as above. I bought a ready-made PCB with all the parts installed and noticed there's no diode across the relay. should I put one? if yes, would a normal 4007 work?
Thank you.
EDIT: there are 2 relays (1 each for both channels)and using a meter, it seems 2 ends of the relay coil are connected to each other. (not sure if this matters but I thought it might be important)
Thank you.
EDIT: there are 2 relays (1 each for both channels)and using a meter, it seems 2 ends of the relay coil are connected to each other. (not sure if this matters but I thought it might be important)
Last edited:
A well-known and beneficial thing is to jump the relay coil as shown elsewhere with a diode.
The cathode always pointing towards the plus-voltage side of the coil.
It helps to protect the transistor driving the relay from being bombarded with a harmful back-voltage pulse when the relay de-energizes.
Depending on the circuit, anything from a 1N4148 to a 1N4oox will do fine.
The cathode always pointing towards the plus-voltage side of the coil.
It helps to protect the transistor driving the relay from being bombarded with a harmful back-voltage pulse when the relay de-energizes.
Depending on the circuit, anything from a 1N4148 to a 1N4oox will do fine.
Yeah. That flyback voltage can be substantial. The diode is needed to prevent zapping the driver transistor (assuming a discrete transistor). If you're using a relay driver such as the ULN2003 you'll find that it has built-in protection diodes.
The diode needs to be able to conduct the same current as the relay coil draws. A small-signal relay might be OK with a 1N4148, but some power relays and relays with low coil voltage will need a beefier diode, such as the 1N4007.
I often just use 1N4007 and call it good.
Tom
The diode needs to be able to conduct the same current as the relay coil draws. A small-signal relay might be OK with a 1N4148, but some power relays and relays with low coil voltage will need a beefier diode, such as the 1N4007.
I often just use 1N4007 and call it good.
Tom
Here's a little test you can do to help you understand the problem. Put a 12V relay on the bench, without a diode connected, and energize it with a lab supply or wall wart or a battery. Now put your fingers across the relay coil contacts, and while still touching them, switch off or disconnect the power supply. You won't notice the 12V with your fingers, but you will feel the pulse of back-emf.
You evil... Or use an oscilloscope. 🙂
I think I was about 9-10 years old when I discovered that transformers can give you quite a zap if you happen to be holding the primary wires while measuring the secondary connections with an ohmmeter.
Tom
I think I was about 9-10 years old when I discovered that transformers can give you quite a zap if you happen to be holding the primary wires while measuring the secondary connections with an ohmmeter.
Tom
It depends on the circuit. If the relay is driven from an emitter follower, the current just keeps flowing through the emitter follower until the magnetic field is gone when the relay is switched off.
Didn't mean to be evil 😉 - I just found that out by accident myself and think it is a pretty enlightening thing to experience hands-on. You won't electrocute yourself after all, and afterwards you won't dismiss any catch diode lightheartedly anymore.
Oh, I was writing that with a smile on my face. I know you won't electrocute yourself if you feel the kick-back voltage, but it can be a rather unpleasant surprise. It's the same operating principle as an electric fence just on a smaller scale.
Tom
Tom
Reminds me of one of my early electrical experiences, maybe 7-8 years old, when I independently discovered inductive kick-back. In my youthful ignorance, I tried connecting a D-cell battery across a shaded-pole phonograph motor. Of course, nothing happened to the motor...until I disconnected the battery while still hanging onto the motor leads. I saw a few stars! It wasn't until years later that I learned what actually happened.
As others have said you need a flyback diode to protect the driver transistor.
I use a diode and a 12 volt Zener to reduce discharge current and make relay turn off faster.
Slow turn off can cause arcing.
I use a diode and a 12 volt Zener to reduce discharge current and make relay turn off faster.
Slow turn off can cause arcing.
Very similar experience and age except I used an old TV vertical output auto-transformer.
Pleased to read the suggestion to use both a diode and 12 volt zener. This is wise advice.
A diode alone slows down the relay switching. It extreme cases (high DC current switching applications) it can lead to welded contacts.
Relays are designed to mitigate this - because they know everyone uses simple free-wheel diodes. This is the reason the contacts are mounted on spring strips, the springs pull the contacts apart only after the armature has accerated away from the yoke and opened the magnetic circuit. The magnetic force drops significantly by this point due to the air gap, so there is significant velocity of the armature even if the current is reduced extremely slowly.
Nonetheless, I've seen a relay application note where reducing the current quickly was highly recommended. In fact, using just a freewheeling diode was referred to as "misapplication".
Yes, it will separate even faster with rapid magnetic field quenching, but standard relays are designed with free-wheel diodes in mind. Solenoids and solenoid valves are a different story, note.
Only yesterday I measured insulation resistance of a 10 kV power distribution transformer. It had first seen a L1-PE shortcircuit followed by a full 3 phase shortcircuit at the secondary side (used as primary…). Insulation resistance meters put higher voltages on the windings with the same results so caution and use of PPMs. The transformer does not have a day off when you are measuring 🙂
Inductive and capacitive parts… can present surprises.
Last edited by a moderator: