1. Home
  2. Electronics Design
  3. Clamp circuit help

Clamp circuit help

I'm working on a little battery-powered project where I need to amplify a low frequency signal by about 1000, and clamp the output of the amplifier circuit to remain above about 1 volt at all times. Also, with no signal the output of this section of the circuit should remain at approximately the clamp voltage. The circuit I came up with uses an NPN transistor in the feedback loop for the clamp - I've included the LTSpice netlist.

The circuit seems to work OK - except for the fact that when the clamp circuit engages the simulation shows a big burst of oscillation at the output. I haven't breadboarded the circuit yet to see if it happens "IRL", and I'll most likely be using a different single supply op amp. Anyone have any suggestions on how I might address this, if it shows up in the real circuit as well? Thanks.

Version 4 SHEET 1 1140 680 WIRE 64 -208 64 -240 WIRE 224 -160 176 -160 WIRE 272 -160 224 -160 WIRE 352 -96 336 -96 WIRE 464 -96 432 -96 WIRE 64 -80 64 -128 WIRE 176 -80 176 -96 WIRE 528 -32 400 -32 WIRE 400 0 400 -32 WIRE 400 0 288 0 WIRE -32 16 -96 16 WIRE 96 16 48 16 WIRE 288 48 288 0 WIRE 336 64 336 -96 WIRE 352 64 336 64 WIRE 464 64 464 -96 WIRE 464 64 448 64 WIRE -96 144 -96 16 WIRE -16 144 -96 144 WIRE 96 160 96 16 WIRE 96 160 48 160 WIRE 112 160 96 160 WIRE 336 160 336 64 WIRE 336 160 256 160 WIRE 352 160 336 160 WIRE -176 176 -256 176 WIRE -32 176 -112 176 WIRE -16 176 -32 176 WIRE 464 176 464 64 WIRE 464 176 416 176 WIRE 288 192 288 112 WIRE 288 192 240 192 WIRE 352 192 288 192 WIRE -96 240 -96 144 WIRE -32 240 -32 176 WIRE 240 288 240 192 WIRE -256 336 -256 256 WIRE -96 336 -96 320 WIRE -32 336 -32 320 WIRE 16 336 16 192 WIRE 288 336 288 272 WIRE 384 336 384 208 FLAG 64 -80 0 FLAG 64 -240 Vcc FLAG 384 336 0 FLAG 384 144 Vcc FLAG 224 -240 Vcc FLAG 272 -160 Vref1 FLAG 224 -80 0 FLAG 288 336 0 FLAG 240 288 Vref2 FLAG -96 336 0 FLAG -32 336 0 FLAG 16 128 Vcc FLAG -256 336 0 FLAG 16 336 0 FLAG 176 -80 0 FLAG 528 -32 Vref1 SYMBOL Opamps\\LT1014 16 96 R0 WINDOW 3 14 116 Left 0 SYMATTR InstName U1 SYMBOL Opamps\\LT1014 384 112 R0 WINDOW 3 16 110 Left 0 SYMATTR InstName U2 SYMBOL voltage 64 -224 R0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName V1 SYMATTR Value 2.4 SYMBOL cap 176 144 R90 WINDOW 0 0 32 VBottom 0 WINDOW 3 32 32 VTop 0 SYMATTR InstName C1 SYMATTR Value 0.1µF SYMBOL res 272 144 R90 WINDOW 0 0 56 VBottom 0 WINDOW 3 36 58 VTop 0 SYMATTR InstName R1 SYMATTR Value 220k SYMBOL res 448 -112 R90 WINDOW 0 0 56 VBottom 0 WINDOW 3 32 56 VTop 0 SYMATTR InstName R2 SYMATTR Value 2.2Meg SYMBOL res 208 -256 R0 SYMATTR InstName R3 SYMATTR Value 22k SYMBOL res 208 -176 R0 SYMATTR InstName R4 SYMATTR Value 100k SYMBOL npn 352 0 M90 WINDOW 3 6 153 Left 0 SYMATTR InstName Q1 SYMATTR Value 2N3904 SYMBOL diode 272 48 R0 WINDOW 3 -89 24 Left 0 SYMATTR Value 1N4148 SYMATTR InstName D1 SYMBOL res 272 176 R0 SYMATTR InstName R7 SYMATTR Value 47k SYMBOL res 64 0 R90 WINDOW 0 0 56 VBottom 0 WINDOW 3 32 56 VTop 0 SYMATTR InstName R10 SYMATTR Value 100k SYMBOL res -112 224 R0 SYMATTR InstName R9 SYMATTR Value 1k SYMBOL res -48 224 R0 SYMATTR InstName R11 SYMATTR Value 1Meg SYMBOL cap -112 160 R90 WINDOW 0 0 32 VBottom 0 WINDOW 3 32 32 VTop 0 SYMATTR InstName C4 SYMATTR Value 0.1µF SYMBOL voltage -256 160 R0 WINDOW 3 -233 60 Left 0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName V2 SYMATTR Value SINE(0 0.0005 50) SYMBOL cap 160 -160 R0 SYMATTR InstName C5 SYMATTR Value 2.2µF TEXT -80 -120 Left 0 !.tran 0.5

Reply to
Bitrex
Loading thread data ...

