I've been trying to find out whether any altcoin community has attempted a hardfork without consensus before.
Can you point me towards an account or analysis of such an attempt?
Only example that comes to mind is Elacoin. Steve Sokolowski covered the fork in his forum post on the block size increase debate:
Everyone can veto the change
A common misconception about this block size increase is that miners will make a decision about whether the change is accepted or not. Miners only have the power to prevent a hard fork, but they cannot force its acceptance even if they mine new fork blocks.
To see why, consider the story of Elacoins. Elacoins were a proof-of-work altcoin created several years ago. After some initial interest, activity declined until there were no more development updates, and difficulty plummeted. However, after about six months, a new developer came onto the scene and decided to fork Elacoin to be proof-of-stake. He released a new version with great fanfare and many future stakers arrived to use the new coin. He set a date in late 2014, at which the chains would diverge. In anticipation of the upcoming change, volatility increased and trading volume rose dramatically.
Exchange Cryptsy, seeing the trading volume, had no reason to make any change to its market, and CEO Paul Vernon ignored customer support tickets from the new Elacoin developers. As price rose in anticipation of the fork, difficulty also rose. The date of the fork came and went, and price continued to rise - but Cryptsy did not upgrade its daemon. As a result, mining continued on the old fork. Huge buy orders of 5+ bitcoins appeared. A large group of people continued to stake the new fork, but the only people who actually mattered - Crypsty, ignored their tickets. Eventually, the hard fork was declared a failure and Elacoins continued to be traded and mined using the old fork, despite almost every user, developer, and staker wanting to use the new fork.
This story demonstrates that exchanges, not miners, will also decide whether to accept any proposed changes. In fact, almost anyone can veto a hard fork. Just 20% of miners can decide not to mine the new blocks, and the fork fails. Just one massive exchange (Coinbase, or the Bitcoin Investment Trust) can continue trading the old fork, and the fork fails. A few of the richest coin owners, like Tim Draper, can decide that they will continue to make their bitcoin investments on the old fork, and the fork fails. Big merchants like Overstock and Microsoft announce that they will only accept the old fork, and the fork fails.
Miners do not have the final word, or even a significant choice, in whether the fork succeeds. There are several lessons to be taken from the Elacoin story: first, miners like Slush and Luke Jr. do not have the power to cause this fork to happen, only to block it. Second, exchanges are more important than miners in whether the fork succeeds or not. Third, a bare majority of people accepting the fork is not sufficient for it to happen, since the default decision is to fall back if it looks like there is even the slightest chance of failure.