Deprived, speaking of investing, have you considered in future offerings (i.e. when the current batch goes to zero) limiting the issuance size? I ask because if this becomes a popular instrument you might end up having cash drag become a problem for later iterations. And, as uncomfortable as this may be, the counter-party risk increases as the issuance grows. There is both increased incentive for your to take the money and run (Rude, I'm sorry) and for hackers and others to try and abscond with the money. The downside is that this would hurt your management fee and limit liquidity and increase miss-pricing of the pre-existing mining and selling (as there is no more ask wall on purchase).
There's an effective cap on capital controlled anyway.
As time passes, sales of PURCHASE will become increasingly small when measured as a percentage of existing sales (that HAS to be the case unless sales of PURCHASE continually increase in volume). At a certain point that percentage will fall below the percentage of capital which is given out as dividends each day.
Capital is kept below 410 days of MINING dividends. So on any given day at least 0.24% of capital will be returned as dividends. That immediately imposes the first cap - that capital will cease to increase once sales of PURCHASE fall below 0.24% of exisiting effective outstanding units. In practice a cap would be reached far sooner than that if difficulty continues rising - as SELLING will also receive dividends (I can't calculate a percentage for them - as that requires knowing future difficulty : and if that were known that this fund couldn't even exist).
As far as CP risk is concerned, the market can take care of that itself. Specifically, if potential investors believe the CP risk for me personally is being neared then they will presumably stop buying PURCHASE and hence impose an effective cap. There's rather obviously no benefit to me in imposing an arbitrary limit below what the market is willing to accept. If someone invests then subsequently sees market cap rise significantly - to above or near their personal tolerance - then that unexpected rise (and it would have to be unexpected or they rationally wouldn't have invested in the first place) will have massively increased liquidity allowing them to sell out at minimal loss. In fact it will probably let them sell out with MORE profit than they anticipated - as the huge sales of PURCHASE that they didn't anticipate will have increased NAV/U a bit (or, more likely, slowed its fall relative to dividends distributed).