Low frequency Noise of Zero Drift Amplifiers

[chuckb]

It appears some choppers do not completely eliminate the basic flicker noise.

We started discussing the low frequency performance of Zero Drift Amplifiers in the "Re: DIY low frequency noise meter and some measurement result of voltage references" topic a month ago.

My interest started with some information on the Analog Devices web site about the ADA4522 AutoZero Amplifier.
https://ez.analog.com/amplifiers/operational-amplifiers/f/q-a/102479/ada4522-1-f-noise

This other paper by David Hoyland (2016) documents the flicker noise performance of several other chopper opamps below 0.1Hz. It also has the current noise spectrums.
https://dcc.ligo.org/public/0126/T1600206/001/Opamp%20Noise%20Test%20Results.pdf

Then Echo88 found this additional paper

Maybe this doc is relevant in this case: http://www.sensorsportal.com/HTML/DIGEST/january_2011/P_745.pdf 

The results from the above papers and my testing are summarized below. I have attached the pdf results from my testing of six different amplifiers. Later posts will have more details of the testing and the results. For my testing of the ADA4528-2, ADA4522-1 and the OPA189 I ended up repeating all the tests with different amplifiers. The results of the second round of tests were within 1dB of the first.

So far -
flicker noise free down to 0.0014Hz
ADA4528 (5v)
OPA189 (36V)

Low Frequency Flicker noise corner, less than 0.1Hz
CS3002
ADA4522
OPA188
OPA180
LTC2057
LTC2058

My impression is some amplifier architectures do not have sufficient gain below 0.1Hz to fully correct for the flicker noise. "Well how good do we need to make it? Well, we need to have good performance for this 0.1Hz to 10Hz spec..."

This extra flicker noise is not a concern when buffering a Zener based voltage reference. The Zener noise is typically ten times the op amp noise.

However if you need stable or Very low noise performance below 0.1Hz don't assume just any chopper will remove the semiconductor flicker noise.

[chuckb]

I tested the amplifiers in a shielded test setup with the op-amps amplifying their own noise 10,000x. My first round of testing just used 1,000x gain. This was not enough gain to be above the HP35665A Digital Spectrum Analyzer (DSA) noise floor. With the higher gain the lowest noise op-amp is 4x above the DSA noise floor.
Gathering the complete noise spectrum of each op-amp took 4 different setups and roughly 30 hours. At high frequencies hundreds of sweeps could be averaged in a short time. At the lowest frequencies it took over an hour per sweep. It usually required 20-30 sweeps to reduce the trace noise to an acceptable level. The DSA results were manually transferred to a spreadsheet.

The output DC voltage was also recorded with a HP34420A 7.5 digit Nano-voltmeter. This data was captured and stored on a PC for later analysis. I usually had a 10 hour data record of samples every 3.3 seconds. This data produced a time record and a Modified Allan Deviation plot.

Attached are details of the testing procedure and the summaries for the first three op-amps.

[chuckb]

I have two strong broadcast transmitters within a few miles of my location. I have learned that I will have issues unless I am proactive with EMI minimization.

My son-in-law gave me 50 of these, very nice, 500g, enclosures. I dedicated one to power input filtering, one to the amplifiers under test and one to DC output isolation. Then I bolted them all together. See attached photos.

I used parts that I had on hand and I have not seen any EMI issues so it looks like the design is good enough.

[chuckb]

One last post for the night.

The stability for the air temperature in this part of the lab is usually 23 +-0.2 deg C. Humidity is held to 35-40% RH.

The air temperature directly next to the heavy metal enclosures was monitored with a HP2804 Quartz Thermometer. The temperature data was recorded with 0.001 deg C resolution. The temperature over 7 hours is included in the time record of the attached pdf. This temperature plot starts a little warm because I had just setup and handled the equipment.

I have not directly measured it but I'm sure the internal box temperature has less variation than the air temperature.

[MatteoX]

I've noticed that you have replaced the floppy drive on the HP35665A with what seems to be usb with some LED display. I am curious what did you use. I have recently got one and was thinking of modifying it.

