Post by thread:78L05-like regulator which sinks excess vout

I've got a circuit which I'm having a bit of an issue with. I have a very small circuit with a single 74HC series IC and a few passives... along with currently a LM2936Z-5 voltage regulator. The total +5V rail draws something well under 1ma. The problem I'm running into is that the function of this circuit involves handling a zero to +24V input very quickly. As such, I have an input circuit which consists of a 10K resistor feeding an input - which has external clamp diodes. I can't really use less than 10K as the input capacitance of the 74HC is high enough that anything more than a 10K resistor results in too much end-to-end delay. The problem I am having is that clamping the 24V input to 5V results in current being sourced into the +5V line on the neighborhood of 2ma. Because the LM2936Z-5.0 that I'm using lacks the ability to drain off excess, the +5V rail sits well over +5V most of the time. I'm considering various options, but my ideal situation would be a drop in three-terminal TO-92 regulator which is capable of acting as a current sink when needed since I already have a quantity boards made and really don't want to scrap them and start over. I believe (but am not sure) that a "real" 78L05 will do this, but I also need this to work in sub-zero temperatures, which a 78L05 is not rated for. So, other than adding a drain resistor, or re-working the board to use a zener regulator circuit instead, does anyone know of a suitable part, or something I've likely missed? -forrest

> So, other than adding a drain resistor, or re-working the board to use a > zener regulator circuit instead, does anyone know of a suitable part, or > something I've likely missed? Afaik a 78(L)05 will *not* sink current to maintain its 5V output voltage. In fact, raising the output above the input is a know way to kill the regulator. Maybe use a bi-directional transzorb or similar instead of the two clam diodes. Wouter van Ooijen -- ------------------------------------------- Van Ooijen Technische Informatica: http://www.voti.nl consultancy, development, PICmicro products docent Hogeschool van Utrecht: http://www.voti.nl/hvu

Wouter van Ooijen wrote: > Maybe use a bi-directional transzorb or similar instead of the two clam > diodes. Hmm.. that's an idea.. I could definately replace the lower clamp diode with something like that.. Or perhaps a unidirectional one (aka 4.7V Zener-ish device) instead so that it still will clamp to zero. -forrest

How about a bipolar transistor (collector to +5V, emitter to 0V) with a 4V7 zener connected base-collector. Should start to conduct in the 5.2 - 5.3V area. But it is 2 components. RP On 01/04/2008, Forrest W Christian <spam_OUTforrestcTakeThisOuTimach.com> wrote: {Quote hidden}> Wouter van Ooijen wrote: > > Maybe use a bi-directional transzorb or similar instead of the two clam > > diodes. > > Hmm.. that's an idea.. I could definately replace the lower clamp diode > with something like that.. Or perhaps a unidirectional one (aka 4.7V > Zener-ish device) instead so that it still will clamp to zero. > > -forrest > > > -

At 02:13 AM 4/1/2008, Forrest W Christian wrote: >The problem I'm running into is that the function of this circuit >involves handling a zero to +24V input very quickly. As such, I have >an input circuit which consists of a 10K resistor feeding an input - >which has external clamp diodes. I see 2 simple solutions - one of those might not even require extra parts. You mention "external clamp diodes". Are these discrete (axial) diodes? If so, simply replace the lower diode with a suitable zener. Then omit the top diode (to Vdd). The other easy solution is to simply add a 5.6V 5% tolerance zener diode between Vdd and Gnd. You should still be within the absolute maximum voltage rating of the CMOS chip. dwayne -- Dwayne Reid <.....dwaynerKILLspam@spam@planet.eon.net> Trinity Electronics Systems Ltd Edmonton, AB, CANADA (780) 489-3199 voice (780) 487-6397 fax http://www.trinity-electronics.com Custom Electronics Design and Manufacturing

On Tue, 2008-04-01 at 02:13 -0600, Forrest W Christian wrote: > I've got a circuit which I'm having a bit of an issue with. > > I have a very small circuit with a single 74HC series IC and a few > passives... along with currently a LM2936Z-5 voltage regulator. The > total +5V rail draws something well under 1ma. > > The problem I'm running into is that the function of this circuit > involves handling a zero to +24V input very quickly. As such, I have > an input circuit which consists of a 10K resistor feeding an input - > which has external clamp diodes. I can't really use less than 10K as > the input capacitance of the 74HC is high enough that anything more than > a 10K resistor results in too much end-to-end delay. How about instead of the diode clamps: Vcc ^ | \ / ~10k \ / signal ---|<------+----------> IC Only issue I see is whenever the signal is low you'll be drawing current. TTYL