H.F. "Super Diode" Problems

Thread Starter

Ragwire

Joined Dec 9, 2013
36
Hi, all, trying to build a precision rectifier with no good results.
I only get VEE (about -9V) on the output pin 6 and nearly the same on input pin 3. Disconnecting the DC filter from the output, and trying with and without the 100K load resistor, still shows no rectified pulses at the output. I tried without the 100K input load, too, and various values of load (DC return) resistors, and feedback resistors too. I also tried without the buffer. Nothing I did actually makes this work. I tried with an LM741 as well, feeding 1KHz at 150mV P-P sine...same behavior. Disconnected the power supply filters and that didn't help.

What am I missing? Thanks...

 

ronv

Joined Nov 12, 2008
3,770
The LT1222 is not stable at a gain of 1.
What is the frequency and amplitude of your input? That's a very high frequency op amp.
 

#12

Joined Nov 30, 2010
18,224
You guys came up with rather good responses, but I'd like to check the basics. Is the +9V supply actually connected? Did you connect the first 100k resistor to the middle of the power supply? The first thing to do here is to resolve the DC issue on the first stage, and yes, it might be oscillating.
 

Thread Starter

Ragwire

Joined Dec 9, 2013
36
It is supposed to be a half wave rectifier, yes. The +/- rails and ground are positively OK (no pun intended), and there is no instability shown on the scope trace. I have seen these break into a 10MHz to 20MHz oscillation plenty of times in amplifiers, but that is not happening. I tried from 1kHz to 100kHzl 100mV to 2V P=P inputs. I subbed out the op amp twice with new '1222s, and even tried a 741. It seems just like the fundamental circuit theory is all wrong somehow. I'm stumped good.
 

Thread Starter

Ragwire

Joined Dec 9, 2013
36
I also tried 1N4148s and with and without the buffer. This is actually my third try building a precision rectifier in the last two years, by the way, and none of the topologies worked...not ever.
 

GopherT

Joined Nov 23, 2012
8,009
Are you using one battery or two? You do realize that the ground symbol is between the two batteries (half of the max voltage).
 

ronv

Joined Nov 12, 2008
3,770
Well for sure the first stage is just a buffer with a gain of 1. You can start there. With no signal applied the input should be at ground thru the 100k and so should the output.. But....

It should be noted that the LT1222 is not stable in
AV = 1 unless CC = 50pF and a 1k resistor is used as the

feedback resistor. The 1k and input capacitance increase
the noise gain at frequency to aid stability.

Try it with a vanilla op amp. Say a LM324 or 358.
 

Thread Starter

Ragwire

Joined Dec 9, 2013
36
Two batteries as in the shematic above. The schematic is mine and is what it is wired to.

Well I'll go get a 324 today and give that a try. I am digging into the datasheet for the LT1222, and I do see what you mean. However, I tried with a 3K and 1K to see if I could get a gain of three and still, no bueno.

Thanks, all.
 

#12

Joined Nov 30, 2010
18,224
Just a note from experience here. Set your scope to read DC when measuring for oscillations. If the DC voltage suddenly snaps to the correct level while measuring, the capacitance of the probe is quenching the oscillations. Been there, done that. :oops:
 

GopherT

Joined Nov 23, 2012
8,009
I think you need to switch the input pins on the second op-amp.
Yes, that circuit has positive feedback so will latch at the rail. You need to interchange the input pins of the second op amp so that the input and feedback goes to the (-) input and the (+) input is grounded.
Good call guys!

#12 was also on the right path when he said we should go back and start with the basics.

Cheers!
 
Top