse the devices internally, maybe some do, I don't know, consult your manual, but based on the video I get the impression that most do not.

I just took a quick peak at a few inverter manuals.

• The Samlex Evo Manual, doesn't mention an internal fuse, and implies there is no internal circuit protection:

Over current protection (fuse) for battery and External Charger circuits has NOT been provided and has to provided by the installer / user. See guidelines at Tables 1.3(a) and 1.3(b) for recommended sizes for installation in free air and conduit respectively. National and Local Electrical Codes will supersede these guidelines.

• The Victron Multiplus makes no mention of an internal fuse, says it has no user serviceable parts, but gives no clear indication one way or the other
• The Victron Phoenix (small inverter) do mention internal fuses but based on the fuse size (200A for a 1000W 12v inverter) appear to be sized based on the peak surge current not the continuous current.

The inverters are fitted with an internal DC fuse (see table above for rating). If the DC cable length is increased to more than 1,5m, an additional fuse or DC circuit breaker must be inserted close to the battery.Reverse polarity connection of the battery wires will blow the internal fuse and can damage the inverter. The internal fuse is not always replaceble (see table above).

 

[Sgt Raven]

If I drop a wrench across the terminals unless the fuse is the terminal how is a fuse gonna help. Please I'm not trying to be argumentative. I just think I want my fuses protecting my devices and by default the wires will be protected.
Trust me I know from much experience the utility of fuses. Every time I trip a breaker I say WOW glad I had that there.

A wrench across the terminals was an example of a "dead short", just as a short in your main cable running from the battery would be.

 

[Sgt Raven]

A valid question, the assumption is based on the video (the link takes you to the relevant part of the video) I linked to from Pacific Yacht Systems / Jeff Cote, where he explicitly states that inverters will exceed their ratings if asked to.

...snip....

That said, it seems logical that inverter manufacturers (at least top tier ones) would fuse the devices internally, maybe some do, I don't know, consult your manual, but based on the video I get the impression that most do not.

I just took a quick peak at a few inverter manuals.
...snip.....

2+ years ago I bought a Bestek 300W pure sine wave inverter that had a bunch of good reviews at the time. It's been sitting in it's box ever since. Yes I know, I'm the dumb guy, who should have plugged it in when I first got it. Last week I cut the cig plug off the end and wired a Anderson Power Pole on it. I checked the polarity multiple times to be sure I had it correct. Took it out and plugged into my new battery build. Heard a large POP and smelled burnt electronics. I wired a Anderson Power Pole on the cig plug and checked it plugged in to my car and I did have the polarity right. Now I have $50 of junk. On this particlur inverter there was no fuse, either in the cig plug or the inverter.

 

[John Frum]

2+ years ago I bought a Bestek 300W pure sine wave inverter that had a bunch of good reviews at the time. It's been sitting in it's box ever since. Yes I know, I'm the dumb guy, who should have plugged it in when I first got it. Last week I cut the cig plug off the end and wired a Anderson Power Pole on it. I checked the polarity multiple times to be sure I had it correct. Took it out and plugged into my new battery build. Heard a large POP and smelled burnt electronics. I wired a Anderson Power Pole on the cig plug and checked it plugged in to my car and I did have the polarity right. Now I have $50 of junk. On this particlur inverter there was no fuse, either in the cig plug or the inverter.

Well that's one data point but I think it may not be representative.
A typical cig socket is probably fused at ~12.5 amps.
I had a really cheap inverter from a company that I will not even name to protect the guilty.
Rated for 500 watts actually capable of ~300 watts and it was fused.

 

[Sgt Raven]

Well that's one data point but I think it may not be representative.
A typical cig socket is probably fused at ~12.5 amps.
I had a really cheap inverter from a company that I will not even name to protect the guilty.
Rated for 500 watts actually capable of ~300 watts and it was fused.

IIRC most Cig sockets are fused @20A, and a 300W inverter can't pull all 300W from one without popping the fuse. That's why they sell a battery clamps to cig plug adapter that is unfused to use with them.

 

[JoeHam]

I have whatever the opposite of “top tier” inverters would be ?.

It is a WZRELB 48v 3KW inverter and I have treated it badly. It came with a bag of fuses and I have used them after shorting out the AC output. Sparks were involved. Like the old TV commercials say “it takes a licking and keeps on ticking”.

If you have seen any of the David Poz YT videos he routinely tries to overload inverters and they usually protect themselves.

 

[JoeHam]

The point I want to make is that the generally accepted wisdom is true:

A proper fuse is present to protect the wire.

A properly designed device protects itself. Period.

I don’t think it makes sense to underrate your wire protection fuse in hopes of adding device protection. Makes no sense to me.

And if the simple purpose of “protecting the wire” isn’t sexy enough then in your mind change that phrase to “protecting the wire from a catastrophic failure that burns down my house and kills me while I sleep”. ?

