relays

This gets better and better. I'm almost glad this silly thing got started.

Apparently a battery-powered product that uses lots of power has no advantage one one that uses very little power. Thanks for the education.

How can you save energy over time without saving power?

The mind boggles.

John

Absolutely. And water-powered closed-loop level controllers have been around for a century or two. They are called "toilets."

John

message

That'd

This *is* getting better and better, but not in a way that makes you look good.

Alas, I suspect that you remain uneducated.

Unless the power required is more than the peak current of the battery supports, it is 100% true that a battery-powered product that uses lots of power has no advantage one one that uses very little power. I have designed created millions of toys that run off of batteries, and never have I or any other Mattel engineer had any cause to minimize the power. We all, however, care a great deal about how much *energy* the toy uses. If not for your willful ignorance concerning the difference between power and energy, this would have been obvious to you. A toy that uses a tenth of a milliamp at a 10% duty cycle has no real advantage over a toy that uses a hunredth of a milliamp at a 100% duty cycle. The *power* will differ by a factor of ten, but nobody cares about that, because the *energy* use is the same, and the battery life is the same. BTW, those are not made up numbers; that's about the current draw and duty cycle that we designed into the 'N Sync Talking CD Case.

(For the pedantic, there actually are some differences, because IR drops are different, batteries act a bit differently when pulsed, and a really weak battery might be able to put out a a hunredth of a milliamp but not a tenth of a milliamp. The above assumes ideal components.)

By changing the duty cycle, as you did when you designed a latching relay into your thermostat. You made no change to the power. None! You reduced the enerrgy by a huge amount. Thus, you saved energy over time without saving power. This has been explained to you before.

It certainly does, but not in a way that makes you look good.

Care to explain why it is that *everyone* here is telling you that you are wrong and *nobody* here is telling me that I am wrong? Nobody seems to have minded correcting my mistakes in the past (and, BTW, I thank them for doing that), so why not now? Could it be because -- try to wrap your mind around this -- because I am right and you are wrong?

Hardly "everybody." Only a few people who can't accept the idea of time-averaging power, and who won't post equations.

So, what's your equation for the power gain of a latching relay?

John

"Average Power" is just another way of saying "Energy." You can't average without time, and time is the difference between power and energy. This sort of modifier is used in many other areas of life; for example, miles per hour is a measure of speed, miles are a measure of distance, and time makes the difference.

Name someone who agrees with you. Anybody? Anybody?

Pout/Pin, using the actual dictionary definition of "power" for P on both sides of the equation,(Not, as you originally posted when you made your mistake about latching relays having infinite gain, joules in and watts out -- the joule is a unit of energy, not power).

Heard of him. never met him. I have been out of Mattel as an actual employee for years, but I have done a bit of contract work for them and other toy companies since then.

You would like Jaaks Pacific. The engineering department is right on the beach in Malibu.

Unlimited numerator, fixed denominator. Sounds pretty big to me.

John

So there's no difference in power gain between a regular relay and a latching relay. Right? So there's no reason to use a latching relay in, say, a battery-powered thermostat, or in a thermocouple switching circuit.

John

And 200 microjoules into a latching relay can switch 4.3 megajoules after one day. Energy gain = 22e9, averaged power gain = 22e9, bandwidth 12 uHz.

Relays are neat.

John

Please tell us more about leftist weenies.

John

Nonsense. Time-averaging is dimensionless scaling. But "include" is algorithmically obscure.

Gosh, thanks.

Sure I can. It's numerically equal to energy_out/energy_in. Gain is dimensionless too. Divide both the numerator and the denominator by the time interval and you get average_p_out/average_p_in, which is the numerically same gain. That's bog simple.

Not many RF engineers define the gain of a power amp at one picosecond instant in time; they average it over the signal waveform. But YANARFEE.

That

Now *you* are confused about energy and power. The ratio of power/energy has engineering units Hz.

As noted, they are numerically equal and dimensionless, so are exactly the same thing.

Always have!

John

Of course not. That would be like an audio amp that's unplugged. You can't measure power gain without a load; or, to be pickey, it's zero, or undefined.

John

K/1 = K for all K.

Whether any number is "infinite" is a philosophical issue. People who do accept "infinity" as a number also accept that infinity/1 = infinity.

As an engineer, I accept an unboundedly large number as infinite.

John

sufficient

Cool stuff. Thanks.

John