diy solar

diy solar

BMS instead of Fuse/Breaker

eXodus

Solar Addict
Joined
Jul 27, 2020
Messages
1,457
This is danger zone - so here it goes:
After watching a few Tesla Teardowns I realized - they don't have fuses anymore in most places.

They call it virtual circuit breaker or e-fuse - a piece of software monitors the current and shuts down the power when it exceeds are certain threshold.

Most modern BMS do that already. When Tesla got approval to put this in a production vehicle - why are we still dealing with fuses? (at least the large one after the battery)

My inverter shuts down the output when I overload it - when I put a 2500w load on my 1500w inverter it shuts down within a second. There is never a danger for the wiring.

What steps would we need to take to use the BMS as a e-fuse:
1. We need to size the wire to rating of the BMS - if you got a 100A BMS - the wire needs to be able to take at least 125A
2. The device (inverter) connected to the battery current must be able to handle that current.
3. ??

I know the BMS could fail and not shut down - but the same is true for a fuse or breaker.

Thoughts, opinions?
 
I'm guessing theirs has passed stringent reliability testing, while a Chinese bms probably has not.

If you want to trust it go nuts.

A 40 dollar fuse is a lot cheaper than my camper
 
Most modern BMS do that already. When Tesla got approval to put this in a production vehicle - why are we still dealing with fuses? (at least the large one after the battery)
I have a class t fuse off the positive terminal because I don't trust my fet based bms to actually work.
I've learned from this forum that the MOSFETs in a the typical FET based bms can fail closed.
Also, the RnD that goes into a Daly, JBD or similar BMS is not on par with the RnD that goes into a tesla.
My inverter shuts down the output when I overload it - when I put a 2500w load on my 1500w inverter it shuts down within a second. There is never a danger for the wiring.
I control my Inverter via its remote switching ports using a Victron smart battery protect which is also controlled by the BMS.
I sill want a class-t fuse just in case.
The wire, BMS, main fuse and shunt add incremental resistance along the main circuit but they are all there for a reason.
I actually use a hall sensor instead of a shunt, part of the reason was to avoid the resistance penalty of the shunt.
What steps would we need to take to use the BMS as a e-fuse:
The e-fuse would need to be demonstrated to be reliable and price competitive.
 
I sill want a class-t fuse just in case.
A 40 dollar fuse is a lot cheaper than my camper
I mean you can always have double and triple safety, nothing prevents you from wearing a belt, suspenders and duct tape your trousers to your belly.

The question is - does a 40 dollar fuse really protect you? If you got a tiny short in a the inverter in a small wire - it will still burn -because even 1A can make a small wire hot.

Lets assume a 100A fuse - that thing will blow around 120A,

The positive wire gets rubbed through somewhere - and contacting a different ground wire, maybe the standard 12 / 16 gauge in your RV - when you only got a tiny contact patch - there will be no 100A flowing. The 16gauge will melt at 50A 12 gauge will melt at 80A and set the whole RV on fire.

What good did the Fuse for you? To get the 100A blowing you need to connect both large conductors to each other, or short inside the inverter - both scenarios are highly unlikely.

The whole story gets even more interesting when you go up in AMPS. When we go direction 200A fuses - even if the positive wire should touch the frame - the contact area will probably instantly evaporate into molten metal - like welding. A fuse wouldn't prevent that, it's to slow and you only need 150A to weld.
 
Last edited:
I mean you can always have double and triple safety, nothing prevents you from wearing a belt, suspenders and duct tape your trousers to your belly.

The question is - does a 40 dollar fuse really protect you? If you got a tiny short in a the inverter in a small wire - it will still burn -because even 1A can make a small wire hot.
Anything inside the box is the responsibility of the manufacturer.
Black box principle applies.
Lets assume a 100A fuse - that thing will blow around 120A,

The positive wire gets rubbed through somewhere - and contacting a different ground wire, maybe the standard 12 / 16 gauge in your RV - when you only got a tiny contact patch - there will be no 100A flowing. The 12 gauge will melt and set the whole RV on fire.