[chuckb]

I've notied you have replaced the floppy drive on the HP35665A with what seems to be usb with some LED display. I am curious what did you use. I have recently got one and was thinking of modifying it.

I used the SFR1M44-U100 Enhanced version 3.5" 1.44MB USB SSD FLOPPY DRIVE EMULATOR GOTEK.
For my analyzer I have to use the high data rate 500kbps and format the USB to 720k (if I remember correctly). The USB needs to be in the drive at boot up and only one 720k drive is available. No matter how big the USB drive is. The DSA reads and writes to the USB and the PC reads and writes to the USB also.

I had to modify the 34 pin ribbon cable to prevent the 5vdc to the drive from shorting to ground. The analyzer was not damaged but the first cable was fried. HP used non standard pinouts to the drives. My HP3589A (same vintage) uses yet another different pinout.

[David Hess]

I am suspicious of the results because a baffle was not used.  This looks a lot like the different results we got with the same part number at different times because the lead frame material changed resulting in very different thermocouple effects with the same assembly.

[Echo88]

Indeed, maybe slight thermal changes within the circuit masquerades as 1/f-noise. Do you have resistors with very low TCR (PTF56 might be cost-effective here) and do you have access to a good heating-controller to stabilize the temp of the setup?
User blackdog posted a few good heater-design-suggestions here somewhere.
Maybe the OP-Amps that appear to have 1/f-noise just have a higher quiescent current, therefore get hotter and lead to more air convection, which results in thermal change? Havent checked the datasheets yet, but might do that tomorrow.

Nonetheless: Thanks for this measurement series. Its impressive how much energy you put in this interesting project. 

[chuckb]

I am suspicious of the results because a baffle was not used.  This looks a lot like the different results we got with the same part number at different times because the lead frame material changed resulting in very different thermocouple effects with the same assembly.

Is there a baffling technique that you would recommend? Is there a particular IC that you think I should retest?

I have not looked at all the chips but most list a maximum of 15nV / deg C for temp co. Of course if they used the wrong material in the lead frame then they could have any temp co. If there was a 2 deg C peak to peak temperature change, then you could have 30nVpp noise from thermocouple effects. This would convert to around 5nV rms at a low frequency. I am seeing less than 1 deg C change over a test series.

The gain setting resistors are 0805 SMD parts with 10ppm / deg C. With the offset voltages I have and the limited temperature change, I have computed less than 1nV of noise from the gain resistors.

Once upon a time I opened an old Zener reference, in a North Hills CC source, and it was full of sand. That's one way to kill convection currents and add thermal mass. However, I would rather not do that here.

[chuckb]

Nonetheless: Thanks for this measurement series. Its impressive how much energy you put in this interesting project. 

I'm retired now so I look for interesting projects to stay busy. Another fellow I worked with retired at the same time and he comes by 20-30 hours a week to help. I would have a lot more unfinished projects if it wasn't for Jim.

[chuckb]

Here are the results for the LTC2057 and the LTC2058. I updated the OPA188 file to add a 5 minute average of the data. This helps when looking for correlations with temperature.

I added a noise density chart with all the op amps. It's cluttered but all the info is in one spot. Also I attached the schematic of the test setup.

There are a few more choppers to test then I will do a good Bipolar and a good JFET.

Sometime later I will update this data with low frequency current noise. The PCB for this has been ordered.

[David Hess]

I am suspicious of the results because a baffle was not used.  This looks a lot like the different results we got with the same part number at different times because the lead frame material changed resulting in very different thermocouple effects with the same assembly.

Is there a baffling technique that you would recommend?

Any minimum volume low profile enclosure over the operational amplifier and its feedback network works.  I usually end up using card stock but fishpaper or a wax paper Dixie cup cut down to size works well also.  The problem is convective air currents.  I used to print out a 1:1 folding diagram with slots and tabs on card stock using an impact printer to make perfect little baffles.

I suspect the major improvement comes from preventing turbulent flow of the air as it rises.

Is there a particular IC that you think I should retest?

Any of the ones which showed high flicker noise.  Pick one and test it again with and without a baffle.

