2
\$\begingroup\$

Recently I designed a circuit to reduce an input signal by 10x. The design uses a resistor-divider network and then a unity op-amp. The signal is appropriately scaled but there appears to be a ton of ripple / noise in the output. The input signal is -1V to 1V @ 100 KHz and output is a motion control DAC.

I am using a TI OPA188 OpAmp with a +12V / -12V power supply. On both power supplies I have a 1uF ceramic, 100 uF electrolytic and 10 uF tantalum capacitor, and then directly next to the OPA188 I have 0.1uF decoupling capacitor. The schematic is this (note the part is OPA188 not OPA140):

Schematic

Thoughts:

  • Do I need to do something with the power (filter?) from the switching power supplies (TDK-Lambda LS25-12)?
  • I do not have a resistor or capacitor on the feedback from op amp output to negative. Does that cause instability? Ditto with the output - I drive the output directly to the remote ADC, no resistor. Error?
  • Did I make a poor choice - is this op amp not suited to unity gain buffer?

The noise is periodic, at least 200 mV pk-pk (a deal-killer as noise like that is directly picked up by the ADC control circuit). The output cable is a typical BNC 50 Ohm cable, ~3' long, high impedance ADC controller at the other end.

Added - picture of noise (op-amp -> 3' BNC cable -> scope): scope

Added - picture of -12V power rail and op-amp output enter image description here

\$\endgroup\$
5
  • \$\begingroup\$ "directly next to the OPA188 I have 0.1uF decoupling capacitor" A capacitor? Not 2? \$\endgroup\$
    – WhatRoughBeast
    Jun 16, 2017 at 23:06
  • \$\begingroup\$ YOu cable is an antenna and your inputs unbalanced therefore you have massive Common Mode 60 Hz modulated SMPS noise? plus motor noise? . Switch to a matched impedance differential amp using shielded twisted pair for both DC power , and motor and also position feedback with shield to nearby source earth ground only. THen add ferrite sleeve to DC twisted pair wires and also to feedback signals. FLoating power sources and switched loads make great noise generators. \$\endgroup\$
    – Tony Stewart EE75
    Jun 16, 2017 at 23:46
  • \$\begingroup\$ I agree with Tony - it does look suspiciously like a switching power supply is generating noise that is being picked up by the scope. Try connecting the input of the coax to itself rather than to the amplifier - leave the ground connected to see if the cabling is picking up the noise. \$\endgroup\$
    – Kevin White
    Jun 17, 2017 at 0:19
  • \$\begingroup\$ As in the comment by @John - you need to set the scope to unity gain probe (it shows 10x) if you are connecting it directly. If you are not connecting directly then you should be to measure noise. \$\endgroup\$
    – Kevin White
    Jun 17, 2017 at 0:22
  • \$\begingroup\$ FYI - yes two 0.1uF decoupling capacitors \$\endgroup\$
    – bode
    Jun 17, 2017 at 0:40

3 Answers 3

Reset to default
2
\$\begingroup\$

I don't think you chose the wrong opamp, at least not based on what you are showing here. It looks like power supply noise to me based on the obvious correlation between supply and output. Also the noise is periodic at about 15-20ms which would match 50/60hz.

Opamps have a specification power supply rejection ratio, which says how well the opamp can isolate the output from the power supply. Looking at the datasheet for this opamp, PSRR is very good, about 140dB, but that is at DC and PSRR is frequency dependent.

Looking at your second picture, I see about 1V p-p noise on the power rail, and 20mV on the output (not 200mV - wrong probe setting?). You are seeing maybe 35dB PSRR in this configuration, from 1V to 20mV. Looking at Fig 12 in the datasheet (PSRR vs Frequency) this ratio happens around 10-100kHz. It wouldn't be surprising if you have that sort of noise on your power rail, switching converters typically run 50khz upwards.

You should probably add some filtering on both power rails to ground using L-C filters. There is no easy answer on what exactly to use and it may not be simple to do this 'properly' - this PDF has some good information. However if it were me, I would grab some power inductors and capacitors, check the cutoff frequency using the formula, and give it a try to see if it helps reduce the noise. For example I might try 22uH/47uF for 5kHz cutoff.

\$\endgroup\$
1
\$\begingroup\$

According to the data sheet the part an auto-zeroing opamp. OPA188 datasheeet

If you don't need the very low offset voltage I don't recommend this type of opamp as they often have a significant amount of clock feedthrough.

(I answered a similar question hereIdentifying Source of Periodic Artifact at Op-Amp Output)

What is "a ton of ripple / noise in the output"?

What noise do you see on the power supply (especially the negative supply as many opamps are sensitive to that)?

Make sure it is not a measurement artifact - connect the scope probe to ground near the opamp, if the noise does not go away it may be due to the way you are grounding the scope.

EDIT : I initially had the wrong part number, the one you have on the schematic.

\$\endgroup\$
5
  • \$\begingroup\$ FYI I am using a BNC (50 ohm coax) directly to the oscilloscope. I will (try) to measure the power supply noise. The noise has been all over the place but it is definitely significantly greater than the source voltage (the PCB has a bypass mode so I can measure both). As far as auto-zero I do not need that feature - is there a better op-amp you'd recommend? \$\endgroup\$
    – bode
    Jun 16, 2017 at 22:57
  • \$\begingroup\$ What is the frequency of the noise? Are you driving the coax directly from the opamp? You may need to add a series resistor (20-100 ohm) between the opamp and the coax to avoid instability. \$\endgroup\$
    – Kevin White
    Jun 16, 2017 at 22:59
  • \$\begingroup\$ Note that with a direct coax you need to set the probe to 1X. \$\endgroup\$
    – John Birckhead
    Jun 16, 2017 at 23:39
  • \$\begingroup\$ To clarify: picture 1 (ch2 only) is @ 1x probe (BNC cable). Picture 2 (ch 1 + ch 2) is @ 10x because I am using a 10x probe on the power supply (the BNC is still plugged in). That's why the CH2 scale is smaller (20x smaller, so spikes are 2x bigger). \$\endgroup\$
    – bode
    Jun 17, 2017 at 0:46
  • \$\begingroup\$ OK, sorry my misinterpretation. \$\endgroup\$
    – Kevin White
    Jun 17, 2017 at 2:03
1
\$\begingroup\$

You have a fair bit of capacitive loading on the op-amp from the scope input and coax.

As @KevenWhite said above in a comment, you should add something like 51\$\Omega\$ in series with the op-amp output (right at the output) if you want to use a direct x1 connection to your scope and further connections to the rest of your system.

enter image description here

As far as what might be triggering the impulses, look for switching power supplies (including possibly a brick on the scope itself- even a Tek scope- it happens) and move any cell phones well away from your lashup.

\$\endgroup\$
2
  • 1
    \$\begingroup\$ Notice OP changed what op-amp they said they're using, and the new one has more overshoot than the OPA140 with 50 ohms between it and a capacitive load. \$\endgroup\$
    – The Photon
    Jun 17, 2017 at 4:03
  • \$\begingroup\$ I do see a ton of noise out of the SMPS. What would the basic idea be to fix it? Is it "filter" or is it "use an LDO to regulate -12V to -10V and 12 to 10V", especially since I do not ever approach the rails? \$\endgroup\$
    – bode
    Jun 17, 2017 at 15:56

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.