8 replies to this topic

[Mark W. Ingalls]

I thought it would be a good idea to begin a discussion on switches in general and power switching with relays in particular. There won't be any product recommendations (I hope) only technology.

Begin with these excellent Wikipedia articles...

Switch

Relay

Solid State Relay

Schottky Diode

[Mark W. Ingalls]

A switch is nothing more than a means of connecting and disconnecting two conductors-- A wire nut on two wires could thus be considered to be a crude switch.

Let's stop to consider the wire-nut-as-switch idea further: Not twisting the wire nut onto the conductors tight enough could cause a connection that gets hot with high current and trips the breaker or even causes a fire. In a similar way, switch contacts must also be intimate if they are to pass high currents reliably. So, just as there might be a proper torque for wire nuts (and there actually is a torque specification for lugs in service panels) there is also a proper contact force for a switch.

Almost everybody knows it takes a larger diameter wire to handle higher currents, but what about the area of the switch contacts? When one stops to consider it, larger contacts for higher current makes sense.

So, a high current switch would have larger contacts and heavier contact force. This means closing and opening a high current switch would probably take more work as well-- we're trying to move a larger object that is connected to a stiffer spring. If we wanted the switch to be remotely operated, e.g., a relay, we would have to expect the relay to perform more work in opening|closing the bigger switch.

(Work is directed energy. Power is energy per unit time. You can convince yourself of this by realizing that a Watt is a unit of power, and the power company sells you electricity by the kilowatt-hour.)

So, big relays consume more power than small ones, regardless of the amount of current passing across the contacts.

There are two other things to consider when interrupting the flow of current using a switch. First, the electricity flowing through the switch has momentum. Stopping the electrical momentum by opening the switch will take more work than closing the switch. Second, this electrical momentum will actually cause a spark to be thrown across the opening switch (but not the closing switch). Thus, turning off lights actually presents more of a fire hazard than turning lights on!

When you start your engine, your starter demands a very high current for a relatively short period of time-- like up to 300 amps when the starter is first actuated -- so the starter solenoid has to be able to slam together with lots of force, but it also has to interrupt the current with a lot of force. Force -> work -> power.

Electrical power can be calculated as:

P = I*I*R

So reducing the solenoid resistance by winding it with bigger wire will give it more contact force, but it will also cause the solenoid to consume more power in doing its job. (All solenoids consume power in order to do their jobs.) But, reducing the solenoid resistance can also cause it to produce more thermal energy, and thermal energy over time can cause the solenoid to wear out.

A continuous duty solenoid either has more winding resistance to limit the current through the windings or has some kind of heat dissipating design (heat sinking, so called) to lengthen its life. The trade-off is lower current handling through the contacts, or a bigger more expensive solenoid.

____________________________________

Before we conclude this post, you might be asking yourself, "Why is it that small wires supplying current to a load, e.g., the "house" battery would get hot, while big wires in the solenoid coil would get hot?"

The answer involves a trick-- In the case of the solenoid, all the charging battery's voltage is delivered to the solenoid coil. In the case of the supply wire and "house" battery, most of the charging battery's voltage is intended delivered to the "house" battery, and only a small amount is intended to be delivered to the supply wire. So the presence of the "house" battery limits the current in the circuit. But the solenoid coil is the load. Making the coil wire smaller limits the current and thus limits the heat dissipated in the wire.

[HERR42]

technically a solenoid has nothing to do with a switch. automotive solenoids are really relays.
I wonder how the name got so....bastardized ?

[Mark W. Ingalls]

Of course you are correct, HERR42. A solenoid is merely a geometric shape.

Similarly, we use the word 'coil' to describe an inductor, a thing which doesn't have to be coiled at all.

[Pete D]

Different names come up for the same thing; a relay or solenoid is merely a remotely controlled switch.

I suspect the term relay came from the telegraph relays in the 1800's, where the weak incoming current from the long line was used to trip a relay connected to fresh power to send out a new strong signal.

Digital transmission lines use exactly that technique today, albeit much faster and using both electricity and light!

