5 ms·

I think the video makes a clear argument to the increase of number of severity of failure modes in the more modern part, I don't feel that it does a great job a

I think the video makes a clear argument to the increase of number of severity of failure modes in the more modern part, I don't feel that it does a great job at a lot of things. So both designs likely have a life limiting factor of oil. It's clear that the older design will have a longer life before running into oil related issues. However, he still needs to modify (punching a hole in the housing) the motor to oil the unit. He makes no affordances to the modern unit for doing so. While it might be a pain to have to re-oil more frequently, it's still likely to be an action that occurs on a multi-year basis. At that point, the difference between re-oiling every other year, and every 10 years feels pretty academic. He discounts the ~6W of energy savings. It's true its not a lot - especially if you take the perspective of - 'let's try to save 6W of power from a 1970's fridge'. However, modern units consume the order of 200W, so an increase of 6W is a 3% increase in energy consumption. Maybe you can justify that for a single component, but how would you justify that across the board? Do wish they used better caps though.

TeMPOraL 7y ago
It's 3% difference in energy consumption, at a cost of replacing a fridge much earlier. Fridges being replaced more often means fridges being made in much larger quantity, which eats up all potential energy savings and then some.
Spooky23 7y ago
The energy argument is valid, but I think the trade off is a result of poorly thought out regulation. For a water heater, there is an energy guide as well as a durability guide. The good/better/best ratings help the consumer make an appropriate decision. Energy star should be similar. I’m not interested in saving $4/yr in electricity if that means buying a new appliance in 4 years, but I have no way to evaluate the quality of the device.
- icegreentea2 7y ago
  That's a reasonable wish. I imagine the usual thought is that warranty periods should provide a reasonable proxy for durability information, but I also understand that it's a pretty imperfect proxy.
  - Spooky23 7y ago
    I'd argue that OEM extended service plans are the real proxy, at least for a service call. Look at Apple as an example, two years is the max service plan for an iOS device. That's the lifetime of a battery. For Dell/HP laptops, they will have 3 year warranties with up to 5 years of extended, but usually exclude the battery and often exclude the screen.
djmips 7y ago
I agree with all your points; it's what I was thinking along while watching the video. I also don't accept that hypothesis that in general newer designed motors fail earlier than older motors. I don't even accept it for being told that these two particular motors chosen at random. Will the newer motor in this video will fail before the older design in this video? Perhaps but I'd like to see some real evidence.
daniel-cussen 7y ago
> At that point, the difference between re-oiling every other year, and every 10 years feels pretty academic. You missed some stuff. It was one oil change for 50 years, not 10, and second, whereas the old fan would stall and heat, but not burn out, when it failed, the new fan would fail irreparably if oil ran out. And it would fail without warning. The type of failure mode is important here. And that's just if the oil runs out. Of course, being crap in an egalitarian way, it is designed so that the capacitor would fail, or the fuse would blow, or the IC's would short, or the windings would burn out, all at precisely the same time. It's mean-time-between-failure engineering.

TeMPOraL 7y ago

It's 3% difference in energy consumption, at a cost of replacing a fridge much earlier. Fridges being replaced more often means fridges being made in much larger quantity, which eats up all potential energy savings and then some.

Spooky23 7y ago

The energy argument is valid, but I think the trade off is a result of poorly thought out regulation. For a water heater, there is an energy guide as well as a durability guide. The good/better/best ratings help the consumer make an appropriate decision. Energy star should be similar. I’m not interested in saving $4/yr in electricity if that means buying a new appliance in 4 years, but I have no way to evaluate the quality of the device.

icegreentea2 7y ago
That's a reasonable wish. I imagine the usual thought is that warranty periods should provide a reasonable proxy for durability information, but I also understand that it's a pretty imperfect proxy.
- Spooky23 7y ago
  I'd argue that OEM extended service plans are the real proxy, at least for a service call. Look at Apple as an example, two years is the max service plan for an iOS device. That's the lifetime of a battery. For Dell/HP laptops, they will have 3 year warranties with up to 5 years of extended, but usually exclude the battery and often exclude the screen.

djmips 7y ago

I agree with all your points; it's what I was thinking along while watching the video. I also don't accept that hypothesis that in general newer designed motors fail earlier than older motors. I don't even accept it for being told that these two particular motors chosen at random. Will the newer motor in this video will fail before the older design in this video? Perhaps but I'd like to see some real evidence.

daniel-cussen 7y ago

> At that point, the difference between re-oiling every other year, and every 10 years feels pretty academic. You missed some stuff. It was one oil change for 50 years, not 10, and second, whereas the old fan would stall and heat, but not burn out, when it failed, the new fan would fail irreparably if oil ran out. And it would fail without warning. The type of failure mode is important here. And that's just if the oil runs out. Of course, being crap in an egalitarian way, it is designed so that the capacitor would fail, or the fuse would blow, or the IC's would short, or the windings would burn out, all at precisely the same time. It's mean-time-between-failure engineering.