 

[FilterGuy]

I am not an expert on UL listings (and less so for certifications in other countries). However, my guess is that to get a UL listing the device has to survive a short without causing a fire. However, I would also guess that a UL listing says nothing about the device remaining functional.

 

[Dzl]

If you have seen any of the David Poz YT videos he routinely tries to overload inverters and they usually protect themselves.

This is good to know, and reassuring! Do you know if he has tested Victron, Samlex, or Giandel?

I also wonder if (like with lifepo4) some damage can be catastrophic and obvious and other damage might be silent and degrade the device over time. I.E. if the fuse protected inverters are fused to prevent catastrophic failure, or to prevent damage to the device. In the case of the Victron Phoenix (the fuse is 200% of the rated output) my guess is the fuse is only there to prevent catastrophic failure, and wouldn't prevent damage to the device.

 

[Dzl]

The point I want to make is that the generally accepted wisdom is true:

A proper fuse is present to protect the wire.

A properly designed device protects itself. Period.

I don’t think it makes sense to underrate your wire protection fuse in hopes of adding device protection. Makes no sense to me.

I get where you are coming from, but I tend to disagree. Or at least I disagree with that as a generalization, if you are sure your device protects itself from damage, then yes, fusing to protect your inverter is probably unneccessary. However if there is any amount of doubt, it seems prudent (and relatively simple) to size your breaker @ 90% (or even 95% or 100%) of your inverters rated output the same way you oversize your wire.

Sizing your inverter the same way you size wire (estimate max current and make sure your wire and your inverter exceed this number with some safety margin) seems smart anyways.

You say "a properly designed device protects itself period." I agree with you in theory and I agree with you that that very much should be a design principle and that reputable companies should have safety margins and device protection built in. But I'm not sure this can be counted on in practice, especially considering that at least one reputable marine electrical professional, is explicitly saying not to count on an inverter to self limit (though to be clear the idea to undersize the breaker is my own, not something he advocates). Maybe he is making an overly cautious generalization, but it seems risky to assume your inverter is fused internally without it being explicitly mentioned in the documentation (especially if you have an expensive inverter). And an overly cautious generalization seems much better than a potentially risky generalization.

I don't want to sound too hardened or certain in my opinion, because (while this is my opinion based on what I know right now), I want to keep exploring the topic, and my understanding may continue to evolve as we explore more and as more info comes to light. But right now it seems to me that in principle, in the case of an inverter, the concept of fusing to protect the wire and the device has merit, is easy enough to accomplish and deserves further consideration.

And to be clear I agree with you that ideally, the builder of the device (the manufacturer) should handle device protection, and the builder of the circuit (us) should handle circuit protection. But because we only have control over the circuit and not the device, ultimate responsibility for both device protection and circuit protection falls on us (either by verifying that the manufacture has included proper device protection, or designing with the assumption that they have not).

I should also probably acknowledge that I'm definitely sometimes guilty of over-caution and over-engineering.

And if the simple purpose of “protecting the wire” isn’t sexy enough then in your mind change that phrase to “protecting the wire from a catastrophic failure that burns down my house and kills me while I sleep”. ?

I'm not sure if this is directed at me, if it is, you will see in the same post that I outlined sizing the fuse to protect the device, I stated that the primary purpose of circuit protection is to protect the wire and have professed that principle many times, so we are on the same page there, I just think an inverter may be an exception to that rule. ✌

 

[JoeHam]

I'm not sure if this is directed at me

Absolutely not.

I try to remember that these posts live on forever and merely want to reinforce that protecting the wire is a noble cause.

I stand by what I said and really don’t have anything else to add.

If a device is poorly designed I will be sure to blow it up early and never look back once the smoke clears. The rest of my system will be properly protected.

 

[Dzl]

I try to remember that these posts live on forever and merely want to reinforce that protecting the wire is a noble cause.

In that case we are definitely in agreement on that point.

I stand by what I said and really don’t have anything else to add.

If a device is poorly designed I will be sure to blow it up early and never look back once the smoke clears. The rest of my system will be properly protected.

Fair enough, I think I pretty much understand your perspective and see where you are coming from. I think I just have a somewhat more cautious philosophy and feel that the cost/benefit of adding a breaker to protect the inverter costs almost nothing (in dollars and in watts) relative to the cost of a replacing a moderately expensive inverter. Even if the device is internally fused, resetting a breaker is much more convenient than replacing an internal fuse. This is just my 2 pence, as I said above, I often err on the side of caution and 'over engineer' things..

edit: and obviously its reasonable that the amount of caution should be somewhat proportional to the cost of the inverter. Destroying or wearing out a couple $100 dollar inverters is a lot different than damaging a $1500 dollar inverter/charger

 

[Sgt Raven]

In that case we are definitely in agreement on that point.



