OK, you've probably already considered this, but:
Maybe you should split the project up into more parts, get a figure for the translation and markup/typesetting and offer the initial .pdf of that as a 'reward' - call it Part 1. Then have a second Kickstarter for getting the actual dead-tree version that does not need to kill electrons to read?
I'd also stick with Kickstarter or a similar funding service that does not charge the donor until the actual goal has been met. A lot less trouble for you to try to refund money if, for some reason, the Kickstarter does not make the minimum goal.
Several other 'services' are particularly associated with scammers who disappear with the funds even when the minimum goal is not met, and charge the potential donor as soon as they sign up.
Thank you for your input, Bruce!
The current planning is to get just the translation and translation editing financed. The problem is that margins in book production and book sales are not very high, so the ratio between what you see as a figure and what actually ends up with me to fund the translation is not good - which means I'd have to set a ridiculously high minimum funding goal, which will decrease the chances of overall success.
It looks like there is a way around, though, and I'm hoping it will work out - we can offer discount codes as rewards, so you as a backer will get the book for say, 20% off. That takes the cost of dead trees out of the equation and makes things much more realistic-looking.
I will definitely stick with Kickstarter, as it's the best-known portal and I have already backed things there. The other platforms seem to have similar charges, too, so there's not much of a difference in that regard. And yes, minimum funding needs to be reached.