I need to 'hard' clip the output at 2.6V.
+-------C---+ | |\\ |-----R--+--|- \\----+----out | / gnd |+/
Conditions Vsuppy= 5V The op amp is a rail to rail output type. Iq op amp =50uA C=33nF R=1k Rsource=50ohm I know 2 ways to do this:
1) Add a clipping cct on the output. (shunt regulator or op amp clipper cct) 2) Reduce the op amp supply to 2.6V (Ex. R divider or voltage regulator.)Any other ways?
D from BC myrealaddress(at)comic(dot)com BC, Canada Posted to usenet sci.electronics.design
If you clip the output, with some sort of shunt or by reducing Vcc, the input side can "wind up" when the output clips, namely drift well off ground. So the integrator may take a long, unpredictable time to recover from clipping.
Diode clamping the inverting node helps that, but the best clamp uses feedback.
Try a PNP transistor: emitter to Vout, base about +2 from a divider, collector to inv pin of opamp. That will clamp about +2.6 (but doesn't clamp at zero.)
Didn't we do a thread on opamp clamping a while back?
John
Two volt zeners back-to-back (2V + .6V) across the capacitor? If it's a single-ended application a 2.6V zener works.
Programmable Zener (TL431) across C. The resistors that program it will have some gain effect, though.
If you need it to clip/clamp accurately, use something like an LMV431 (or other CMOS equivalent) Ref from a resistive divider from Out, Anode to summing node, Cathode to +5V. Whopping amount of loop gain, may need taming for stability.
...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC\'s and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I love to cook with wine Sometimes I even put it in the food
Low-voltage zeners are awful.
John
iirc...low voltage zeners act like 'warped' resistors.
I should mention that my app needs to function like an integrator when the voltage is below 2.5V.. So I'm not confident about using a zener in the feedback path.
D from BC myrealaddress(at)comic(dot)com BC, Canada Posted to usenet sci.electronics.design
ah..ic.. With the output clipped post circuit, C might keep charging and charging and charging.. Once the input is at a level that no longer causes clipping, it takes time for C has to discharge and that maybe a problem..iows..perhaps call it integrator lock up.
The PNP solution sounds good. It puts a limit on the capacitor voltage. I might try that. Thanks..
Nope..can't recall other clamping threads..
D from BC myrealaddress(at)comic(dot)com BC, Canada Posted to usenet sci.electronics.design
Sure, use the 2.5V version of the LM385. My point was to simply kill the gain where one wants to clamp.
A LMV431 has a Iz(off) of 0.001uA
0.001uA * 1k = 1uV 1uV * Avopen(opamp) = perhaps 1V So that's a 1V offset. I can deal with that.Interesting.. :)
D from BC myrealaddress(at)comic(dot)com BC, Canada Posted to usenet sci.electronics.design
I'm thinking about that.
D from BC myrealaddress(at)comic(dot)com BC, Canada Posted to usenet sci.electronics.design
...maybe a low power reference diode (2.5V prolly as close as you can get onthem)?
Low voltage zeners ae sooooo slowww that a forward diode is almost better....
Maybe this?
Bob
If this is not just a floobydust experiment, think it over carefully, and add diodes appropriately, to avoid offset.
No time this morning to show you, I'm off to an appointment. Maybe this PM.
...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC\'s and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I love to cook with wine Sometimes I even put it in the food
Call it integrator lock up if you will, but the rest of the world has called it integrator wind up for decades, if not over a century.
Besides, "lock up" implies that it never unwinds.
-- http://www.wescottdesign.com
I thought John was just being metaphorical with wind up. I didn't know that was the official name.
D from BC myrealaddress(at)comic(dot)com BC, Canada Posted to usenet sci.electronics.design
That clips?
D from BC myrealaddress(at)comic(dot)com BC, Canada Posted to usenet sci.electronics.design
I think I can figure out how to null out the LMV431 Iz(off) current.
D from BC myrealaddress(at)comic(dot)com BC, Canada Posted to usenet sci.electronics.design
Google for it, and you'll find all sorts of material on it's deleterious effects on the stability and settling of control loops that include an integrator.
It's definitely something that you have to pay attention to if you're going to do serious control system design.
-- http://www.wescottdesign.com
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.