[RE-wrenches] Fuse sizing in battery circuits
Kent Osterberg
kent at coveoregon.com
Sat Apr 6 12:21:08 PDT 2013
Phil,
Thanks for the code references, they are very useful and point out an
area where I fear many installations of battery based inverters are
deficient. You are right the manufacturers have not made this easy!
First, data on the short circuit current available from most batteries
isn't easily available, if available at all. Concorde publishes short
circuit current numbers for the SunExtender batteries. Neither Trojan,
Rolls/Surrette, nor Outback publishes short circuit current data for
their batteries. Full River publishes an internal resistance number that
can be used to estimate the short circuit current. Taken as a whole,
battery manufacturer's certainly haven't made it easy to comply with 690.71
Second, sometimes interrupting capacity data on the breakers being used
in BOS equipment is also missing. Take a look at Outback's catalog and
try to find the AIC for the breakers used to protect the inverter
wiring. Is it published anywhere else? Should a system designer have to
buy the Outback breaker, find out who manufactured it, and then search
for the correct ratings in the in the breaker manufacturer's catalog?
Obviously not.
Third, BOS manufacturers often connect circuit breakers for charge
controllers or dc utilization circuits that have AIC ratings of only
5000 amps directly to the battery. So while the inverter may be
protected with a breaker having an AIC rating 25,000 or 50,000, other
devices don't. BOS manufacturer's should include on the dc breaker panel
data showing the lowest AIC breaker that is connected directly to the
battery. Then installers would have at least half of the information
needed to decide when a fuse is needed to backup the dc breakers. And
shouldn't the instructions for BOS equipment point out that a fuse may
be necessary at the battery.
The short circuit current from a single string of golf cart batteries or
L16 batteries is probably less than 5000 amps. Fuse(s) are probably
needed in most installations with larger batteries or battery systems
with multiple strings of batteries. 2011 NEC 240.21(H) allows
overcurrent protection to be installed as close as practical to battery
terminals - even in classified locations.
Kent Osterberg
Blue Mountain Solar, Inc.
www.bluemountainsolar.com
On 4/6/2013 7:29 AM, Phil Undercuffler wrote:
> Ray,
>
> 690 does offer the protection, it's just that most maufacturers
> haven't made it easy.
>
> 690.71 Installation
>
> (C) Current Limiting. A listed, current-limiting, overcurrent device
> shall be installed in each circuit adjacent to the batteries where the
> available short-circuit current from a battery or battery bank exceeds
> the interrupting or withstand ratings of other equipment in that
> circuit. The installation of current-limiting fuses shall comply with
> 690.16.
>
>
> The telcom industry has been doing this for years. Their norm is to
> have each string of batteries on a shelf, and terminate at a breaker.
> Each string then is combined at a bus bar or plate, to help ensure
> equal current on each string. The loads and charging sources all come
> to that same bus bar. It has some advantages, as individual strings
> can be monitored and serviced without taking the entire system down.
>
> FWIW, OutBack makes a battery rack for AGM batteries with series
> string overcurrent and disconnects on every string. It's also to my
> knowledge the first UL1741 Listed device available -- the other stuff
> I've seen is either not listed, or is trading on old 508a standards.
> AGMs might not be for every installation, but for jobs where they are
> appropriate this could be a good, code-compliant solution.
>
>
> Phil Undercuffler
> OutBack Power
>
>
>
>
> On Friday, April 5, 2013, Ray Walters wrote:
>
> I just finished a rewire and we kept the Ananda power center,
> Allan :-)
> Back when we used class T fuses more they were always over sized
> relative to Heinemann breakers.
> 400 amp class T fuse = 250 amp breaker= 4/0 cable
> 200 amp class T fuse = 175 amp breaker= 2/0 cable
>
> I still think class T fuses are superior in some ways to breakers:
> they have better interrupt capacity and trip faster in a short
> circuit condition.
>
> However, If you spend too much time thinking about and looking at
> the trip curves, you'll break your brain and special order some
> weird fuses and breakers. (anybody need a 350 amp class T fuse?)
> Just follow the inverter manufacturer's recommendations and all
> will be well.
> On the other hand, William has brought up a topic I have harped on
> for years: Having the OCPD in the cabinet doesn't protect the
> majority of the circuit. Class T fuses at the battery terminals
> do, but they're not rated for the corrosive environment. I have
> thrown a few class T fuses away that had acid eating away at the ends.
> I don't have the solution, but I will continue to point out that
> this is a real problem. Dropping a wrench across the battery
> terminals can lead to a spectacular failure that not only can
> cause a fire, but might even cause a battery explosion, yet NEC
> offers no protection.
> We use insulated wrenches from experience, and hope for the best.
>
> R.Ray Walters
> CTO, Solarray, Inc
> Nabcep Certified PV Installer,
> Licensed Master Electrician
> Solar Design Engineer
> 303 505-8760
>
> On 4/5/2013 8:29 PM, William Miller wrote:
>> Friends:
>>
>> Good topic. Some questions:
>>
>> 1. Most manufacturer's present an installation guide that shows
>> one OCPD in the battery circuit and that is in the BOS cabinet.
>> This means the battery leads are unprotected. Do we need an OPCD
>> at the battery terminals?
>>
>> 2. Class T fuses are generally recommended for this application.
>> The data shows them as "fast acting." Is this a problem? Will
>> they act too fast and open during normal surge loads?
>>
>> Thanks in advance!
>>
>> William Miller
>>
>>
>>
>>
>>
>>> Troy,
>>>
>>> Overcurrent device size is matched to the conductor size. The
>>> inverse time constant nature of an overcurrent device can
>>> typically handle the surge currents as long as conductor sizing
>>> has truly been done correctly for the conductor. Circuit
>>> breakers are preferred to fuses because they can be reset.
>>>
>>> There has been volumes written on this issue. The constant
>>> current at lowest battery voltage should be used, plus the ac
>>> ripple content on the battery circuit. This is usually a much
>>> larger conductor than your average designer will plan for. The
>>> best thing is to look at Midnight, Outback, and Schneider and
>>> see what size overcurrent devices they require for their
>>> products. That will give you a good clue as to how to size the
>>> conductor and overcurrent device.
>>>
>>> Bill.
>>>
>>> *From:* re-wrenches-bounces at lists.re-wrenches.org
>>> <javascript:_e({}, 'cvml',
>>> 're-wrenches-bounces at lists.re-wrenches.org');>
>>> [mailto:re-wrenches-bounces at lists.re-wrenches.org
>>> <javascript:_e({}, 'cvml',
>>> 're-wrenches-bounces at lists.re-wrenches.org');>] *On Behalf Of
>>> *Troy Harvey
>>> *Sent:* Friday, April 05, 2013 3:38 PM
>>> *To:* RE-wrenches
>>> *Subject:* [RE-wrenches] Fuse sizing in battery circuits
>>>
>>> I've got a question about battery string fusing. Typically we
>>> size the wire from the batteries to the inverter based on
>>> continuous rating procedures (max power/efficiency)*125%.
>>>
>>> However a 6kW inverter, can peak at 12kW for 5-10 seconds,
>>> doubling the source current. That is no big deal for the wire,
>>> because it is a short time frame... little heat will be
>>> generated. However, in fusing the sub-strings, you need to
>>> account for that peak surge current so you don't blow fuses all
>>> the time. But if you put a 500-1000 amp fuse on a 4/0 wire,
>>> above the max surge draw of the inverter, the wire will be
>>> under-protected for its ampacity rating. Any thoughts on the
>>> catch-22?
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org/attachments/20130406/e054e4a1/attachment-0003.html>
More information about the RE-wrenches
mailing list