What good did the Fuse for you? To get the 100A blowing you need to connect both large conductors to each other, or short inside the inverter - both scenarios are highly unlikely.
@FilterGuy posted about this aproximately a year ago and its been rattling around between my ears ever since.
I guess I'm just doing the orthodox thing but I'll keep the e-fuse in mind.
The whole story gets even more interesting when you go up in AMPS. When we go direction 200A - even if the positive wire should touch the frame - the contact area will probably instantly evaporate into molten metal - like welding. A fuse wouldn't prevent that, it's to slow.
Are you saying that fuse won't blow in this scenario?
 
I mean you can always have double and triple safety, nothing prevents you from wearing a belt, suspenders and duct tape your trousers to your belly.

The question is - does a 40 dollar fuse really protect you? If you got a tiny short in a the inverter in a small wire - it will still burn -because even 1A can make a small wire hot.

Lets assume a 100A fuse - that thing will blow around 120A,

The positive wire gets rubbed through somewhere - and contacting a different ground wire, maybe the standard 12 / 16 gauge in your RV - when you only got a tiny contact patch - there will be no 100A flowing. The 16gauge will melt at 50A 12 gauge will melt at 80A and set the whole RV on fire.

What good did the Fuse for you? To get the 100A blowing you need to connect both large conductors to each other, or short inside the inverter - both scenarios are highly unlikely.

The whole story gets even more interesting when you go up in AMPS. When we go direction 200A fuses - even if the positive wire should touch the frame - the contact area will probably instantly evaporate into molten metal - like welding. A fuse wouldn't prevent that, it's to slow and you only need 150A to weld.
Not sure what any of that has to do with... anything I said that you quoted. Especially since the statement is that I still wouldn't trust it and would still have that fuse... just like you suggested, which you conveniently left out.

Your original post was about using these instead of a fuse or breaker.

This whole comment shows a fundamental failure to understand the function of the thing we are discussing.
An electronic overcurrent limiter would not solve a single one of those situations either, because it would break the circuit for a current beyond the intended only. It's function would be identical to the large fuse you just described and explained wouldn't protect anything because that's all it would be. Just another overcurrent device.

Why do you think an electronic overcurrent device would do anything differently?

How would that stop any of the scenarios you described when its
effectively just an electronic fuse?

Regarding "fail and not shut down", that's the thing:

That's on you, as the designer of the system.
The only time a fuse will fail to stop current is when the current/voltage (or some combination) vastly exceeds the interrupt capacity of that fuse. There's a whole system in place for identifying how much potential current is available on a given circuit and then the fuses or other interrupt devices need to be capable of interrupting the arc that forms when they blow.
That's why a class T is a thing for lithium. Because it can break thousands of amps, which a "regular" fuse cannot.
So the only time a fuse "fails to fail" is when the user/operator/person who installed that fuse failed to understand the circuit.

Does a $40 fuse really protect me?
Yes. Technically. Not directly, but yes.
Fuses and overcurrent devices are not designed to protect people. They are designed to protect things. As a nifty side effect they tend to prevent fires from occurring in the situations you are describing as well.

If you want to protect yourself directly: ground fault devices.


Is more interrupts a good thing?
Yes.

Does my BMS do this?
Yes.

Am I glad my BMS does this?
Yes.

Do I still have a $40 class T fuse in the mail?
Yes.

Do I still have fuses on the main feeder circuits?
Yes.

Do I still have more fuses on the individual auxiliary circuits?
Still yes.

I never said you can't have it. I simply said it would need to pass a ton of qualifications first before being officially considered a "safety" device.
 
Are you saying that fuse won't blow in this scenario?
depending which kind of fuse/breaker you pick, there are slow and fast acting fuses and breakers. Since the iron of the frame already almost instantly melts above 120-150A a slow blow fuse will not catch that. Only like Class T.

