Standard non-inverting opamp with a cap across + and - terminals...... Why?

Its possible the (-) input was meant to be at common. Unless it had some other biasing network attached to the (-) input other than that cap, I'm sure it was intended to be at common. This would imply that the op-amp is being supplied with a +&- rail with respect to common in which case, one of the inputs if not used as an input will also be going to a common of some sort.

For basic description of op-amp operations, it is common to find material assuming the use of a dual supply rail. With this in mind, a basic inverting op-amp would have the (+) input tied to common and the (-) input being loop back biased etc...

In your case, the inputs are switched.

ing

de

g an

al.

Sometimes placing a series RC across the inputs of an opamp is done as a way to lower the loop gain. This may be the intended purpose.

Also the capacitor may be keeping RF and spikes from messing with the op-amp at the cost of making the loop less stable.

I've placed 100pF or so there but 0.1uF is a bit much. If cell phone RF and noise like that is a concern it is usually better to migrate to a CMOS opamp. This gets rid of rectification effects in the BE junctions because then there ain't no BE junctions no more ;-)

I've seen this a couple of times, but it never made sense.

Just today, I was thinking about hanging a cap to ground from the output of an opamp. I've seen that done, too. I need a voltage-adjustable node that's stiff from zero Hz to GHz. A National C-load opamp would work, but they're expensive. So a cheap opamp with a cap or two to ground is interesting.

I suppose I'll have to Spice the damned thing.

John

hing

ode

ng an

ial.

Wouldn't this be subject to the same issues as designing a LDO, i.e. in terms of stability. Driving a capacitive load is pretty ugly for anything other than a OTA where the C load provides compensation.

It is a mistake, from the point of view of any normal AC response.

What was the expected response of the A/D converter, in terms of speed or output? Oscillation above a certain frequency might be ignored, though there would be severe compression and loss of accuracy if the op-amp output pegged on the rails, while wobbulating.

I've seen DC reg loops that had oscillating intermediate gain stages, in production. Same kind of undamped C across op-amp input pins, but more modest in value, by an order of magnitude or so. It was not the biggest worm in the can.

Can you reference the publication title / author?

RL

ive

tching

al.

Bode

ting an

nd

ecial.

Actually, it doesn't get rid of rectification effects (because there aren't any) but increases the linear signal voltage range at the op amp input.

MOSFETs typically offer one or two orders of magnitude less transconductance than bipolar transistors, so that where the bipolar input stage produces a substantially non-linear ouput for input signals over 26mV (at room temperature) effectively "rectifying" larger input excursions, the MOSFET input stage stays linear for input signal ranges of up to about a volt.

Joerg has never got stuck into the mathematics of what is going on, so he still thinks that high frequency noise is being rectified by the BE junction, and consequently can't believe that the gate of a MOSFET can "rectify" a (bigger) signal in exactly the same way.

-- Bill Sloman, Nijmegen

Surf back thru previous posts in regard to OpAmps driving capacitive loads... stability achieved with a simple feed-forward RC.

BTW: Many CMOS OpAmps ARE transconductance-style... gain dependent on load... look for output stage with drain-to-drain.

...Jim Thompson

No, one order of magnitude is MOSFETs vs. BJTs... two would be BJTs vs. tubes!!

Incidentially, I bet the "input linear range" on the old K2-W was pretty wide...

Tim

He doesn't just think that, he's experienced it. Or rather, a client has and now it's gone :-)

Oh, and it's got nothing to do with 26mV.

As usual, Slowman, utilizing inadequate knowledge of semiconductors, obfuscates.

