protection diode circuit for dual direction LEDs

[unknown_h]

Hi all,
I have a circuit I am using to show the direction a particular motor winding is energized by a controller chip and I have been told that I need a protection diode to prevent reverse voltage damage to the LEDs.
I have tried all the possible diode and zener diode configurations that I could think of, but they all result in something blowing up according to the stress simulations I have performed. I have been told that the diode should set the current direction somehow, but this just confused me more. Any Help would be great, the motor coils are being energized at 12Vs.

[inse]

I don’t know why this should blow up.
Based on the efficiency of your LEDs, 510Ohms might get a blinding indicator, better go for 1k.

[unknown_h]

Thanks for reminding be about that resistor,it's on my list of things to fix. I think they might be worried about reverse transients and the 5v max reverse voltage?

[gabiz_ro]

Connected anti-parallel, reverse voltage for either led is forward voltage of other led.

[Ian.M]

i.e. each LED die in the antiparallel dual LED is the other die's protection diode!

However, as Inse pointed out you are driving the LEDs rather hard.  If the motor voltage spikes over about 17V e.g. due to regenerative breaking or just a crappy controller design that doesn't clamp back-EMF from switching the motor, the peak LED current could exceed 30mA and possibly damage the LEDs (if they are the typical 5mm through-hole parts with max 30mA If).  Decreasing the LED current to 10mA by increasing the series resistor, to allow more headroom for spikes would be advisable, and if the controller doesn't clamp the back EMF, you may need to split the series resistor in two, and add back-to-back Zener clamping.

[djsb]

Why not put a snubber capacitor across the series resistor as well? Not sure if it would work but worth a try?

[Ian.M]

That would make it worse as a capacitor across the LEDs current limiting resistor would dump any transients straight into the LEDs as large current spikes- you have to split the resistor and put the capacitor (or pair of opposite direction Zeners in series) across the series combo of the LEDs + half the resistor.  The other half of the resistor then limits transient currents into the capacitor (or Zeners), so it (they) can function to smooth (limit) the voltage, and thus the LED current.

N.B. a capacitor (unless very large) directly across a LED die is minimally effective for smoothing the LED current due to the low slope resistance of the LED's Vf vs If curve.

[unknown_h]

Thanks all for the help seem falstad sims agree, seems that Micro-Cap smoke sims (which I probably have set up wrong) had me barking up the wrong tree. While the software seems really powerful all the documentation has disappeared now that it's discontinued.

[mariush]

The basic through hole leds have a low reverse voltage value ... often 5v or less. Adding just a 1n400x diode in series with each led would probably solve this issue.
Agree with higher resistor values to limit current to something more reasonable like 5-10mA

[wasedadoc]

The basic through hole leds have a low reverse voltage value ... often 5v or less. Adding just a 1n400x diode in series with each led would probably solve this issue.
Agree with higher resistor values to limit current to something more reasonable like 5-10mA

There is no low reverse voltage issue in this configuration where the LEDs are connected inverse parallel.