Especially since the statement is that I still wouldn't trust it and would still have that fuse... just like you suggested, which you conveniently left out.
I like to make thought provoking statements :) no personal offense was intended. Accept my apologies.

I simply said it would need to pass a ton of qualifications first before being officially considered a "safety" device.
That gives me some hope that a BMS could be potentially be used a overcurrent safety device with additional testing and maybe redesign to fulfill this duties.
 
That gives me some hope that a BMS could be potentially be used a overcurrent safety device with additional testing and maybe redesign to fulfill this duties.
It already is, it just usually not the only over-current protection device for the battery or the main circuit.
I wonder in Battleborn would bet your house on their bms.
If you rely on the BMS as your "Armageddon" over-current protection does that mean that all the branch circuits should be fused on the negative lead?
 
@FilterGuy posted about this aproximately a year ago and its been rattling around between my ears ever since.
I guess I'm just doing the orthodox thing but I'll keep the e-fuse in mind.
Since I posted about this it has been 'rattling around between my ears' as well. Since then I have identified *many* similar scenarios. The only way I can think of to use fuses/breakers to reduce the risk is to put the devices on both the negative and positive..... That is not done in any typical system I am aware of. The only way I can think of to use passive components to manage this is to use super large wires on everything... and that is not going to happen either.

Where I finally ended up was
1) We need to be careful with the physical layout of the system to avoid the possibility of a high amp circuit shorting to a low amp negative circuit.
2) If this were a large issue, we would see the standards bodies start putting safety requirements in to address it. Since we are not seeing this, I can only conclude that this is not a common problem.

BTW: This is definitely more likely to be an issue with battery systems because of the huge differences in the currents on various DC circuits. On the AC side, the most likely result of a 30A circuit (10AWG) shorting to an 18AWG negative is a popped breaker. On the DC side we see 125A 4AWG circuits and 22AWG control wires.....
 
Where I finally ended up was
1) We need to be careful with the physical layout of the system to avoid the possibility of a high amp circuit shorting to a low amp negative circuit.
Agree.
High current busbars at system core and a fuse_block with built-in negative terminals keep things isolated and organized.
Like zoning for cities.

2) If this were a large issue, we would see the standards bodies start putting safety requirements in to address it. Since we are not seeing this, I can only conclude that this is not a common problem.
Agree.
BTW: This is definitely more likely to be an issue with battery systems because of the huge differences in the currents on various DC circuits. On the AC side, the most likely result of a 30A circuit (10AWG) shorting to an 18AWG negative is a popped breaker. On the DC side we see 125A 4AWG circuits and 22AWG control wires.....
Yep.
 
I wonder in Battleborn would bet your house on their bms.
If you rely on the BMS as your "Armageddon" over-current protection does that mean that all the branch circuits should be fused on the negative lead?
as soon as you change wire sizes you need to add an overcurrent device. The question is which direction.

I'm really worried about all those very high current application, I don't like dealing with high current (thick wires, lot's of heat when you got a bad connection) I'd rather go with "higher" voltage 24v or 48v.

When going that way - is maybe a good idea to have the battery system floating? No chassis connections.

But to loop back to E-Fuse integrated into the BMS - the software could be programmed to detect even less then catastrophic events.

A regular fuse - is only good against one short. While a software defined - could also detect sparking through movement.
 
as soon as you change wire sizes you need to add an overcurrent device. The question is which direction.

I'm really worried about all those very high current application, I don't like dealing with high current (thick wires, lot's of heat when you got a bad connection) I'd rather go with "higher" voltage 24v or 48v.

When going that way - is maybe a good idea to have the battery system floating? No chassis connections.

But to loop back to E-Fuse integrated into the BMS - the software could be programmed to detect even less then catastrophic events.

