From: Fred Bloggs
User-Agent: Mozilla/5.0 (Windows; U; Win 9x 4.90; en-US; rv:1.0.1) Gecko/20020823 Netscape/7.0
X-Accept-Language: en-us, en
Subject: Re: Driving the HSDL4230 AIGaAs Infrared LED
Date: Fri, 25 Oct 2002 14:42:22 GMT
NNTP-Posting-Date: Fri, 25 Oct 2002 07:42:22 PDT
Organization: EarthLink Inc. -- http://www.EarthLink.net
Paul Mathews wrote:
> "Markus" wrote in message news:<email@example.com>...
>>I want to drive the HSDL4230 between 4 an 5 Mhz at its max rated current of
>>What component would be the best choice, a high speed mosfet ? or something
>>Have seen it done with a bunch of TLL buffers in parallel, but apart from
>>price tag, i didn't like that method much
> This is normally done using a simple bipolar transistor current
> source. Connect the LED cathode to NPN collector, LED anode to a
> well-bypassed positive voltage source (e.g., 12V), and connect the NPN
> emitter to a resistor Re. You then drive the NPN base with a voltage
> source, for example, a CMOS buffer output. The value of Re can be
> calculated as follows:
> Re = (Vin - Vbe) / I
> where Vin is the drive voltage, e.g. 5V for 5V CMOS logic, Vbe is the
> base-emitter forward voltage on the NPN (use 1.0V for this power
> application), and I is the desired LED current. For 5V drive and
> Re = (5 - 1) / 0.5A = 8 Ohms
> This circuit will deliver more current at higher temperatures, due to
> tempco of Vbe. However, LEDs generally deliver less light output at
> higher temperatures, so this provides some amount of compensation.
> You may also need some small amount of base resistance to stabilize
> the circuit (prevent ringing on edges). Use Rb = 47 Ohms.
> Choose a transistor that has high current gain at 500mA and good
> speed. Have a look at the Zetex line if you're trying to keep size
> down. They have SOT23 transistors with current gain of 350 at 500mA!
> Paul Mathews
He needs to be very quick about it too because 500mA is the "absolute
maximum" repetitive current the device is rated for and this at a
maximum of 20% duty cycle at room temperature. This puts him at
200n/5=40ns pulses at 5MHz PRF. One distinguishing characteristic of
this LED is low Vf at high currents such as 2.5V at 500mA so he
should be able to make it work at 5V with say 1.5 from B to GND leaving
1V for Vce where minimum beta remains high.