Fair enough, I think I pretty much understand your perspective and see where you are coming from. I think I just have a somewhat more cautious philosophy and feel that the cost/benefit of adding a breaker to protect the inverter costs almost nothing (in dollars and in watts) relative to the cost of a replacing a moderately expensive inverter. Even if the device is internally fused, resetting a breaker is much more convenient than replacing an internal fuse. This is just my 2 pence, as I said above, I often err on the side of caution and 'over engineer' things..

edit: and obviously its reasonable that the amount of caution should be somewhat proportional to the cost of the inverter. Destroying or wearing out a couple $100 dollar inverters is a lot different than damaging a $1500 dollar inverter/charger

Another thing about a circuit breaker at the inverter end. A properly constructed DC circuit breaker has a way of dealing with a internal spark. Using the circuit breaker as a switch, should take care of the spark you get when connecting a inverter to it's power source.

At least I think it would, if wrong please point out where I am.

 

[Craig]

Another thing about a circuit breaker at the inverter end. A properly constructed DC circuit breaker has a way of dealing with a internal spark. Using the circuit breaker as a switch, should take care of the spark you get when connecting a inverter to it's power source.

At least I think it would, if wrong please point out where I am.

I have also noticed with my SSR's there is no spark either.

 

[Dzl]

One thing I don't think ever got fully covered was proper circuit protection for the solar charge circuit(s). There will naturally be some variability based on PV array size and arrangement. But lets start simple. One Panel, One SCC, One Battery.



I see three logical potential places for circuit protection, from top to bottom they are:

1. A breaker (dual pole? single pole?) between the Solar panel and the charge controller. For reasons discussed on page 1 of this thread and in this blog post, overcurrent protection might be impractical and unneccessary at this leg of the charging circuit, but being able to isolate the system from the charge source is useful.
2. A fuse or breaker close to the SCC sized for the wire between the battery and the SCC. In the event of a short between the battery and the SCC this would stop current flowing from the SCC.
3. A fuse or breaker close to the Battery side of this circuit, sized for the wire between the battery and the SCC. In the event of a short between the battery and the SCC this would stop current flowing from the battery.

I suspect that I might be misunderstanding how current would behave in the event of a short between the battery and SCC. If the positive wire got severed and both sides of it shorted to ground (for instance chassis ground of a vehicle), would that create two energized and shorted circuits (one energized by the battery and the other by the charge controller/pv panel?

What should the best practice or default practice for OCP of the solar charging system be?

 

[Craig]

I do not know what is correct but I like the idea of having a breaker on each end of the wires between SCC and battery it may be overkill but for 40 bucks I don't see a problem.

 

[FilterGuy]

A lot of the Charge controllers have built-in over-current protection. In this case, just the fuse at the battery should be sufficient. If there isn't built in protection then yes, a properly sized fuse would need to be placed on the output of the charge controller.

1. A breaker (dual pole? single pole?) between the Solar panel and the charge controller. For reasons discussed on page 1 of this thread and in this blog post, overcurrent protection might be impractical and unneccessary at this leg of the charging circuit, but being able to isolate the system from the charge source is useful.

Correct, for a single panel (or a set of panels in series, a fuse/breaker is not needed (and would be difficult to size correctly if it was needed).
Protection is not important till you have multiple panels or strings of panels in parallel.

As you point out though, some folks find it valuable to have a switch or breaker in the circuit for disconnect purposes. I have not done this, but if I felt the need, I would probably switch both legs of the circuit.... but don't ask why... it just seems like the right way to do it.

1. A fuse or breaker close to the SCC sized for the wire between the battery and the SCC. In the event of a short between the battery and the SCC this would stop current flowing from the SCC.

A lot (most?) of the SCCs have built in over-current protection. However, if it doesn't then Yes, a properly sized fuse at the SCC is needed in order to keep things safe.

1. A fuse or breaker close to the Battery side of this circuit, sized for the wire between the battery and the SCC. In the event of a short between the battery and the SCC this would stop current flowing from the battery.

Correct.

I suspect that I might be misunderstanding how current would behave in the event of a short between the battery and SCC. If the positive wire got severed and shorted to ground (lets say the chassis of a vehicle), would that create two energized circuits (one energized by the battery and the other by the charge controller/pv panel?

Yes.... you have two current sources tied together (SCC and Battery). If the connecting wire is severed, both remain hot.

I would say you have a pretty good understanding of it.

 

[Dzl]

A lot of the Charge controllers have built-in over-current protection. In this case, just the fuse at the battery should be sufficient. If there isn't built in protection then yes, a properly sized fuse would need to be placed on the output of the charge controller.

One consideration here (specific to Victron components (with the exception of their smallest controllers), but maybe others too) is that the internal fuse is non-user replaceable (soldered). But now that I'm thinking more about it, I'm realizing that even in a short circuit situation the SCC side of the shorted circuit couldn't exceed the output current its rated for right (unless the SCC itself was the problem)? So you are sortof in the same position as with the panels, their wouldn't be enough difference between the short circuit current and the max normal output current for a fuse to be effective in a short circuit situation without tripping in normal use all the time. Does this logic make sense or am I misinterpreting something?