A regular fuse - is only good against one short. While a software defined - could also detect sparking through movement.
Yep, with a signal processor lots of possibilities including AFCI and GFCI.
 
Yep, with a signal processor lots of possibilities including AFCI and GFCI.
now we are getting somewhere. Awesome idea.

What else could a measuring device like the BMS get used for?



Could we have a BMS with multiple side circuits ?
Agree.
High current busbars at system core and a fuse_block with built-in negative terminals keep things isolated and organized.
Like zoning for cities.
Currently a BMS is one in - one out.

Why not one in many out? Have 200A BMS - which has a 100A tap 2x 50A taps 4x 20A whatever?
 
now we are getting somewhere. Awesome idea.

What else could a measuring device like the BMS get used for?



Could we have a BMS with multiple side circuits ?

Currently a BMS is one in - one out.

Why not one in many out? Have 200A BMS - which has a 100A tap 2x 50A taps 4x 20A whatever?
Being able to do signal processing to detect various effects depends on the fidelity of the original measurement/signal as well as what and where the measurement is taken. The current sensors used in a BMS might be adequate for some types of processing, but are probably to slow to do things like arc fault detecting. Obviously, if there were a demand for it, the proper sensors could be added to do a lot of different things.... for a price.

We already have BMSs with multiple circuits..... one for charge and one for discharge. (Ok...OK.... that is not what you were thinking :) ). However, I am not sure how useful having multiple output ports would actually be. For a home system, there is usually only one load: The inverter. For mobile systems, the inverter load is almost always vastly larger than any other load so most people will be looking for one of the ports to be able to handle nearly 100% BMS capability. I guess having a 2nd port that takes the place of a battery protect on the DC loads would be useful, but everyone will have a different need for the current on the 2nd port so the number of different BMSs could start to get large...making it a difficult business proposition.

Having said that, from an engineering point of view it is all possible. All it takes is a market willing to pay for the extra functionality and someone willing to take a chance on a new market segment. I am just not sure there would be a large enough demand. Take a look at the common BMSs we are using in the DIY community. A lot of the BMSs discussed on the forum are re-purposed from other markets like scooters and medical devices to sell into the DIY market. That implies the demand for DIY BMSs is still somewhat small.
 
I mean you can always have double and triple safety, nothing prevents you from wearing a belt, suspenders and duct tape your trousers to your belly.

The question is - does a 40 dollar fuse really protect you? If you got a tiny short in a the inverter in a small wire - it will still burn -because even 1A can make a small wire hot.

Lets assume a 100A fuse - that thing will blow around 120A,

The positive wire gets rubbed through somewhere - and contacting a different ground wire, maybe the standard 12 / 16 gauge in your RV - when you only got a tiny contact patch - there will be no 100A flowing. The 16gauge will melt at 50A 12 gauge will melt at 80A and set the whole RV on fire.

What good did the Fuse for you? To get the 100A blowing you need to connect both large conductors to each other, or short inside the inverter - both scenarios are highly unlikely.

The whole story gets even more interesting when you go up in AMPS. When we go direction 200A fuses - even if the positive wire should touch the frame - the contact area will probably instantly evaporate into molten metal - like welding. A fuse wouldn't prevent that, it's to slow and you only need 150A to weld.
You don't just fuse the battery cables, you fuse every smaller wire that feeds power to a device. This is how proper wiring works.

Look at your car's fuse box. It's fed by a fat wire coming off the battery. That wire is fused. Then, every wire feeding every electrical circuit gets fused going out of the fuse box.
 
You don't just fuse the battery cables, you fuse every smaller wire that feeds power to a device. This is how proper wiring works.

Look at your car's fuse box. It's fed by a fat wire coming off the battery. That wire is fused. Then, every wire feeding every electrical circuit gets fused going out of the fuse box.
The poster was talking about an unusual short circuit scenario. Consider this hypothetical:

1632521733896.png
The circuits are all properly fused and the wires are all properly sized, but you end up with a 125A fuse dumping onto a 16AWG wire..... There is a good chance the wire will burn to protect the fuse.

