diode across solenoid coil

[fuseshut]

Conventionally I have always used a 1n4004 diode across a relay coil to prevent any back EMI. I have this solenoid application and was going to put diode across its coil. I have read previously that a fast recovery diode is recommended, rather thant the standard 1n4001-4004 series. Any thoughts on this?

 

[OperaHouse]

A fast recovery diode only helps when the voltage immediately reverses on the diode. Since this only happens after you have removed power from the solenoid, it doesn't matter if the diode becomes a dead short for a short period of time. I assume no one turns off a solenoid for 10ns and turns it back on. That advice while well intentioned is bogus.

 

[itsmoked]

1N4004 is just fine.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Comcokid]

Fourty years ago standard rectifier diodes were slow so faster diodes were recommended for inductive kickback supression. Today, it's no issue. Use the 1N4004.

 

[Skogsgurra]

Fourty years ago, we used selenium rectifiers for that. And 1N4004. And sometimes the 914 for real low power coils. They were slow - true. But reverse recovery speed has nothing to do with how well a diode takes care of inductive kickbacks.

Let's not feed and care the myths.


Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[bogeyman]

Fitting a commutation diode across the soleniod will slow the release time of the soleniod when switched off.
The magnetic field will decay much more slowly than without the diode, the decay of flux will be aproximatly exponential with a time constant of L/R. Note that the vale of L will change as the magnetic circuit varies due to armature position.
I once tried to improve contact life in the fuel pumps of a Jaguar E type with diodes but the delay made the solenoid operated pumps cycle time to long to give good fuel delivery at large (heavy footed) throttle openings. The final cure was an arc suppressor of 10 ohms and 0.01uF in series over the points, just like SU used to fit. An alternative cure would be a voltage dependent resistor or possibly a zenner over the switching device.

 

[OperaHouse]

People forget that just a resistor across the coil is more than sufficient and more reliable in many applications. Many automotive relays now include a resistor, less probability of failure and they can't be put in backwards. I used to sell diode snubbers with internal resistors.

 

[btrueblood]

Opera,

The problem with a resistor is the power draw goes up whenever the solenoid is energized. In this day & age of energy conservation, that's not necessarily a good thing.