This is not a simple question, despite how simple it may seem...  The right reference is your governing Timber Design Standard/Code.

You should probably be looking for real world help, unless this is in a mechanical device, at which point real world help would probably be very much worthwhile in any case...

Good luck,

YS

B.Eng (Carleton)
Working in New Zealand, thinking of my snow covered home...

If it is joist in bending, then the repair should generally be full length, using bolts, not nails.  If the member is in tension or compression only (truss member) you can "scab" repairs onto the side, usually 18 inches beyond the break or splice point, and nailed at 4 inches o.c. staggered top and bottom.  Check the stresses before you do that though, so that you can size the nails. Generally won't be less than 12d nail though.  The National Design Standard (NDS)for wood in the US, provides guidelines for the shear capacity of nails, thru-bolts, and lag bolts in wood.  If you don't have access to a copy, you might try a local university.

These are very general guidelines and you should make sure you check the stresses and if you have questions, check with a local structural engineer to check the analysis and code provisions.

Good luck.

The fasteners on each side of the splice must be designed for the bending moment, shear, and axial force at the splice.

DaveAtkins

You have to remember that wood creeps with time and a moment splice probably isn't advisable.

And ron9876 has it...

This is possible, but the design standards which I am familiar with and/or have used (CAN/CSA, AS, NZS, BS, EN) either DO NOT permit this type of connection, or more commonly, severely penalize and apply high load factors with low resistance factors.

Good luck!

YS

B.Eng (Carleton)
Working in New Zealand, thinking of my snow covered home...

Nails?   Unless you are not going to carry best.

very, very low loads and small bending moments for a short time, nails are not the best solution.

Through-Bolted, with a splice plate on BOTH sides - and the splice plate glued/epoxied to the two wood beams, is

The glue prevents rotation of the plate against the wood at the surface.  
The through bolts "compress" the wood beams over a wider area than bolts and washers (and much, much more than one-sided screws or lag bolts - both of which will loosen as the wood shrinks over time.)  They carry shear load.  
The moment loads are resisted by the area glued and the pins (bolts) on both sides of the splice.

How many bolts of what diameter?  How thick/wide/long a splice plate?   Depends on your loads.   At least 4 bolts (two on each side of the splice) - probably more.  

Maybe.  8<)

Thanks to all responders.  While my Kent Mechanical Engineering Handbook has all sorts of formulas for nails and bolts subjected to tensile and shear loads perpendicular and parallel to wood grain, it seems that I have not come across analytical work with nailed connections.  I will be checking that Timber Design Standard,though.