Why have a Resistor across a Relay Coil?

I have a IO-360 with Aerosance FADEC. The Aerosance wiring shows a 2200 ohm 1/2 W resistor across the Boost Pump Relay (That coil measures 78 ohms). 14V is supplied through a Boost Pump Sw. and the low side ground is supplied by the ECUs to turn on the electric fuel pump as needed. This has worked great for 9 years but on a couple of recent flights it logged a BOOST PUMP RELAY SHORTED error that lasts about 20 minutes then goes away until the next cold start flight. So as I try to figure out what is happening I wanted to understand why the resistor is in the circuit.


It's been about sixty years since electronics school but I knew some of you will know why that resistor is in the circuit.

I would go back to the Aerosance and ask them for clarification.

Normally any relay should have a protection diode across it. Not a resistor. This prevents kickback voltage spikes caused by the collapsing field damaging control circuitry.

Are you sure its a resistor or just a failed diode?

It's probably about error detection and reporting

The ECU "output" that controls the boost pump relay would be protected, in that it has a "current limiting" capability. If the current the ECU is sinking is beyond a certain threshold, it will indicate that the boost pump is on the way out. Although it might report a "short", it might be a partial short of the windings, the boost pump might be still "working" but it is almost certainly headed for a hard failure and should be checked out.

Given that the ECU can sense the current it is sinking for the boost pump, consider now a boost pump failure where the motor windings are "open circuit". The pump has failed, but how does the ECU know the difference between a pump failure and an electrical supply failure (CB tripped) for the boost pump? That 2200 ohm resistor is the answer. If the boost pump has gone open circuit, but the supply is OK, then the ECU will sink a bit over 6 mA of current. It'll recognize this and deduce that the supply (circuit breaker + switch) is OK but the boost pump itself has failed open. That's the error it will report.

If the resistor fails, no big deal, but the ECU won't be able to tell the difference between a boost pump open circuit failure and a supply/switch failure, so it'll probably report a boost pump supply voltage error even when the supply is OK (but the pump failed).

A.

Resistors and diodes are both commonly used across the coil in a relay to suppress voltage transients that happen when the coil is turned off with a switch. The magnetic field through the coil doesn't allow instant current changes so the voltage across it increases to the point that a conductive arc forms and allows the current to jump the most convenient gap back to the other side of the coil until the field totally collapses. Semiconductors could be destroyed and switch contacts damaged in the process.

A diode across the coil provides a very low resistance, controlled path for the coil's current to travel instead of arcing. Since the resistance is low, it will take a while for the current to decrease and the relay will open slower than without the diode. This can cause some arcing inside the relay's switch and even cause the relay to weld into an undesirable position. The diode is saving external circuitry at the expense of possibly reducing the relay's own lifespan.

A resistor across the coil also provides a controlled path for the current when the switch opens. The resistor will convert the energy of the magnetic field to heat and allow the relay to switch faster than if it had a diode. Depending on the resistance, it will be almost as fast as if there was no protection. The downside is the resistor will always be wasting some power as long as it is on and there will still be some voltage transient. Depending on the resistor used, it will probably be 100+ volts, but not the 1,500+ of an unsuppressed relay.

gammaxy said:

... The downside is the resistor will always be wasting some power as long as it is on and there will still be some voltage transient. Depending on the resistor used, it will probably be 100+ volts, but not the 1,500+ of an unsuppressed relay.

Click to expand...

Actually wouldnt the peak voltage across the resistor be the normal energize coil current times the resistor value? The lower the resistor the faster the relay is de energized at the expensive of higher quisence current.

PilotjohnS said:

Actually wouldnt the peak voltage across the resistor be the normal energize coil current times the resistor value?

Click to expand...

I left this out of my explanation, but you're correct.

PilotjohnS said:

The lower the resistor the faster the relay is de energized at the expensive of higher quisence current.

Click to expand...

This seems reasonable, but where does the energy go? It's just a loop of a resistor and inductor. The current everywhere in the loop is the same, so energy is not being sent back to the battery or somewhere else. Without any resistance (either the resistor, wires, or the coil's resistance), the current will keep flowing and the relay won't open. A resistor resists current flow and emits the energy as heat according to the equation P=I^2*R. Notice that more resistance means more power is dissipated.

Thanks for the replies. I better understand the use of a resistor across the relay coil.

I can't contact Aerosance since Continental bought them and promptly dumped support for the FADEC. They try to get Bonanzas with FADEC to remove them if they have problems. It's a shame because it is a great system.

One thing I did not make clear; the boost pump relay gets the 14v from the Boost Pump Switch Auto Position and the ECUs provide the low or ground side to pick the relay as needed. The actual 14V to the Boost Pump is provided through the N/O points of the Boost Pump Relay so the ECUs would not know the condition of the Boost Pump, just the Boost Pump Relay Coil low side indications.

I prefer to use a bidirectional transient suppression diode or "transorb" for relay coils. Unlike a resistor, it draws essentially no current while the relay is energized. A diode is ok but the low voltage (about 0.6 V) while suppressing the turn-off spike means that the current decays relatively slowly. This causes the contacts to slowly separate which increases arcing. A transorb will break down at a higher voltage depending on the which part you select. The higher voltage will cause the relay to open quicker but this voltage is limited so that arcing in the switch driving the coil is avoided. These parts install just like a diode but are even easier since you don't have to worry about polarity.

Here's a part I bought just a few weeks ago:
https://www.arrow.com/en/products/1.5ke22cahe3-ac/vishay

This transorb breaks down at a minimum of 18.8 V. I've been going to Arrow electronics because they have a huge selection of electronic components and give free shipping for any size order.

Bob Nuckolls tested relay contact opening speed in his lab. He found that there is no difference in contact opening speed regardless of the type of arc suppressor used. It is true that a diode may delay contact opening. But once the contacts start to open, they open just as quickly no matter what the arc suppressor. The purpose of the diode is not to protect relay contacts, but rather to protect the contacts of the controlling switch from arcing. The voltage across the controlling switch contacts, when first opening, is system voltage plus diode forward voltage drop.

Thanks for the reply Kevin.

I'm reluctant to change the circuit components since this is the components recommended by Aerosance and their ECUs provide the low or ground side of the relay coil. I just did not understand why the resistor was there in the first place. There are other relays in other circuits that don't have a resistor or diode across the coils.