[snip]

Bias the positive OpAmp Input to +1V, then clamp the right hand end of C1 keeping it from going above +1V... an open-collector comparator is ideal for such "DC restorer" work, articularly at your speeds... us an LM339 :-)

You have too much loop gain in your scheme. Maybe just take collector of Q1 to right end of C1? ...Jim Thompson

--
                  [On the Road, in New York]

| James E.Thompson, CTO                            |    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.
Reply to
Jim Thompson

Would be nice, but unfortunately for this design I can't add another chip! One quad op amp is all I got to work with, and the other sections are spoken for!

Wow, works perfectly! I don't completely understand why, though...

Reply to
Bitrex

Sure you do. The cap rolls off the "jerk" ;-) ...Jim Thompson

--
                  [On the Road, in New York]

| James E.Thompson, CTO                            |    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.
Reply to
Jim Thompson

Yup, Collector of Q1 to right end of C1 stops the ringing. ...Jim Thompson

--
                  [On the Road, in New York]

| James E.Thompson, CTO                            |    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.
Reply to
Jim Thompson

Scaling down R1 and R2 by 10:1 seems to fix the burst oscillation. A hundred ohms in the NPN base would be prudent, too, to prevent RF oscillations that Spice might miss.

But using the bias current of the opamp into R11, to make the input offset, works I guess, but makes me feel a little insecure.

John

Reply to
John Larkin

Thanks, I'll add the base resistor. Unfortunately, I don't think I can reduce the resistor ratio, because I need a high pass 3dB point below 10 Hz. I'd like to keep the capacitor value the same since I'm trying to use as many 0.1 uF valued caps as possible.

For what I'm doing I only need the positive half cycle of the input signal anyhow; if the negative going portion gets clipped, say because the DC voltage generated by the bias current against R11 is to low, it doesn't really matter.

Reply to
Bitrex

It certainly warms the cockles of my heart to see Larkin stammering along trying to pretend competence at circuit design. Anyone have a clue what he's trying to say? The bias current of an LT1014 is 30nA MAX.

Maybe next time Larkin will ask Slowman for advice before exhibiting his ignorance. ...Jim Thompson

--
                  [On the Road, in New York]

| James E.Thompson, CTO                            |    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.
Reply to
Jim Thompson

I just noticed that C5 is wrong, it should be 22uF not 2.2uF.

Reply to
Bitrex

You sure don't. Bitrex does. Maybe somebody will take pity and help you.

And your clamp suggestion was idiotic. Spice it and see, since you apparently don't understand it.

John

Reply to
John Larkin

Oh boy...I think I'm just going to get some popcorn and sit this one out!

Reply to
Bitrex

Thompson disagrees with me whether it makes sense or not. He has killfiled me, and hangs on my every word. Crazy old coot.

John

Reply to
John Larkin

Of course I did Spice it... right there, your schematic, just moved the colector, and you verified. ...Jim Thompson

--
                  [On the Road, in New York]

| James E.Thompson, CTO                            |    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.
Reply to
Jim Thompson

he's

He has killfiled me, he wants to debate me, and he is obsessed with me. How sick and stupid is that?

John

Reply to
John Larkin

OK, so long as you use that same opamp and the bias current isn't too high. Using the LT1024 worst-case bias current over temperature, the opamp output is long-gone saturated to its + rail.

JT rants at me for beta biasing, even when it's safe. He missed the fact that this circuit is beta biased, and isn't safe.

John

Reply to
John Larkin

Oh it will definitely break into oscillation for real. You have a HUGE feedback gain of 220K/rE ...not a good way to clamp. What is the problem with the conventional diode feedback off an output volatge divider?

Reply to
Fred Bloggs
[snip comments by village idiot]

While you're munching on your popcorn, here's another way to DC restore...

formatting link
...Jim Thompson

-- [On the Road, in New York]

| James E.Thompson, CTO | 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

formatting link
| 1962 | I love to cook with wine. Sometimes I even put it in the food.

Reply to
Jim Thompson

The first stage works without a pull-up?

John S

Reply to
John S

Oh! I see it now. Clever.

John S

Reply to
John S

And if I should want to use an inexpensive opamp like the LM324, with its higher bias currents, I'm well hosed. In that situation, if I do need a gain of about 1000 from only 2 opamp sections, it seems my only options would be to redistribute the gain more evenly between sections, say 32 and 32, decrease the size of the input resistors, and try to do the "match parallel resistance value to resistor on other input" thing on the resistors in the feedback loop.

Reply to
Bitrex

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.