Folks:
I'd appreciate a quick education on the value of using diodes in source selection circuits. I've built simple relay-based source selection boards for several preamp and integrated amp projects. My boards are simple enough -- a rotary switch on the front panel toggles power between the relays to activate one input and deactivate the others. But I've seen photos posted by other members that include a diode at each relay, and I'm unclear why. Is the diode in series with the relay and intended to block reverse voltage? Is it in parallel with the relay (and if so, for what purpose)?
Thank you in advance for the education.
Regards,
Scott
I'd appreciate a quick education on the value of using diodes in source selection circuits. I've built simple relay-based source selection boards for several preamp and integrated amp projects. My boards are simple enough -- a rotary switch on the front panel toggles power between the relays to activate one input and deactivate the others. But I've seen photos posted by other members that include a diode at each relay, and I'm unclear why. Is the diode in series with the relay and intended to block reverse voltage? Is it in parallel with the relay (and if so, for what purpose)?
Thank you in advance for the education.
Regards,
Scott
The back emf can be very high when the impedance is very high. Approaching a kVpk from some relays.
Shorting the relay coil at the time of switch off changes that to a very high current.
This shorting slows down the release of the contacts.
Instead of using a short to avoid the very high voltage, one can use the Vf of a power diode to reduce the voltage and pass lots of current. This still slows down release. ESP has given some data in a paper.
But instead use a diode + Zener. This develops a back emf voltage roughly equal to the supply voltage saving the switching transistor from damage and reduces the current significantly. So much so that a 1n4148 signal diode works reliably. And it allows a slightly faster release.
For a 12V relay I use a 10V to 12V Zener + 1n4148 or 1n914
Shorting the relay coil at the time of switch off changes that to a very high current.
This shorting slows down the release of the contacts.
Instead of using a short to avoid the very high voltage, one can use the Vf of a power diode to reduce the voltage and pass lots of current. This still slows down release. ESP has given some data in a paper.
But instead use a diode + Zener. This develops a back emf voltage roughly equal to the supply voltage saving the switching transistor from damage and reduces the current significantly. So much so that a 1n4148 signal diode works reliably. And it allows a slightly faster release.
For a 12V relay I use a 10V to 12V Zener + 1n4148 or 1n914
AndrewT & DF96:
Thank you. The ESP article was helpful. It seems my simple rotary switch-controlled circuit (which lacks a transistor) might benefit from a snubber at each relay instead of a diode/zener combination. Rod Elliott suggests the "rule of thumb" for snubbers is a resistor of 0.5 to 1.0R per contact volt and a capacitor of 500nF to 1uF per contact amp. I will be using 24 VDC to operate the relays (Fujitsu RY-24WZ-K). I have plenty of .25VA 15R resistors on hand. I understand the resistors will be fine for this application but am unclear about calculating the "contact amp". The coil resistance on these relays is 2880 ohms. 24/2880 = .00833. I have a few .01uF monolithic ceramic caps on hand. Are they suitable?
Regards,
Scott
Thank you. The ESP article was helpful. It seems my simple rotary switch-controlled circuit (which lacks a transistor) might benefit from a snubber at each relay instead of a diode/zener combination. Rod Elliott suggests the "rule of thumb" for snubbers is a resistor of 0.5 to 1.0R per contact volt and a capacitor of 500nF to 1uF per contact amp. I will be using 24 VDC to operate the relays (Fujitsu RY-24WZ-K). I have plenty of .25VA 15R resistors on hand. I understand the resistors will be fine for this application but am unclear about calculating the "contact amp". The coil resistance on these relays is 2880 ohms. 24/2880 = .00833. I have a few .01uF monolithic ceramic caps on hand. Are they suitable?
Regards,
Scott
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.