To fuse or not to fuse and where

[Lurkin]

Just considering the appropriate fusing points for a large pack.

Options seem to be:
- Fusing every cell, Tesla Style (won't be doing this)
- Fusing each parallel group (i.e. PPTC fuses) (this is what I am debating doing)
- Fusing on the positive on the output/input of the entire battery (some seem to use a circuit breaker) (Luke has recommended a fast blow fuse for this elsewhere https://endless-sphere.com/forums/viewtopic.php?f=2&t=28601&hilit=circuit+breaker)

I realise for every fuse it also becomes a potential problem in the middle of the pack, but also increases the protection for a short within the pack itself. I will probably fuse each parallel group then the overall pack, just curious to see what others think and/or are doing.

Are blade fuses fast enough/appropriate for an overall pack fuse?

(Yes, I have already used the search function already. Most of the threads are pretty old and may be referring to old technology).

 

[DVDRW]

The more are in parallel the more likely you need to fuse every cell because shorting probability increases.
Every wire must be fused for safety.
I dont understand why BMS wires are rarely fused.
I always fuse my BMS wires (1A glass quick fuse)

And i dont use common BMS wires (-+-+-+-+-) my next pack because if poor connection, the BMS may malfunction by over voltage or burst into flames.
I use 2 wires every cell like (-++--++--++--++-)

I also use fuse end of pack but 72+ voltage DC fuses are bit bulky, expensive and hard to find though.
AC circuit breaker is fire risk because of arcing.
24v car fuse melts and burst into flames @72v

 

[Lurkin]

Does it matter where physically it is connected, i.e. closer to the cells or closer to the BMS? I would have thought the cells.

(Yes, its a decent sized pack with a BMS, or will be. 13S15Pish to be confirmed when the frame finally turns up)

 

[Punx0r]

I dont understand why BMS wires are rarely fused.

I think they tend to be considered self-fusing as they're typically very small gauge wires. Not ideal, though, I agree.

 

[Lurkin]

Direct from em3ev's website:

PPTC thermal fuses used on all signal wires. If a fault were to ever occur on any of the small balance wires which run around the battery pack, the fuse simply shuts down the connection and keeps everything safe. Once the fault condition is removed, function returns to normal.

http://em3ev.com/store/index.php?route=product/product&path=35_53&product_id=161

I take it signal wires = balance wires from BMS? This is exactly what I am planning to do.

 

[Lurkin]

Interesting. So I've been reading up to work out what fuse is correct (yes, covered many times on this forum, but not always particularly well) for 48v.

Many because there aren't many options that are specifically rated for 48v. Most seem to be happy with a blade or ANL style 12v fuse and the concern is when the fail occurs and the fuse wants to open, using a higher than rated voltage means there is a risk that despite breaking the circuit, due to the higher voltage, it will just arc across the open fuse anyway.

The answer to this is usually downplayed, stating that's an irrelevant issue that doesn't occur until the voltage is approaching 100v +

Has anyone deliberately destructively tested a auto blade fuse to be sure of this theory? I would really like to use a regular 40a blade fuse because they are accessible, cheap and easy to replace if it blows?

 

[Syonyk]

Hm. I probably could blow a low amperage one on high voltage to check.

I don't think 48v is enough to really change the arc risk from 12v. I wouldn't use one on 240v or something, but 48v is probably fine...

 

[Lurkin]

Would be awesome if you could. Happy to buy some and post them to you for testing if that would help (depending on where abouts you are).

I've actually read through your blog before as I was considering buying one of the ebay 18650 tester boards - the one that had data logging and you found it to be inferior. Any suggestions on a better way to test for capacity/IR with logging? I tried unsuccessfully to buy an RC charger for this purpose, but HK kept misadvertising their crappy chargers, stating they had USB logging but on arrival would just have empty USB ports.

 

[Syonyk]

Would be awesome if you could. Happy to buy some and post them to you for testing if that would help (depending on where abouts you are).

I can just run by NAPA and grab some 3A or 5A ones to blow. Not going to break the bank, though I'm not sure how I'll connect to them. Probably just solder to the leads.

I've actually read through your blog before as I was considering buying one of the ebay 18650 tester boards - the one that had data logging and you found it to be inferior. Any suggestions on a better way to test for capacity/IR with logging?

No. :/ I haven't found one with logging yet, and while I will at some point, it's likely to be a higher voltage unit aimed at testing full packs.

 

[oatnet]

- Fusing on the positive on the output/input of the entire battery (some seem to use a circuit breaker)

I believe best place for a pack fuse is in the middle of the pack.

-JD

 

[Lurkin]

?

What does that actually mean? between the series connections?

 

[oatnet]

?

What does that actually mean? between the series connections?

Yep. I learned this from professionally built packs in '06. For example, if you have a 72v pack made up of (2) 36v sub-packs, put the fuse between them. In the event of a failure, this divides the voltage into (2)36v packs which is much more manageable. Say you have a fuse on the positive lead of a 24 cell LifePO4 pack, but have a frame short to cell #23, or a pack short from cell #1 to cell #13 - that fuse will not blow, so your pack will remain shorted and unpleasant things will happen. With the fuse in the middle, it will trip in both scenarios and protect the pack without any additional hardware. If you look at my beach cruiser build, I built two 7s packs into the same brick, and connect them with a fuse at one end and power takeoffs at the other.

If cell #12 shorts to the frame or cell #1, then even a mid-pack fuse won't help you - but now you have divided the voltage so your largest risk is a dangerous 36v short instead of a deadly 72v short, and half of your pack might survive.

The next level of protection is to also put fuses on both ends as well as the middle, to reduce the risk of a single-cell to frame short, so your largest exposure is within a subpack. That is what I use on my 42s dune buggy pack, and I also have a fuse on every balance wire as close to the cell as possible.

Ultimate protection would be a a fuse on every cell, but I have not found that to be practical. :lol:

-JD

 

[amberwolf]

?

What does that actually mean? between the series connections?

That's where the fuse is on my A123 EM3EV pack--between the two halves.

 

[Lurkin]

Is there a degree of resistance incurred with fuses?

The plan is to fuse each sub pack before putting into series, each balance and an overall fuse as well.

I'm having a hard time knowing which fuses are appropriate. There seems to be slim pickings at 48v, most seem to be using 12v fuses. Luke has advised in other threads to use fast blowing fuses, which seems like a good idea. But which fuses to use?