I have not looked at all the chips but most list a maximum of 15nV / deg C for temp co. Of course if they used the wrong material in the lead frame then they could have any temp co.

That specification excludes thermocouple effects and represents the potential performance under ideal conditions.  It does not include the thermocouple junctions between the leads and circuit.

If there was a 2 deg C peak to peak temperature change, then you could have 30nVpp noise from thermocouple effects. This would convert to around 5nV rms at a low frequency. I am seeing less than 1 deg C change over a test series.

External thermocouple effects will always ruin the low frequency noise performance of even a low noise precision bipolar part (0.1uV/C) with its higher albeit low 1/f noise never mind a chopper stabilized part.

The gain setting resistors are 0805 SMD parts with 10ppm / deg C. With the offset voltages I have and the limited temperature change, I have computed less than 1nV of noise from the gain resistors.

Resistor selection is of course important but you probably cannot go wrong with precision metal film or wire wound parts unless they are defective which sometimes happens.  A thermally symmetric layout with phantom parts to equalize the number of junctions in series with both inputs can help.

Once upon a time I opened an old Zener reference, in a North Hills CC source, and it was full of sand. That's one way to kill convection currents and add thermal mass. However, I would rather not do that here.

Sand would sure work and it would dampen vibration.  Potting in gel or epoxy also works.  I am not so sure about immersion in oil but I know that has been done also.  An air baffle is the simple if perhaps lower performance way and allows an easy comparison of performance with and without the baffle.  In production make the baffle conductive for free shielding at the expense of more common mode capacitance.

[chuckb]

Thanks for the guidance. I have some thin, flexible GAP PAD that I will cut and place over the sensitive parts.

[RandallMcRee]

I suggest testing the OPA140. I have used it for low-pass filters and it does seem to have good low-noise properties. The datasheet says

Very Low Offset Drift: 1 μV/°C Maximum
• Very Low Offset: 120 μV
• Low Input Bias Current: 10 pA Maximum
• Very Low 1/f Noise: 250 nVPP, 0.1 Hz to 10 Hz
• Low Noise: 5.1 nV/√Hz,

[RandallMcRee]

Chuck,
I am puzzled when comparing the graph of the ad4522-1 to the opa188. The opa188 seems to have sample measurements less than 1uV but greater than 950e-9 volts? The 4522-1 has samples around 500e-9. Yet both have 0.1 -10Hz noise in the 0.1-0.2uV range according to the datasheets. Can you shed some light on this? (Graphs from post #1 and #3). What is being measured in those graphs?

[chuckb]

The opamps are configured for a gain of 10,000. The output voltage passes through a 230 Hz single pole filter before leaving the enclosures. This DC voltage (around 10mV) is measured by an HP34420A 7.5 digit meter. The meter does not have any filters enabled. It's configured for 100 NPLC and offset compensation AutoZero is on. The meter integrates the input voltage for 1.67 seconds. Then it measures it's internal offset and zeros the meter. The meter supplies a reading every 3.3 seconds. The software takes that reading divides it by 10,000 and stores it in a spreadsheet along with time and temperature.

I don't know how the Multislope Integrator responds to higher frequency noise. This meter is taking a sample every 1.67 seconds and the response probably rolls off quickly above that frequency.

[Kleinstein]

The multi-slope, like other simple integrating converters should give an sin(x)/x type frequency response. So higher frequencies should be reasonably well suppressed, especially the 60 Hz and related as they fall to zeros of the filter.

The absolute position should be just the offset of the OPs - not directly connected to noise. So the 2 OPs would be at around 1 and 0.5 µV offset.

[David Hess]

I don't know how the Multislope Integrator responds to higher frequency noise. This meter is taking a sample every 1.67 seconds and the response probably rolls off quickly above that frequency.

Everything you need to know is here:

It's configured for 100 NPLC and offset compensation is on.

You are in the US so 60Hz power means the integration time is 1.67 seconds.  That produces a sin(x)/x frequency response with a first null at 0.6Hz and a -3dB bandwidth of like 0.29Hz (1) although I am not sure what significance that has with a non-linear roll-off.

