Joe--What causes this???

[laatsch55]

A waveform within a waveform--the top pic is a magnified version of the bottom one.

 

[Gepetto]

a high frequency oscillation riding on the main signal Lee.

 

[laatsch55]

One of the RC4739's IS BAD then.

 

[Gepetto]

Possible but not certain Lee. Decompensated 739 op amps are very sensitive to layout due to wide bandwidth

 

[Gepetto]

Hi Lee
Which segment of the 4739 are you noticing the oscillation on?
Z1 pin 1?
Z1 pin 13?
Z2 pin 1?
Z2 pin 13?

 

[Gepetto]

Hi Lee also please give me the time base setting on your scope for the magnified version of you scope trace (100_846.JPG) so I can approximate the frequency it is oscillating at.

 

[Gepetto]

Lee
Sorry for all the questions but...
From a decoupling capacitor perspective, it appears that Z1 is bypassed by C6 on pin 14 (B+)and C13 on pin 7 (B-). I notice no high quality ceramic or film bypassing across C6 or C13 which is questionable design practice on a wideband op amp with 3MHz GBW product. There should be a 0.01uF or 0.047uF or 0.1uF cap from pin 14 (across C6) and pin 7 (across C13) of Z1 to ground.

I see no bypass at all on Z2 and there should be at least the 0.01uF type bypass as recommended above. My best guess is that pin 1 of Z2 is your oscillation culprit but that is just my guess based on the fact that it is a unity gain stage. Unity gain configurations are the least stable op amp configuration.

 

[laatsch55]

Joe, i've been busy getting stuff ready for tomorrow. .2ms was the setting on the scope, and I was scoping the outputs. I can scope the pins here in a bit.

 

[Gepetto]

Take your time Lee, it's your project after all

 

[Gepetto]

Based on your scope setting of 0.2ms per division, it looks like a 30Khz oscillation

 

[laatsch55]

OK, just got in from our oil recycling project 120 miles away. I'll get that info to ya Joe. Regardless of whether It's Z1 or Z2, should I bypass the pins as you suggest just for good measure? And where do I tag ground? On the IC ground pin?

 

[Gepetto]

For Z1 bypass right across the leads of C6 and C13. For Z2, bypass across the leads of C40 and C41 which are close to Z2. In addition I recommend a ceramic 0.1uF from pin 7 to pin 14 on Z1 and Z2. This is a does no harm type of improvement. I notice that these big electrolytic bypass caps are a bit away from these high speed ICs.

This treatment will provide both common mode and differential bypassing. A preamp really should have regulated power supplies for better noise rejection but I noticed they flew economy on this design.

 

[laatsch55]

Pin 1 of Z2 joe, scope setting---1us

 

[laatsch55]

Pin 1 is where the little dot is correct??

 

[laatsch55]

Pin13 also.

 

[laatsch55]

Would MKT's work for the .1??

103 KCK on the ceramic??

 

[Gepetto]

Correct Lee the dot is pin 1. So the time base was 1 usec instead of 0.2 msec?

If that is the correct assumption then the oscillation is at ~ 6 Mhz

Is it more pronounced (higher amplitude) at pin 1 than at pin 13 of Z2? My suspicion is yes but need you to confirm that.

 

[laatsch55]

That is correct Joe. Pin 1 much higher.

 

[Gepetto]

either 103K or 104K will work Lee.

after you do the bypassing, it may be gone but I suspect not. Recommend you try putting 33pF cap across R49 which is a 22K resistor on the schematic.

 

[Gepetto]

the 33pF should kill it if my guess is right. If it works, I will explain what is up. If not, I will go silently into the night