Hello all,
I "upgraded" two quad opamps in a simple 4 channel synth mixer.
TL074c= out
AD8513 = in
(soic)
AD8513 datasheet.....LINK
I then checked for stability on my scope. I did not see "ringing" but did see "overshoot" at the transients, generally under 2kHz. Over 5kHz, they mostly even-out.
Here are two screenshots about 1kHz.
Yellow = source
Green = mixer output
(I made green's gain is a little higher for visual clarity)
Second one is the same, just zoomed-in.
Gain pot is around typical listening level, no load. Cables are short.
Here is the Mixer schematic...... https://github.com/pingdynasty/Mix/blob/master/hardware/LinearMixerRev04-schematic.pdf
Notes:
There is no overshoot out of the first-stage quad opamp, only out of the second quad opamp, at the output....before (or after) the 1k ohm output resistor.
I tried putting a snubber network on one output, as suggested by the AD8513 datasheet, with no affect at all.
I also added 0.1uf psu bypass caps on the affected quad opamp, before testing, as a precaution.
I added some 4 foot length cables to the output just to see what would happen.....no effect/difference.
Frankly, it "sounds" fine. Should I even worry about this ?
I "upgraded" two quad opamps in a simple 4 channel synth mixer.
TL074c= out
AD8513 = in
(soic)
AD8513 datasheet.....LINK
I then checked for stability on my scope. I did not see "ringing" but did see "overshoot" at the transients, generally under 2kHz. Over 5kHz, they mostly even-out.
Here are two screenshots about 1kHz.
Yellow = source
Green = mixer output
(I made green's gain is a little higher for visual clarity)
Second one is the same, just zoomed-in.
Gain pot is around typical listening level, no load. Cables are short.
Here is the Mixer schematic...... https://github.com/pingdynasty/Mix/blob/master/hardware/LinearMixerRev04-schematic.pdf
Notes:
There is no overshoot out of the first-stage quad opamp, only out of the second quad opamp, at the output....before (or after) the 1k ohm output resistor.
I tried putting a snubber network on one output, as suggested by the AD8513 datasheet, with no affect at all.
I also added 0.1uf psu bypass caps on the affected quad opamp, before testing, as a precaution.
I added some 4 foot length cables to the output just to see what would happen.....no effect/difference.
Frankly, it "sounds" fine. Should I even worry about this ?
Last edited:
Can you post the schematic here?
Are there any adjustable tone circuits in the path? There also seems to be some LF phase lead.
Are there any adjustable tone circuits in the path? There also seems to be some LF phase lead.
No, they were 14pin solder-only SOIC....so yeah, I took the risk, and measured afterwards.
The schematic is linked in my prior post. (you can view it without downloading it)
I've been watching this video/training series on "Op Amp Stability" from TI...... but did not see my "exact problem".... which I originally thought was capacitive loading or stray-capacitance (pcb layout).
It was mentioned somewhere that the small overshoot without "ringing" (bouncing) is not really a problem.
Found it....
Do-it-yourself: Three ways to stabilize op amp capacitive loads
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
I do have an identical mixer that still has the original TL074's in place..... so maybe tonight I'll do an A/B/X sound comparison, and also some scope views.....under some real-world loading and cabling.
How do I do A/B/X you might ask ?
Well thanks to the miracle of modular synthesizers.... there are a few ways to "switch/swap signals" as desired, or automatically.
One method I use, is a module by Doepfer called the A151 Sequential Switch. It can be bi-directional, so either 1 signal sequentially switching to up to 4 outputs....or (up to) 4 inputs sequentially into 1 output. The "rate" or time held for each switch is determined by an adjustable clock or trigger or gate....pretty-much click-free. I can also configure an audible pause or ping denoting a change.
So if I don't look at the led's on the module and how I remember hooking it up.......I won't know which source I'm listening to except for a ping. Perhaps leave the room, come back, listen, and write down some notes.
It seems that you ignored some of the ad8135 datasheet.You get better explanations in Barrie Gillbert and Scott Wurcer datasheets for Ad745 and Ad797.