I'm always pleased to note that I'm the highest standard for Slowman's disdain, but please don't feed the jerk. Let him die that most unpleasant of deaths... alone ;-)

...Jim Thompson

You seem to insist on using actual real-world results for your appraisals rather than just a bunch of math and theory on paper. Most annoying. :-)

On a sunny day (Wed, 08 Apr 2009 09:59:47 -0700) it happened Joerg wrote in :

I could be wrong, but I think this depends as much or more on the circuit topology then on the transfer curve of the semiconductor in question. The word 'common mode' comes to mind. Although Bill Slowman may have a point somewhere mathematically, nobody is driving RF into the base of a transistor with the emitter grounded. In a normal diff amp with current source things are different. Again, too much book reading and too little hands on experience from Bill. The Electroshock did not rectify that (pun intended). I would like to hear about this from Winfield Hill actually. I work all the time here with strong RF fields, and in the past with even stronger ones. to the point where transistors simply died in the box.... But I have also been successful in getting rid of the RF so far in millivolt circuits. RF decoupling and filtering is an art too.

Well, problem is nobody told the cell phone RF field. And then there are those pesky little thingamagics in chips: Parasitic capacitors. At

stronger ones.

circuits.

Sometimes a microvolt of rectified RF can throw a signal aquisition scheme completely off the rocker. That's what happened at this client and I actually had to come out there to figure it out. When I approached their building and saw a large airport and several big jets really close to their site my first thought was "Oh, oh!"

The best was when I was in the middle of audio measurements and told the group around me "One of you must have a GSM phone" .... "Ahm, ahem, yeah, me. How the heck do you know? It's in my briefcase. How'd you see it?"

On a sunny day (Wed, 08 Apr 2009 13:04:49 -0700) it happened Joerg wrote in :

stronger ones.

circuits.

Yes, cellphones, GSM, that is well known, my old walkman could hear them. A XkW CB transmitter with a 25 meter high antenne nextdoors neighbour will light a light bulb here.... connected to the coax. I asked him about that..... well.. I think my GPA at that distance was just the director .... perfectly tuned too. My own transmitter makes stripes on the LCD!!!!very fine pattern you just have to look close. The HDSPA modem for the eeePC does much the same as a GSM, indeed breaks through the audio sometimes.

s and

-)

You can get good real world results with remarkably inadequate theory. Jeorg under no obligation to justify his (good) results with any kind of theory that makes sense, and his proposition - that base-emitter junctions are directly rectifying millivolt levels RF signals and turning them into curents flowing the through the base-emitter junctions - doesn't make any kind of sense at all.

The proposition that an NPN long-tailed pair follows the positive envelope of an RF signal does make sense and does provide a non- magical explanation the "rectification" that we've all seen.

I didn't invent this theory - though I'm damned if I can remember where I came across it - and I got to believe it when I had to stop some sensitive front end from picking up the local radio station, and used it work out effective remedial action. It's all about 25 years ago now, so the details are a little vague.

-- Bill Sloman, Nijmegen

The topology was a long-tailed (differential) pair.

Parasitic capacitance at the emitter node makes that pretty much the case at UHF, right? You'd be driving RFI into the base(s), with the emitter(s) quasi-grounded. The transistor's live, making it a pre-biased detector.

If you've got equal antennae at the inputs then you'll rectify a common mode signal. If unequal, a normal mode voltage.

Even if the induced signal is common-mode balanced, it's likely still a problem: op-amps' CMRRs generally suck @ 1GHz, so common mode RFI shows up at the output, peak-detected & low-pass filtered.

No 26mV required.

Cheers, James Arthur

stronger ones.

circuits.

It's probably for the same reason, rectification in BE junctions.

light

One of those Italian power amps that routinely get confiscated by the Rijkspolitie?

If he has lots of tattoos it's probably wise not to ask too much ...

through

Yeah, all this didn't happen in them tube days :-(

Exactly. Continuous duty RF often doesn't present a problem. Gets rectified but nobody notices. When the RF pulsates like in a GSM sync phase, that's when things hit the fan. Still, it took me a while to convince the client that this was what happened. When the engineer finally agreed to swap for a CMOS opamp and the problem completely disapeared that cinched it for them. Initially he offered a wager in the currency "beer" but since I knew I'd win I thought this wouldn't have been the proper thing to do as a consultant.