Obviously this will not work for measuring noise up to 10Hz or even 1Hz.  When I have done this in the past, I had to settle for higher noise and lower resolution but faster measurements and then amplify before measurement accordingly because I was interested in noise up to 10Hz but I got excellent results.

(1) The 0.35 relationship between transition time and bandwidth becomes 0.442 or 0.468 depending on your perspective.

[chuckb]

I performed a test on all three channels to test for convection air currents interacting with dissimilar metal junctions on the PCB. I tested the LTC2057, LTC2058 and the OPA188. All three had previously shown low frequency flicker noise.

I cut some small sections of Berquist Gap Pad HC 5.0 that is 1mm thick. One thin slice covered the gain setting resistors. Then I placed two move layers over the resistors and the opamp covering as much adjacent space as I could. This material is slightly tacky and very compliant. It’s like a wet blanket. If there was any convection current happening before I estimate it is over 10 times less now.

With this modification the parts were retested. The low frequency noise of the OPA188 was within +- 0.5 dB of the previous open air test (inside the metal box). The LTC parts were monitored with the HP34420A and the peak to peak noise along with the Allan Deviation were within 10% of previous open air test.

With the Gap Pad, the LTC2058 offset shifted from -240nV to -160nV. I don't know why it shifted. Maybe it was from the temperature change. The LTC2057 offset voltage shifted less than 20nV. I did not check the OPA188 for an offset voltage shift.

Conclusion - The low frequency flicker noise that I am measuring is not caused by convection currents in the air interacting with dissimilar metal thermocouple junctions. That type of noise absolutely happens in other installations, it is just not happening here in this small, low power density enclosure.

The ADA4522 was tested in open air with +-2.5v and +-5v power. The noise amplitude and spectrum was the same as the internal power dissipation, and chip temperature rise, doubled. If there were convection currents the noise level would have changed.

Next testing will be for some conventional opamps just for comparison. Then I will build up a new pcb to start current noise measurements.

[chuckb]

I tested three normal amplifiers for low frequency noise. This was to provide a baseline for comparison to the Zero Drift amplifiers.
All three amplifiers were tested at +-5Vdc and a Gain of 10,000. I used the same 10 ohm and 100k ohm feedback resistors to set the gain as I used on previous tests.

The OPA227 (CORRECTED, WAS OPA277) bipolar amplifier was tested. It's low frequency flicker voltage almost reached the performance of the ADA4522 chopper!

I also tested two JFET amplifiers, the ADA4622 and the ADA4625. The ADA4625 has a very good 3nV / rt Hz white noise floor and very good flicker noise performance for a a JFET.


I made a new PCB for current noise testing last week. It came in Friday and Jim built it up for me. The first board was 2 layer with no flood fill of copper. This new PCB board has the same basic layout but I made it 4 layer with grounded copper flooding all the unused spaces. We will see if it makes a noticeable difference when I rerun a few of the voltage noise tests. In actual real world applications the Zero Drift amplifiers may not be in a heavy metal enclosure like I have and they may be subjected to faster temperature changes. A PCB with more copper to equalize the temperature may be useful for an environment with a faster ambient temperature change.

The current noise testing PCB started with a 1Meg ohm resistor for the current sensing resistor. This quickly proved to be too large, the op amp output was on the 15v rail. I changed it to a 100k resistor. That seems to work. I elected to only measure the current noise in the positive input of the opamp. When other people (and manufacturers) test the current noise they test both inputs at once. For me, 90% of my circuits will have a low impedance in the negative input and a higher impedance in the positive lead so that is what I tested. For this test my negative input is connected to the 10 ohm feedback resistor. Does anyone see a problem with this approach?

I have provisions to test the impact of 20pf and 2000pf across the 100k resistor. Reference the attached schematic.

When I first powered up the pcb it worked fine up to +-10V then the negative rail started drawing 1 amp (PS limit) and the voltage fell to 3V. Later I could take it up to 13V before it started drawing 1 amp. Well, when you put 5 solid tantalum capacitors in backwards on the negative rail, they will do that. I need to fix my engineering drawing.

The low frequency current noise testing has started and the results will be provided soon.

