I am using a DYAD CP Clare DSS4 1A05, normally open, relay. The datasheet can be downloaded

I am new to using relays. These seem to be the important specifications:

Coil Voltage: 5V Typical -- 10V Maximum
Operate Voltage & Release voltage: 0.8V Minimum -- 3.75V Maximum

Here is my circuit:

Relay Circuit

- R0, R1, R2, and R3 are relays of concern
- I0, I1, I2, and I4 are inputs which can be 5V to close the switch or 0V to open the switch.
- The left and right motors are 12 volts dc motors.

The problem is, the relays got destroyed too often, usually after a day of use. That is, the two armature terminals of 1 or 2 relays get normally open (and cannot be closed anymore), and some get normally closed (and cannot be opened anymore). I don't know what I lack here.

I tried reducing the +12 and -12 volts to +- 6V, and the dc motors replaced with 3 volts, but still the same problem.

What do you know about this matter?

  • \$\begingroup\$ In what way were they "destroyed"? Coil burned out? Contacts welded or fused together? What? \$\endgroup\$
    – Majenko
    Jan 12 '12 at 16:20
  • \$\begingroup\$ Also, what is the current draw of the motors? \$\endgroup\$
    – Majenko
    Jan 12 '12 at 16:21
  • 4
    \$\begingroup\$ Your link doesn't go to the datasheet. It seems to go to some datasheet search engine. Do it right and fix your question to point to the actual datasheet. \$\endgroup\$ Jan 12 '12 at 16:23
  • \$\begingroup\$ @Majenko. Edited. Some relays gets their contacts welded and some are fused together. Current draw of the motor is the typical one, range of 0.3-1.0 A. \$\endgroup\$ Jan 12 '12 at 16:33
  • 3
    \$\begingroup\$ @Kortuk: While anything is possible, it's very rare for a relay coil to break. It's basically impossible to hurt the coil of a 12V relay by driving it on/off with 12V as often as you want. Intead of taking the relay apart, listen for a click. Even if the contacts have welded, there should still be something audible as the coil is switched on and off. When relays break, it's just about always the contacts or part of the mechanical mechanism. Coils of wire don't break unless you so something very unusual or way out of spec. \$\endgroup\$ Jan 13 '12 at 14:17

There are two obvious possibilities:

  1. You aren't doing break before make properly. You need a delay between turning off one relay and turning on the next. It's quite possible that one relay can turn on faster than the other turns off, leaving a dead short thru both relays for a little while.

  2. You are switching a inductive load but not giving any place for the inductor current to go when switching off. Relays are intended to handle this to some extent, but at the least this will cause arcing when the contacts open. Two reverse Schottky diodes to the two supply rails for each motor will fix this problem.


Here is a schematic showing how the reverse diodes should be connected:

  • \$\begingroup\$ For #1. I intentionally tried opening all 4 relays together at one time and close them together. I did it many times. The relays were not destroyed at those instants. During normal operation (as you suggested by adding delay), at some unknown instance, the relays got destroyed. For #2. Can you give example circuit or a link for that? \$\endgroup\$ Jan 12 '12 at 16:36
  • \$\begingroup\$ What's with the downvote? What exactly do you think is wrong with what I said? \$\endgroup\$ Jan 12 '12 at 18:07
  • \$\begingroup\$ @OlinLathrop, I am not sure why, but at least 4 others disagreed, a win of 38 rep. \$\endgroup\$
    – Kortuk
    Jan 12 '12 at 19:32
  • 1
    \$\begingroup\$ @Kortuk: That's not the real point. This stack exchange system is broken as long as someone can anonymously downvote without explanation. We see random unexplained downvotes occasionally, and there is no upside to them. I notice a correlation between having a disgreement with someone in one place and downvotes suddenly appearing elsewhere. This system is being abused. It's time the stack exchange folks took this seriously and put a stop to it. \$\endgroup\$ Jan 12 '12 at 23:14
  • \$\begingroup\$ @OlinLathrop: I used to be in StackOverflow and we felt the same way when random downvotes come in. But don't worry, there is nothing wrong with the system and there is no abuse. A one or two downvotes plus three or more upvotes proves that your answer is correct. \$\endgroup\$ Jan 13 '12 at 2:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.