Alt-BEAM Archive
Message #05420
To: beam@sgiblab.sgi.com beam@sgiblab.sgi.com
From: Sean Rigter rigter@cafe.net
Date: Monday, 26 July 1999 6:32
Subject: Re: h-bridge problems
>The bottom line is: LED series current limiting resistors of 500 ohm to
>1K are fine for visual indication of the output state of 74HC logic at
>Vcc=5V. At lower Vcc, you can reduce the resistor values over a range
>from 500 ohms at down to 0 ohms at about 2.0V (the latter strongly
>dependend on the type of LED).
>
>Want to know why? Read on!
>
>For reasons of meeting requirements of logic levels the voltage drop on
>the outputs of 74HC logic caused by output current must not exceed
>certain limits. Based on Ohm's law, the maximum allowable output current
>is therefore determined by the HC logic output resistance . An 74HC240
>can sink 3 times as much current as a 74HC14 and so can have three times
>smaller LED resistor and brighter LEDs. The lowest practical super
>efficient LED current is about 2 ma. With a 74HC14 and a Vcc of 5V,
>there is a typical 1.9V drop across the LED which leaves 3.1V across the
>resistor and HC driver output resistance requiring about 1.5K to limit
>the LED current to 2 ma. So at 5V use 1.5K max and 390 ohm minimum for
>LED currents from 2 ma to 8ma. With a 74HC240 a minimum LED resistor
>value can be reduced to 200 ohm which results in about 20ma of LED
>current which may be required for very low efficiency LEDs. Without a
>resistor at Vcc=5V, the LED current can be 70-150ma which is enough to
>damage the LED and the HC logic in a very short time if not instantly.
>
>Minimum resistor values at the other end of the Vcc spectrum are
>(approximately): Vcc=4V - Rled = 330 and Vcc =3V - Rled = 150 ohm. At
>Vcc voltages between 1.8 and 2.2V the value is a little uncertain since
>it strongle depends on the forward voltage of the LED. Iit may even be
>necesary to remove the LED resistor to get enough LED current since the
>HC driver output resistance limits the current. Owing to the relatively
>high LED Vf compared to Vcc= 1.8-2.2V, the effect of output loading on
>logic voltage levels is minimized.
>
>I found that the absolute minimum supply voltage that can be used with a
>2ma LED is about 1.65V for some types of red LEDs. . More typical red and
>green LED minimum voltages at 2 ma is around 1.9-2.0V.
>
>I recently posted a 100ma Low Drop-Out voltage regulator design that
>exploits this low Vcc concept to the max by using an actual LED as the
>regulator voltage reference. That way the LDO regulator holds the 74HC
>Vcc supply voltage at the optimum level for HC outputs to drive LEDs
>directly. In addition, the LED Vf referenced regulator tracks Vcc with
>temperature for constant LED brightness over a wide input voltage range
>of 2.1V to 9V.
>
>As Steven Bolt has pointed out many times, for low power applications
>there is a big advantage to operating HCMOS at low voltage ie Vcc<2.5V
>compared to Vcc=5.0V. For example, at Vcc=5.0V active oscillators (Nv)
>and voltage detectors (Nu) can produce peak supply currents bumps (Ipeak)
>of 20 ma per stage and average currents of many ma. But at Vcc<2.5V
>these are typically reduced to Ipeak<1ma and I average <100ua. This
>means that using a low power LDO voltage regulator can significantly
>reduce overall power and give consistent performance for many BEAM
>applications over a wide range of supply voltages.
>
>enjoy
>
>wilf
>
>SG wrote:
>
>> 10K ? wow! i was told just 1K at the highest, 500ohm at the lowest.
>> -Sparky
>>
>> At 10:26 PM 7/23/99 +1000, you wrote:
>> >done that, 10k, nothings caught on fire yet.
>> >
>> >
>> >-----Original Message-----
>> >From: SG
>> >To: beam@sgiblab.sgi.com
>> >Date: Saturday, 24 July 1999 12:48
>> >Subject: Re: h-bridge problems
>> >
>> >
>> >>>Please help, you wouldn't want all the
>> >>>prizes to go to Darrel :-) David Perry
>> >>
>> >>be sure to put resistors on yur LEDs!!!
>> >>heh heh
>> >>-Sparky
>> >>
>> >
>> >
>
Home