Post by thread:[EE] Relay coil diode specifications

I have never had any problem with just 'grabbing a diode' and using it. I may be wrong but for the amount of time its going to be carrying current it may not matter much. Do correct me if I am wrong and have just been lucky previously. Steve -----Original Message----- From: spam_OUTpiclist-bouncesTakeThisOuTmit.edu [.....piclist-bouncesKILLspam@spam@mit.edu] On Behalf Of Goran Hosinsky Sent: 31 May 2006 08:24 To: Microcontroller discussion list - Public. Subject: [EE] Relay coil diode specifications What voltage and current specifications should the diode across the relay coil have for a relay pulling 400 mA at 15V dc? Goran, La Palma

> What voltage and current specifications should the diode > across the relay coil have for a relay pulling 400 mA at > 15V dc? A coil tries to maintain its current, so when the driver switches off the current will be 400mA. When the driver is switched on the diode will see a reverse voltage of 15V. So a 15V/400mA diode would do. But I would pick an 1N4004, becase that is my 'mid-current' yellybean diode. Wouter van Ooijen -- ------------------------------------------- Van Ooijen Technische Informatica: http://www.voti.nl consultancy, development, PICmicro products docent Hogeschool van Utrecht: http://www.voti.nl/hvu

Wouter van Ooijen wrote: > So a 15V/400mA diode would do. But I would pick an 1N4004, becase that > is my 'mid-current' yellybean diode. As long as the relay is not going to get turned on again before the current from the last on phase decays and the diode turns off. The reverse recovery time of 1N4004 is rather slow, so otherwise you get some nasty effects if you turn the switch on while the diode is still conducting. ****************************************************************** Embed Inc, Littleton Massachusetts, (978) 742-9014. #1 PIC consultant in 2004 program year. http://www.embedinc.com/products

> > So a 15V/400mA diode would do. But I would pick an 1N4004, > becase that > > is my 'mid-current' yellybean diode. > > As long as the relay is not going to get turned on again > before the current > from the last on phase decays and the diode turns off. The > reverse recovery > time of 1N4004 is rather slow, so otherwise you get some > nasty effects if > you turn the switch on while the diode is still conducting. What would be a sure-reliable suggestion? something like an 1N5819? Wouter van Ooijen -- ------------------------------------------- Van Ooijen Technische Informatica: http://www.voti.nl consultancy, development, PICmicro products docent Hogeschool van Utrecht: http://www.voti.nl/hvu

Wouter van Ooijen wrote: > What would be a sure-reliable suggestion? something like an 1N5819? A schottky would better for spped if the voltage is low enough and the leakage acceptable (which I think is the case here). However I would probably make sure in the firmware that there is always a minimum off time for each relay to guarantee the diode is off before the relay is switched on again. ****************************************************************** Embed Inc, Littleton Massachusetts, (978) 742-9014. #1 PIC consultant in 2004 program year. http://www.embedinc.com/products

On Wed, 31 May 2006, Wouter van Ooijen wrote: >>> So a 15V/400mA diode would do. But I would pick an 1N4004, >> becase that >>> is my 'mid-current' yellybean diode. >> >> As long as the relay is not going to get turned on again >> before the current >> from the last on phase decays and the diode turns off. The >> reverse recovery >> time of 1N4004 is rather slow, so otherwise you get some >> nasty effects if >> you turn the switch on while the diode is still conducting. > > What would be a sure-reliable suggestion? something like an 1N5819? Either that or 1N4004 and series RC snubber in parallel to slow the pulse down a little so the 1N4004 can cope with it. The 1N4004 is not just slow, it has a relatively large Rs which causes the reverse voltage to be much higher than you think it is, at least initially. This will result in an impressive narrow peak that can raise as much as 1/2 Vcc (in the conduction direction of the diode of course). The series RC snubber will deal with that. Peter