[Kleinstein]

I see no problem with testing only the current at the non inverting input.  The current noise can be correlated between the inputs. So for a full test one would need a test with resistors at both inputs too.

Because of the correlation in current noise the test with resistors at both inputs at the same time can not be replaced with tests with the resistors one at a time.  It's kind of measuring different things.

I also prefer the current noise for just 1 input over the date for the uncorrelated part only.

Wether the 100 K resistor is right for the test likely depends on the OP to test - some might need an even lower resistance and others might want the 1 M back. For the low voltage noise types the 100 K could well be sufficient. Low bias AZ OPs (e.g. AD8551) might want a higher resistance.

[chuckb]

Current noise testing of Zero Drift amplifiers continues. I will be traveling so this will not get updated for awhile.

Attached are some low frequency current noise spectrums for the ADA4528, the OPA189, the ADA4522 and the LTC2058. These were all with no decoupling capacitor across the 100k current sensing resistor. The spectrums are about 3 dB better with a decoupling capacitor. Charts to follow when I have time.

This is the current noise of just the non-inverting input. The inverting input has 10 ohms to ground. With the 100k current sense resistor, the current noise spectrum is at least 10dB above the voltage noise spectrum.

I started current noise testing with a 100k resistor as the current shunt. The next step was to add some filter capacitance across the resistor and note the change. The change was larger than I expected. Take the ADA4528 for example. The offset voltage was -1uV. When I added just a 100k resistor the offset raised to -20uV. Well that's the bias current... Then I added a 2nF COG cap across the 100k and the offset dropped to -6.5uV.

My current noise results for the ADA4528 do not match Hoyland's. Hoyland has a flicker noise rise starting at 0.1 Hz. In my testing the rise is very slight at the lowest frequencies. I will have to look into that. Maybe it's because I'm just looking at the current on the non-inverting input.

[EmmanuelFaure]

Nice PCBs! Where do you get them manufactured?

[chuckb]

ExpressPCB has a quick turn small sized 3 board deal for about $100. You use their pcb software and you don't have to mess with gerbers, drill files etc.

[chuckb]

Thanks for the work combining those graphs. Making a combined chart like that is on my to-do list.

[chuckb]

Here are the combined results of the Modified Allan Deviation analysis for the OpAmps.

[Inverted18650]

Nonetheless: Thanks for this measurement series. Its impressive how much energy you put in this interesting project. 

I'm retired now so I look for interesting projects to stay busy. Another fellow I worked with retired at the same time and he comes by 20-30 hours a week to help. I would have a lot more unfinished projects if it wasn't for Jim.

thanks chuckb and jim, great project.

[chuckb]

The third version of the pcb has been built and data collection is happening now. The components were rearranged to minimize trace length between the Op-amps under test and the DIP switch that changes the configuration. This is a 4 layer PCB. Reference the attached photos

Three changes were made, just to make things better.
1. Two of the regulators were moved to a different corner of the pcb to spread out their heat.
2. A low frequency ferrite core was added to the input power line to help with 1 MHz interference.
3. Each power pin to each Op-amp now has a 100 ohm series resistor and a parallel combination of a 100ufd Tant Polymer cap and a 0.1ufd X7R cap for filtering. This filtering starts working (-3dB) at 20Hz.

Jim added heaters to the three metal enclosures so I can control the housing temperature better. I will also be able to test how some parameters change with temperature. I plan to test at 30 and 40 deg C. The heaters have 100k thermistors on the same PCB for quick heater temperature feedback. I monitor this resistance with a K2015 (same as K2000). I also added a TI TMP235 bandgap temperature sensor to the PCB. The ground pin of the TMP235 is soldered directly to a via connected to the three ground planes. The sensor was then covered with hot glue for mechanical strength and isolation from air drafts. This provides 10mV /K to another K2015 DVM. The thermistor on the heater is in the inner temperature control loop. Because of the tight thermal coupling the heater temperature is tightly controlled. I also added a slow outer integration loop with +-0.1 deg C authority that corrects if the PCB temperature is off of target. An insulating cover slows down the rate of temperature change to help the thermal control works better.

