Was wondering what common practice is for doing an arc flash analysis where some of the PD are soo old there`s no digital records available. I recently came across some CEFCON fuses and everyone I talked to said there was no data available other than possible on paper in a filing cabinet somewhere.

I found documentation (TCC, etc.) for Aeroflex HRCI-J fuses through Fusetek (thanks again Malcolm!). They sent me a scan of the datasheet.
It's not 100 % digital, but it's still usable.

If digital records are not available, I presume some paper records are. Using a printed TCC you should be able to determine the clearing time for both max available fault current and arcing fault current levels. The clearing times may be entered manually into an analytical program to obtain the results.

Or, find a modern fuse that is within the software library, that has a TCC that matches the old fuse; and use that.

You can try [url='http://www.electricalmanuals.net/']http://www.electricalmanuals.net[/url]. If not there I generally indicate it is an obsolete PD and do not enter an PD info. I recommend it be replace with something newer.

I would use the 2 seconds rule. It's usually assumed that the outer limit for the arcing time is no more than 2 seconds. Although this is not a hard and fast rule, it accounts for the likelihood that the arcing material in an arcing field will likely be either burned off or expelled by the force of the blast. In any case, this would extinguish the arc event. Also, IEEE 1584 states "If the time is longer than two seconds, consider how long a person is likely to remain in the location of the arc flash. It is likely that a person exposed to an arc flash will move away quickly if it is physically possible and two seconds is a reasonable maximum time
for calculations." [ IEEE 1584 Annex B page 76]

The fuse is a CEFCO . Two solutions:

1 Use a Bussmann FRS, K9 TCC.

2. Replace with a RK1.

For old circuit breakers, I assume the breaker would take 2 seconds to operate. In some cases I have seen old power circuit breakers that have not operated in 30 years. For these, I have been modeling them as switches. Monster Fuses has a good cross reference for old fuses. If you can not find one on their website, their customer service folks have been very helpful.

1: I don't get that implied attitude of "if it isn't digital, it doesn't exist."
So what if you don't have an electronic copy?
Model it into your software from the paper copy.
I have modelled lots of breakers where I could find no digital copy of the tccs.

2: As for Arcad's comment, there is no "2 second rule". There is a comment, not a "RULE". Personally I think it is a bad idea to have that in the standard. There are lots of situations where 2 seconds isn't near long enough, and others where 2 seconds is way too long. The way I read the IEEE text, it is requiring that you "consider " the time, not take 2 seconds as a blanket "rule".

I really appreciate all the great responses. Lucky the client was willing to replace the fuses. But I`m starting a new project now where part of the facility was built in the 60`s so all these new resources will be invaluable.

Hi, I use the approach if I have a paper copy is to scan and make it a PDF, copy with as large a scale as possible and paste into Excel. I then recreate the axes and the data using a chart and overlay it over the image. The image needs to be positioned behind text. I can then fine tune my points to get operating curves to suit and make them as close as possible.
If you send me the details, I am happy to plot for you, if I am also able to keep a copy for my records.
Other options for sourcing data is to find out what happened to the company, chances are they were bought out by another company and that new company may have the records somewhere. Or if the fuse are still available, use google and find a supplier who may have the data for these items.

I guess this is an excellent solution, especially if you do not have to many situations like this in a project.

But, if you have many unknown models, the cost of searching and developing curves will be too high.

When this is the case, I only dedicate a maximum of 30 minutes to find the curve. If I do not find it in that time, I do not assign a library to the OCPD.

You may consider using Arc Flash Analytic V1.0 mobile app for your application. The app features build-in and extendable circuit protection device library. You can populate the app with fuse and circuit breaker data including time-current characteristics for the devices that are not already listed in the built-in protection device library. All you need to add a fuse is a set of five (5) points from the fuse average melt time-current characterstics. The app will use the points to analytically interpolate the characteristics in the 0.01 to 10 seconds time interval and determine arc duration for both the predicted arcing current and the reduced arcing current.