There are not a whole lot of viable ways to deal with this.
- Make all the wires 1 AWG...... not going to happen.
- Put fuses on both the positive and negative.... I am not aware of anyone doing this.

As I said in an earlier post I have settled on this reasoning:

1) We need to be careful with the physical layout of the system to avoid the possibility of a high amp circuit shorting to a low amp negative circuit.
2) If this were a large issue, we would see the standards bodies start putting safety requirements in to address it. Since we are not seeing this, I can only conclude that this is not a common problem.
 
However, I am not sure how useful having multiple output ports would actually be. For a home system, there is usually only one load: The inverter. For mobile systems, the inverter load is almost always vastly larger than any other load so most people will be looking for one of the ports to be able to handle nearly 100% BMS capability. I guess having a 2nd port that takes the place of a battery protect on the DC loads would be useful, but everyone will have a different need for the current on the 2nd port so the number of different BMSs could start to get large...making it a difficult business proposition.
in my future predictions - there will be a lot more DC appliances and inverters will get smaller again.
DC Air conditioners, DC refrigerators, DC Microwave. DC Induction cooktop

Currently we have those things AC - because it's legacy/cheap and at 12V the wiring is unpractical. But when we hit 24V or 48V DC as standard. It makes a lot more sense having those things directly connected.
 
The circuits are all properly fused and the wires are all properly sized, but you end up with a 125A fuse dumping onto a 16AWG wire..... There is a good chance the wire will burn to protect the fuse.
This is avoided by proper wire layout. There are many theoretical scenarios one could speculate on that never happen in the real world, unless the builder does not know what they are doing. And in that case, they probably have bigger problems to be concerned about.
 
This is avoided by proper wire layout. There are many theoretical scenarios one could speculate on that never happen in the real world,
Counter question - Is a fuse needed with a proper wire layout? Or is the BMS as overcurrent device sufficient?

I'm trying to come up with case - where the large fuse really would blow. Aside of connection both poles of the battery - or a internal inverter failure I haven't found one yet.

Please describe the scenario when the fuse would do something better then a BMS would do.
 
The poster was talking about an unusual short circuit scenario. Consider this hypothetical:

View attachment 66193
The circuits are all properly fused and the wires are all properly sized, but you end up with a 125A fuse dumping onto a 16AWG wire..... There is a good chance the wire will burn to protect the fuse.

There are not a whole lot of viable ways to deal with this.
- Make all the wires 1 AWG...... not going to happen.
- Put fuses on both the positive and negative.... I am not aware of anyone doing this.

As I said in an earlier post I have settled on this reasoning:

1) We need to be careful with the physical layout of the system to avoid the possibility of a high amp circuit shorting to a low amp negative circuit.
2) If this were a large issue, we would see the standards bodies start putting safety requirements in to address it. Since we are not seeing this, I can only conclude that this is not a common problem.
I have been around enough to see that a tiny wire can pop a big fuse but not enough to say for certain that it will always do so.

Fairly recently I blew a 150 amp ANL by touching my amp main positive with my then unknowingly grounded 20 awg remote turn on wire.

The resistance of 20awg over 15 feet is such that it shouldn't have even been possible unless the fuse essentially failed at half it's rating.

That said, in your example 20 feet of 16awg should have such a resistance that you'd expect it to "pull" 150 amps and that fuse should still pop. It makes me wonder if there's a study somewhere on this kind of short scenario.

As you state, though a bit exaggerated, ensuring a minimum wire size may alleviate this. 20 feet of 12awg would be 0.032 ohms on paper and @ 12v should see 375 amps and blow virtually any fuse you'd realistically use in a 12v system.

And it's not that big a deal to run 12awg everywhere in a build.
 
Back
Top