3 ms·
> That a new fridge is more energy efficient than an old one is not the point, because, of course, you can always choose to have a new one while the old one sti
by Paul-ish 9y ago
> That a new fridge is more energy efficient than an old one is not the point, because, of course, you can always choose to have a new one while the old one still works, for whatever reason. The point it: that things was built in a way that it lasted 60 years.
Isn't it possible that the complexity that makes something energy efficient also causes it break sooner?
- TeMPOraL 9y agoIt is. For instance, if elements that move under power are changed from metal (heavy) to plastic (light), you need less power to move them. However if making something more energy efficient causes it to break more often, you have to expand the scope to include the energy costs of manufacturing a replacement copy. Even if the marginal costs of manufacturing an appliance are so low repairs no longer make sense, I still doubt that e.g. more efficient washing machine that you have to replace 3 times over 10 years uses, in total, more energy than one less efficient washing machine that doesn't break for 10 years.
- logfromblammo 9y agoNo. The complexity does not make anything break. You only have to engineer something once, and coordinating the failure rates of the different subsystems is part of the engineering. If anything, managing the additional complexity by skimping on the engineering work is what causes the failures--that is, engineering a more complex device with the same or smaller engineering effort. The ideal situation when engineering a new device is that every part of it is likely to break simultaneously, the day after the warranty runs out. If that happens, you know that your device was made as cheaply as possible, but no cheaper, and that you will never need to repair such a device, because buying a new one will always be cheaper. Repairs are only an issue when different parts are likely to break at different times.