So far the TMP235 on the PCB and the control loops hold the PCB indicated temperature to +-0.02 deg C. The TMP235 is powered from 2.5v regulated supply and it draws 9uA so the temperature rise is less than 0.01C. This low current also makes it a noisy temperature sensor so I may change it in the future.

I added more selections for capacitance across the 100k shunt resistor that measures Bias current. Reference the attached schematic.

I also added a selectable 100k shunt resistor in the non-inverting input of the Op-amp. This allows testing similar to how other researches have tested the low frequency current noise.

I am currently checking the ADA4528 (5V), OPA189 (30V), ADA4522(30V), and the LTC2058(30V) Op-amps. So far the noise spectrums are within 1dB of previous testing. The gain is the same at 10,000 (80db). For some tests I am collecting data on one Op-amp for 2 days. These are not quick tests. A 2 channel spectrum analyzer and two K2015 DVMs are used to collect the data.

Future testing -
Low bias current choppers.

I have the OPA2189 (dual OPA189) Op-amps for testing. These are not stocked in Mouser or Digikey yet. I had to get them from the factory web site.

I also have an engineering sample of a very old Maxim MAX426 Super Op-amp from the early 90s. It has not been powered for close to 30 years. This Op-amp was way ahead of it's time but it never made it into production. This device used internal 16 bit DACs to store the offset voltage instead of caps. Caps take up a lot of space on the die. There is a note about this chip in AoE3 but I could not find it. I might be mentioned in the 1st or the 2nd book as the best Op-amp available...

[serg-el]

Max426 datasheet 

[SilverSolder]

That's pretty impressive specs for the 1990s!

[serg-el]

Max426 datasheet part 2 

[chuckb]

This data collection takes a while when you have two temperatures and two different current measurement techniques to check. Along with different capacitance values for filtering. I'm about halfway through the testing with these 4 Op-amps. I'll generate some plots tomorrow for the data I have collected so far.

The MAX426 is an interesting chip for the electronic historians. Attached is some more info about it.

[chuckb]

The preliminary Datasheet for the MAX426.

[chuckb]

Here is how Cirrus built a Zero Drift Op-amp with 300dB gain at 1 Hz, the CS3001. Also attached is some chopper history.

[serg-el]

First mention MAX425/426.
And datashit in better quality 

[Echo88]

Interesting datasheet and autozero-concept, thanks! Its a bit like the OP mentioned in AoE 3 (dont remember the name) which eliminates its own Offsetvoltage once when its powered up.

[chuckb]

When I went to the Rev C PCB I added temperature control to the three metal housings.

I started with the TMP235 sensor from TI because it has a very low operating current of 9ua. However it has a surprising amount of noise. I measured 0.030 K peak to peak. This sensor was better than no temperature control but I thought I could find a sensor with lower noise.

After a week of testing Jim changed the Chopper PCB temperature sensor to a 300ua LM335. He kept the sensor ground lead as short as possible for best heat conduction to the three ground planes. I should have also glued the TO-92 sensor down to the PCB. That will be the next revision.

The LM335 sensor is much quieter than the TMP235; it had less than 0.001 K peak to peak noise. The indicated temperature was within 0.5 K of the TMP235 sensor. Note: I used a K2015 DVM with a moving average of 20 samples for both sensors. The TMP235 used the 1V range and the LM335 uses the 10V range.

The normal AC coupling of the HP35665A Digital Spectrum Analyzer works down to 1 Hz. I needed AC coupling down to 0.001 Hz so I built a custom coupling capacitor. I used eight of the KEMET Polyester 47 ufd 63 V, R60DR54705050K film caps. The 400ufd was 0.5dB down at 0.001 Hz. For faster settling I added a switch for selectable low frequency limits of 0.1 and 0.01 Hz. See photo and schematic. I use the 0.1 Hz position for quick charging and settling when a configuration changes.

I completed current noise testing for the ADA4528 and the OPA189. The first test (labeled Single) was with a 100 kohm sense resistor on just the non-inverting input. A 111 nF C0G was across the resistor to average current spikes. The second test (labeled DIFF) kept the first resistor and cap but added a second resistor and cap in series with the inverting input and the feedback network. The two current curves are on the attached plots.
The 111 nf causes the rolloff of the blue trace at 10 Hz. When a 10ufd is placed across the 100k the current noise roll off happens at 0.1 Hz.

