2 diodes, 1 driver? Power help.

[bhwollen]

Ok im trying to run 2 diodes off of a 3.5A power supply. If i were to run 2 445 diodes in series would i get 1.75A each diode?

 

[anselm]

No.
When driving two diodes connected in series, you'll need twice the voltage, but
the current needs to stay the same as for one diode alone.
What driver are you going to be using?

The way you are thinking would be the case if you wire them in parallel.
In any case, you cannot connect the diodes directly to the power supply.

 

[Leodahsan]

edit: read more about voltage/current.

 

[anselm]

I've really had it with you, leo!
Please just stop posting if you have only false information to contribute. I understand
you like to help newbies and most of the time your spot on, and I appreciate your enthusiasm.
But sometimes you just need to control your urge to post BS.

if you wire two diodes in series one of they can get a higher current (because of lower resistance) than other,

Series and parallel circuits - Wikipedia, the free encyclopedia

Learn some basic electronics, man!

 

[Leodahsan]

you said "UTFSE" more than 5 times (that I saw), that really pissed me off but I think someone would politely tell you "anselm, don't do it, it is wrong", I didn't neg rep you anyways.
if he "utfse" as you said, he had answered his own question.

 

[anselm]

Dude, drivers aren't mean't to power different diodes in series. I'm not posting bs, I'm trying to say that he is more likely to get damage with two diodes in series than with two diodes in parallel.

Just goes to show your ignorance.
Inside the projector where these diodes originally come from they are, indeed, driven in series.
As well as parallel.

OP:
remember to use balancing resistors on each diode if you do end up setting them up in parallel,
cause if you don't, then yes, one diode will have more current flowing through it than the other.
Again: what's your driver?

 

[Leodahsan]

I'm not posting in this thread or correcting you anymore, mr. neg rep.

 

[bhwollen]

Woah guys, I didnt mean to start an argument. All I wanted to use a FlexmodP3 and split the current between the two. I still don't understand the best way to do this.

 

[anselm]

Don't worry, you didn't start any argument.

split the current

-> parallel

 

[bhwollen]

So i set my driver at the desired current for only one diode, not double it?

parallel

 

[anselm]

No, you'll need twice the current than you'd need for one diode.
What else did you mean when you said "split the current"?:thinking:
Please read that wikipedia link above.

 

[bhwollen]

Ok so what i got from that is if I wire in parallel then I should double the current. So set my driver to 3.5A and set the diodes on their own individual circuits.

No, you'll need twice the current than you'd need for one diode.
What else did you mean when you said "split the current"?:thinking:
Please read that wikipedia link above.

 

[anselm]

.... and use balancing resistors (for example 0.2Ohm) for each diode.

And don't start off with 3.5A.
Start low, try higher and higher currents and observe how the diode's behave.
Make sure they are both at the same brightness.
1. 500mA
2. 800mA
3. 1500mA
and so on until you get close to your goal.

Also, you should read this and grasp the idea behind it:
http://en.wikipedia.org/wiki/Current_divider

 

[rhd]

bhwollen:
Don't disregard the balancing resistors part. That was the complete key to unlocking a problem with a parallel driver build I did (different objective, but same concept).

I would also think (though I'll disclaim that I might be wrong - and you should defer to anyone who says otherwise on this point) that you could do it in series with a DIY driver, provided that:

- The IC (say an LM317) could handle double the normal voltage requirement. So for 445s, that would be like 10V plus the LM317 voltage drop. You'd need like 3x cells for sure.
- You set the current at the same level you'd set it at for a single diode.

My gut feels happier with a series approach, but I think in reality it would be much harder to implement. I think that with our usual drivers, it's easier to get more current out of them, than more voltage.

 

[bhwollen]

Thanks for your help, but this was an abandoned project. Im now working on my RGB scanner

bhwollen:
Don't disregard the balancing resistors part. That was the complete key to unlocking a problem with a parallel driver build I did (different objective, but same concept).

I would also think (though I'll disclaim that I might be wrong - and you should defer to anyone who says otherwise on this point) that you could do it in series with a DIY driver, provided that:

- The IC (say an LM317) could handle double the normal voltage requirement. So for 445s, that would be like 10V plus the LM317 voltage drop. You'd need like 3x cells for sure.
- You set the current at the same level you'd set it at for a single diode.

My gut feels happier with a series approach, but I think in reality it would be much harder to implement. I think that with our usual drivers, it's easier to get more current out of them, than more voltage.

 

[Leodahsan]

just a heads up, RHD, I've runned a 445 in 12v and 24v (adapter) and everything went fine, just heated more..