When two of the same vowel come in a row, the sound simply lasts twice as long. The vowel sound doesn’t change - it just lasts longer.
The extended version of e is usually written ei (not ee)
Likewise, the doubled version of o is usually written ou (not oo)
There are a few exceptions to this, but the rule holds true 95% of the time.
In hiragana, you would add another character to extend a sound.
In katakana, you extend a sound by simply writing a horizontal bar.
Ex. ああ and アー
Double consonants have the same concept as doubled vowels. When two of the same consonant appear next to each other, the consonant lasts for 2 beats instead of one. You do this by pausing on the consonant for an extra beat. Therefore, in the word metta, you would pronounce me, then take your tongue up into a position for the ta sound, but before doing that you freeze right there for one beat. Finally, you say the ta sound.
When doubling the consonant n, you would not use the small tsu. This is because there is a character that makes the n sound all by itself.
せんえん and せねん may both seem like they should be written the same way in Roman letters, however, this is not the case. The pronunciation of these two words are also different.
The character n can only be used as the end of a unit of sound, not the beginning of a new one. If the next character happens to be a vowel sound, like in せんえん, the characters do not merge into an n-series sound (n plus e does not equal ne).