Working on an existing product which has had field failures. Design is a short (60ms) pulse is sent via microcontroller to activate / deactivate a latching switch. I have determined failure is due to the fact that (coupled with external variables) there is no repulse upon sending of short pulse. Therefore if latching switch happens to not activate upon short pulse, product is dead in the water until power is cycled.

I have a design workaround via software to generate a pulse / repulse strategy. Basically stating I want to repulse every so often, just in case latch switch did not respond (I have no other way of sensing other the external variables as design is existing).

I have cleared this with the latch manufacture to state it is not detrimental to repulse the switches. I am now questioning what the down side of this pulse / repulse might be on the 4700uF capacitor (see schematic).

Bench testing shows 4700uf cap once fully charged is 28VDC and upon pulse, drops to 14VDC, then recharges to ~28VDC in ~15 seconds.

Cap using is NRLM472M35V25X25F which is 35V rating which I am also questioning. I would have used a 50V rating but there is some timing involved so possibly they had to use the 35V. I'm looking into that right now.

My investigation of the cap spec sheet has lead me to estimate 3.36 years for this cap if @ 35V and 113°F(installed outdoors) environment (doubling life expectancy every 10°C reduction from 85°C). I am looking to better understand the negative implications (if any) of the additional pulses are to the capacitor.

Anyone have any input to offer??

enter image description here

  • 1
    \$\begingroup\$ Not sure what you mean by repulse. If the back story isn’t relevant, ditch it. \$\endgroup\$ – Andy aka Dec 7 '18 at 19:49
  • \$\begingroup\$ If it fails the 1st time, how likely is it to succeed the next? There is an empirical solution to this problem if you know the L, DCR, V, t guaranteed to switch the Relay with a drive of C, V and ESR and have a safety factor for aging. \$\endgroup\$ – Sunnyskyguy EE75 Dec 7 '18 at 19:52
  • \$\begingroup\$ Andy, once pulse is sent. Repulse again and again. Note taken on back story. \$\endgroup\$ – becjasl Dec 7 '18 at 20:01
  • 1
    \$\begingroup\$ But this part has 1/2 the Tan delta so will last much longer and give more reliable results. But again do the math. digikey.ca/product-detail/en/epcos-tdk/B41888C7478M000/… \$\endgroup\$ – Sunnyskyguy EE75 Dec 7 '18 at 20:01
  • 1
    \$\begingroup\$ A flyback diode would be an important component when using a coil, but I don't see it on the circuit diagram. Is there one? \$\endgroup\$ – Andrew Morton Dec 7 '18 at 20:06

Your Answer

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

Browse other questions tagged or ask your own question.