2 diodes, 1 driver? Power help.

[rhd]

24V using an LM317? It was able to dissipate that much extra voltage?

 

[Leodahsan]

I can't see why not. With no reason to use it like that, I tested a little time so, it is not recommended, but works

 

[rhd]

I thought the limit was 15V or 16V (or maybe even 12V) ?

Perhaps it's a limit that reacts to heatsinking.

 

[anselm]

The LM317 can take up to 40V input.
http://www.national.com/mpf/LM/LM317.html#Parametrics
To prevent it's thermal protection from triggering, you just have to keep it cool enough.

 

[rhd]

I had no idea - that's phenomenal.

 

[BShanahan14rulz]

put 'em in series and I personally promise that they will see the same current.

 

[Bionic-Badger]

Leodahsan, your current analysis is incorrect. You're thinking of diodes as resistive elements, which they are not. Furthermore, even if the devices were resistors you're confusing current with voltage. As per Kirchoff's Current Law, the sum of all currents entering and exiting a node is zero. This implies that the current flowing through all devices in a series-topology circuit is exactly the same for each device; the same amount enters as exits. With resistive elements the resistance will determine what voltage drop each resistor causes in the voltage divider; this does not hold in any shape or form for current in series elements.

Diodes drop their forward voltage and that's all you should worry about. Put them in series, and set the current with your driver. The forward voltage drop of the diodes will add, and the driver must provide enough voltage to cause current to flow through the diodes. Manufacturing differences will affect your circuit if you wire the diodes in parallel because their voltage drops, etc. will be slightly different. So don't wire diodes in parallel unless you have separate regulating elements, which can include resistors or transistors or other elements.

 

[Cyparagon]

Yup, mister leo is wrong once again. The different current is only a problem when running them in parallel. You can get rid of that problem with a current-balancing resistor on each diode, though. It has been done many many times. Typical values are 1-2 ohms.

Go with series if you can, though.

 

[Leodahsan]

@BB
@Cyp

I noticed my error and edited that
I'm taking my time to study that. You'll feel happy to not see me in LPF for a 'long' time. :yh:
I'll miss you boys