[NewCandle] Relaxation oscillator for LED laser driver

[Keith Nagel]

Hey Horace,

What you've drawn is basically Ron's LED driver
circuit! At the frequencies Ron is working at, it may be
better to view C1 as being in parallel with L1. Also, Ron
does capacitive coupling rather than inductive coupling
as you show. These construction details don't detract
from the fact that Ron has built test circuits that ought
to answer any of your questions. 

The issue of the diodes is unclear to me. In an ideal circuit
the capacity of the LEDs in reverse voltage condition
would be zero, thus forcing all current through the
forward conducting branch. In a real circuit, that capacity
might be quite large, and at high enough frequencies
would compete with the forward conducting branch for
current. 

Say Ron, have you measured any of the HF properties of
LEDs? Such parameters as the junction capacity, and
light output vs frequency, would be quite helpful for
engineering the circuits you are experimenting with.
Last time I checked, I had difficulty finding any
of these parameters in the literature. The capacity
would be fairly easy to measure. The light output vs
frequency would be even easier if you have a photometer
that itself is very fast responding. Drive the LED
with a fast risetime pulse, and look for unexpected overshoot in
the resulting photometer output.

BTW Horace, I got that jameco email as well, and got
the right answer! Did you? (grin) Frankly, I've given
up on them as a supplier, as I've gotten burned buying
cheap switch components that were, well, cheap. My
fault I suppose for being cheap, but still the parts
weren't even usable and I would not myself sell them.
Makes me wonder about the quality of many of their
other inexpensive components.

Back in Oct of last year we here were discussing
using avalanche transistors for Ron's circuit. George
Holz was the one making the suggestion. I am more of a
spark gap guy, but there is a lot to be said for these
semiconductor equivalents if you can suffer the lower
power handling.

K.
  

-----Original Message-----
From: newcandle-bounces at ipdiscover.com
[mailto:newcandle-bounces at ipdiscover.com]On Behalf Of Horace Heffner
Sent: Wednesday, February 04, 2009 1:14 PM
To: New energy for the new world.
Subject: Re: [NewCandle] Relaxation oscillator for LED laser driver



On Feb 4, 2009, at 11:40 AM, Dr Stiffler wrote:


> Why not just disconnect C1 and leave this connection open, but you  
> will indeed need D1 & D2. Works well with 80 super white LED's.

C1 exists at some value whether shown or not. By pushing up the value  
of C1 the current duration is pushed up and the frequency dropped,  
and the peak voltage controlled/limited.  The current still sloshes  
back and forth, but the experiment is more controlled.

D1 and D2 might not be needed in the configuration shown depending on  
the values of L1 and C1.  Also, the fact the current path through an  
array has low resistance takes the reverse voltage load off the  
opposed array.


>
> D1 and D2 can be Vishay 1N4148's which are indeed rated at 500mA  
> forward peak and 300mA If.

I don't see the significance of those specific values. I proposed no  
specific values.


> These are the diodes you felt did not exist or at least questioned  
> my specs; in a comment on vortexx.

I don't recall this at all.  Do you have a reference?  This doesn't  
sound like me. I doubt that I would question the existence of some  
kind of diode.  In fact, I think I have a supply of 1N4148's on hand  
somewhere, or something similar with a 1000 PIV rating.


> I know of over 100 people that have driven at least 50 SW's this  
> way. Why close the old loop.

To measure the source of apparent brightness effects.  Also, it would  
eliminate energy loss due to capacitive linkage to the environment  
and thus heating of the environment.

>
> Buy the way JameCo sells the Vishay 1N4148 and they take CE well  
> over 350 volts. Strange little guys.
>
>
> -- 
> ********************************************************************** 
> **
> CONFIDENTIALITY NOTICE
> This e-mail and any attachment are confidential and may be privileged
> or otherwise protected from disclosure. It is solely intended for the
> person(s) named above. If you are not the intended recipient, then
> reading, use, disclosure, copying or distribution of all or parts of
> this e-mail or associated attachments is strictly prohibited. If you
> are not the intended recipient, please notify the sender immediately
> by replying to this message and delete this e-mail and any attachments
> permanently from your system.
>
> NON-DISCLOSURE AGREEMENT
> All parties under a non-disclosure agreement with Dr. Ronald Stiffler
> and/or Stiffler Scientific are hereby notified that the contents in
> this e-mail and any attachments hereto are covered under one or more
> non-disclosure agreements and restrictions as set forth therein.
> ********************************************************************** 
> **
>
> _______________________________________________
> NewCandle mailing list
> NewCandle at ipdiscover.com
> http://ipdiscover.com/mailman/listinfo/newcandle_ipdiscover.com

Please be advised that the content of any correspondence to me,  
Horace Heffner, is placed into public domain unless otherwise  
specified by prior written agreement. I can be contacted at:  
hheffner at mtaonline.net. However, by sending unsolicited information  
to this address, the sender agrees to place it in public domian and  
to make it available for immediate posting to public news forums.

Best regards,

Horace Heffner
http://www.mtaonline.net/~hheffner/





_______________________________________________
NewCandle mailing list
NewCandle at ipdiscover.com
http://ipdiscover.com/mailman/listinfo/newcandle_ipdiscover.com