[Mark W. Ingalls]

Different names come up for the same thing; a relay or solenoid is merely a remotely controlled switch.

I suspect the term relay came from the telegraph relays in the 1800's, where the weak incoming current from the long line was used to trip a relay connected to fresh power to send out a new strong signal.

Digital transmission lines use exactly that technique today, albeit much faster and using both electricity and light!

A good lead-in, Pete!

Actually, electrical and optical digital signals use solid state amplification today, but at one time the only way to boost a weak (telegraph) signal was to run it through the coil of a solenoid that switched power from a fresh battery or generator.

Wireless signals were first generated by wheels containing banks of switches, called a spark-gap transmitter

[Mark W. Ingalls]

The other kind of relay, the one that doesn't involve contacts made of metal controlled by an electromagnet, is the solid state relay.

Here's a data sheet for a 100A, 200V DC solid sate relay: Teledyne

We observe that the "on state" resistance is 22 milli-Ohm. (A milli-Ohm, is 1/1000 of an Ohm, so 22 milli-Ohm = 0.022 Ohm.)

Assume we want to pass 30A through this piece of equipment, then we must accept a 0.66 Volt drop across the relay, something we may or may not want to do.

Here's a data sheet for a 70A, 12 or 24 Volt automotive relay: Tyco

We observe that the "Initial Voltage Drop" is less than 0.20 Volt. The contacts are specially plated to be self-healing after current begins to flow through them, so after a short period of time the voltage drop is negligible.

We also note that the Tyco product is rated for 100,000 operations at 70 Amps, while the Teledyne product is not rated for a maximum number of operations.

That is because the Teledyne product, which is solid state, should survive an unlimited number of operations, so the spec doens't apply.

[Pete D]

T1 digital carrier, the basic building block of the US digital network (other parts of the world use different T-schemes), doesn't amplify digital signals, rather it regenerates them new using the old incoming bit as a prompt to launch a brand new clean bit. Even the fiber optic cables have 'noise' and need regeneration of bits at certain intervals.

That's probably the main superiority digital transmission systems have over analog ones, because the analog carrier systems merely amplified both the signal and the noise until the noise overwhelmed the signal, whereas the digital ones keep leaving the noise behind.

Of course, the current process uses solid state micro-components instead of the old WU mechanical repeaters, but the digital regeneration is essentially the same process at significantly faster rates.

[Mark W. Ingalls]

{Note to the reader: I apologize for not trying harder to keep this thread on-topic. I'll try to do better in finishing up.}

Thanks for that reminder, Pete. Since I am in the amplify-weak-signal business, I sometimes forget that every digital copy is essentially indistinguishable from the original, and so of course the repeater is a better alternative to amplification when you have the choice.

In the case of the received satellite signal, where we don't have that choice, we still have to use amplification followed with correlation to deal with the bit errors, eh?

Anyhow, back to power switching...

As was noted before, the Schottky diode has less forward voltage drop, and relatively recent technology advancement has resulted in vastly improved reverse leakage current. This, and other advancements, has allowed manufacturers to offer all kinds of power management products to isolate and maintain our truck's batteries.

Because it was mentioned previously in another thread (belated thanks for that post, 5 Speed) we will review the specifications provided by Hellroaring Technologies for their battery isolator|combiners. In particular we note the following power dissipation comparison between a mechanical relay ("the solenoid") and the isolator|combiners:

Power dissipated with 120 Amps from the alternator divided with 60 amps in main and 60 amps in auxiliary system-- Manual relay: <10 Watts, BIC-75150A: About 36 Watts, BIC-75300A: About 36 Watts
or About 7 Watts (depends on which configuration you use)

So we can conclude that recent advances in solid state battery isolators make them just as efficient, power-consumption-wise, as the tried-and-true relay. We can also conclude that for simple systems, the tried-and-true relay is still a good choice.

The prices quoted for the solid state products are all in excess of $100, while the relay I use came supplied with my truck's towing package. If you are trying to manage a complex system, the isolator|combiner is the way to go. If you just want to charge a "house battery" off your truck's alternator, then the solenoid|relay will probably do the job for less money.