Relays with diodes

[boris_thespider]

On what circuits should I use a relay with a diode? And, is there ever a bad time to use this type of relay? Thanks...

Mike

[Gunner]

Your question is too vague to answer except in equally vague terms.

A relay should be used any time you want maximum current to a lamp, lamp string, heater motor, etc. without passing it through long wiring and switch.

Diodes are rarely needed except in those cases where you need to have multiple sources drive the same lamp (e.g., brakes, turn signal, emergency flashers). Diodes are kind of a kluge and can be avoided with correct wiring in most cases. They're used to patch in a new circuit without rewiring other parts of the system, such as when you add emergency flashers to an already-wired car.

If you have a more specific question I can try to give a more specific answer.

[kobrabytes]

http://www.kpsec.freeuk.com/components/relay.htm

[ccomt]

Here's a brief primer on relays that discusses the reason for diode protection:

http://www.rallylights.com/hella/Relays.aspx

Here's a link to specs and info on a Hella diode protected relay used in automotive apps:

http://www.rallylights.com/detail.aspx?ID=412

[IndyCobra]

Actually, voltage spikes (the perceived need for the diode) are mostly a figment of the imagination. The mostly likely scenario to produce a spike is when the alternator is removed and then added back into a charging circuit when the engine is running. In the aviation community, there's typically a switch that allows the pilot to remove the alternator from the circuit in order to run the electrics from the battery only. This is done for safety and redundancy reasons. Cycling this switch "may" cause a spike, but even that has been elusive to prove scientifically. In automobiles, the alternator is wired to always be in the circuit so there's no way to turn it off and then turn it back on again.

Bottom line is whether you believe in voltage spikes or not, I wouldn't worry much about diode protection in your relay. It won't hurt anything to have it, but it likely isn't doing anything to begin with.

[vettestr]

Boris,
Seems we have some who do not see need for diode but I don't agree with statement spikes are "are mostly a figment of the imagination" but like Indy mentioned they don't hurt anything. Let me toss in my 2 cents for what it's worth and do what works for you.

When you have a load on the relay that generates a flux field like on a motor running your cooling fans it collapses to throw feedback at the relay contact/s. A load like driving lights does not have mass that is still turning like the armature in a motor, alternator or ? so does not have the same spike potential but contacts do arc in the relay when cycled on/off, all these things lead to eventual failure. All relays with or without diodes or load types will fail sooner or later so just a matter of cost and benefit with things like reliability and amount of cycles expected as points to consider. There was an article in the links of the cooling fan thread that I thought explained this well, it is on my other computer but will try and dig it out.

Driving lights are higher amperage draws but the number of times it will turn on/off in a year of driving plus no circuit feedback ..... I would not drive to parts house to get a diode protected relay if I had proper rated relay in my hand. If installing a high load cooling fan like the Lincoln fans drawing 40 plus amps plus considering how many times it will turn on and off through the t-stat in a year of driving I would only install a diode or a relay with a diode built in. The non protected relay would work for sure but I want to include as much reliability on this device as possible.

[Gunner]

Originally Posted by IndyCobra

Actually, voltage spikes (the perceived need for the diode) are mostly a figment of the imagination.

It might seem so in auto/aero electric circuits. They're certainly real when driving relays with electronic components - I've seen many a transistor or IC power output damaged by driving a relay without a protection diode. I suspect they're just as real in vehicles but there are no components sensitive enough to be affected by them. It becomes a "might as well" sort of protection under those conditions, probably never serving any purpose. As they don't cost anything (maybe a buck, for a new, high-power diode) and can't hurt... "might as well."

[IndyCobra]

I worked closely with two experts in the aviation electrical field a while back. The issue of voltage spikes is a big concern in the aviation community because you don't want a spike to destroy $50,000 worth of sensitive avionics if a simple diode can prevent it.

They tried repeatedly to create one in numerous aircraft under strict scientific testing and could not. That doesn't mean they don't occur, but they were sufficiently satisfied that current generations of personal and commercial aircraft are sans diode protection.

I admit that this doesn't necessarily transfer to automobiles where the comformance to strick electronic standards for component parts is probably less rigorous. So install a diode and drive happy.

[Gunner]

I suspect you're right, that avionics and aeromotor electricals are already heavily protected and designed to produce no unwanted noise and interference. They thus don't need the extra protection of a coil diode.

I've seen and measured (and repaired the damage from) e-coil spikes in both cars and circuit boards. Many times, in the latter.

[PDUB]

Originally Posted by IndyCobra

Actually, voltage spikes (the perceived need for the diode) are mostly a figment of the imagination.

If this in fact were true, then your ignition system would not create a sprark! Engine no runny...

Quick primer:

Anytime the magnetic field surrounding a coil collapses, i.e. when you shut it off, you get an electric field, and thus a voltage, induced that is associated with said collapse. This is associated with inductance. The induced voltage attempts to try and keep the same amount of current flowing in the circuit as just before the switch was opened to turn off the relay. These induced voltage spikes can reach extremely high values, depending on conditions. This is basically how an ignition system generates a spark. And, it is this induced voltage that can damage componants.

The diode, often called a fly-back diode, is mostly important when you are using an IC to turn the relay on and off, as in the diagram for which Kobrabytes provided the link. In this case the transistor is in an open collector configuration as a low-side driver, and operates as the switch control for the relay coil circuit. Keep in mind that we are talking about the relay coil and not the relay load being switched, which could also have a separate inductive spike issue. The diode provides a discharge path for the inductive spike when the relay coil circuit opens, thus protecting the transitior switch. Again, this has nothing to do with the load switching.

If you want to see just how big inductive spikes can get, look here:

http://www.youtube.com/watch?v=0pRZ-...eature=related

or here:

http://www.youtube.com/watch?v=pqMYYlYq4Ho

and don't forget to duck!

[trularin]

Generally, we use a diode on relays that are driven.

An open collector driver where the relay acts as Rc, should have a diode placed across the coils or after a period of time, a nice hole will get blown through the transistor.

You do not need a diode for switch driven relays.

Just my $0.02 worth.