I plotted the Voltage noise of the ADA4522-1 and the OPA189 at different power supply voltages. All power supplies were symetrical. For example the
30 V curve is configured as +-15V power. The    ADA4522-1   ADA4522-2 picks up a little low frequency noise at 30 V. The chip has    6   12 times the power dissipation at 30 V compared to 5 V. It's just a sample of one also.
   

[SilverSolder]

Can we be sure that the HP35665A  is as quiet at those low frequencies as the very high quality parts under test?

[chuckb]

Can we be sure that the HP35665A  is as quiet at those low frequencies as the very high quality parts under test?

That's a great question. See the attached file.

I first started evaluating the amplifiers with a gain of 60 dB. Then I saw that all three amplifiers had the same low frequency noise. So I increased the gain to 80dB. I only use metal thin film resistors for this low level work. The 10 and 100,000 ohm resistors in the feedback network are 5  10 ppm/ deg C, 0.1% accuracy parts. So the gain is stable. The temperature is also stable to minimize thermocouple and potential issues with op amp offset voltage drift.

The downside of using that much gain is that it limits the high frequency performance of the op amps because of GBWP limitations. I collect data out to 1KHz but because of two different high frequency roll off issues I only report performance out to 100 Hz. The OPA189 with a 14MHz GBWP has much better performance at 1 kHz than other op amps with a 1 or 3 MHz GBWP.

Part of this research is to decide which chopper to use as a preamp for my DSA. Just for fun, I attached is a photo of the 1.5F coupling cap for that preamp. There are a lot of 22,000 ufd caps in that box! It uses a triax connector to connect with the preamp. The switch adds a 10k resistor in series to limit charging current.

[SilverSolder]

Interesting, so it seems the HP35665A is more or less out of steam at 0.001Hz for this kind of testing? (Which is impressive, still!)

Impressive capacitor box...  I'm beginning to see why filtered references might not be completely practical, if done right!

[alanambrose]

Super interesting, many thanks for this work

[chuckb]

...
I plotted the Voltage noise of the ADA4522-1 and the OPA189 at different power supply voltages. All power supplies were symetrical. For example the
30 V curve is configured as +-15V power. The ADA4522-1 picks up a little low frequency noise at 30 V. The chip has 6 times the power dissipation at 30 V compared to 5 V. It's just a sample of one also.
 

Correction, the noise plot at 30V is using the dual version of the chip, the ADA4522-2. So the 30V curve has 12x the power dissipation of the 5V curve (single opamp).

[Inverted18650]

I’ve downloaded all the files. Thanks again, this is great work!

[chuckb]

I got busy with other projects and I just let the system collect data for 12 days. This is over 900,000 data points for the ADA4528 and the OPA189. At 1 day the Allan Deviation stability of some of the op amps is better than 1nV. That pretty amazing! See attached.

The zero input stability of the ADA4528 and the OPA189 are similar to the HP34420A Nanovoltmeter. All three have a flicker noise floor around 0.3nV at 1000 seconds.
Link to DVM stability results plot - https://www.eevblog.com/forum/testgear/nanovoltmeters-performance/?action=dlattach;attach=501782

I tested the ADA4522-2 for current noise with a single and dual 100k input resistors. See Attached. With a single 100k resistor on the non inverting input I recorded the current noise with a 0.1 uF C0G and a 10 uF Film capacitor across the resistor. During differential current noise testing both 100k resistors were bypassed with 0.1 uF C0G capacitors. The 100k resistors have 0.1% tolerance.

The testing continues...

[chuckb]

Just wanted to note that I corrected this old post. I updated the name on the file also.

I tested three normal amplifiers for low frequency noise. This was to provide a baseline for comparison to the Zero Drift amplifiers.
All three amplifiers were tested at +-5Vdc and a Gain of 10,000. I used the same 10 ohm and 100k ohm feedback resistors to set the gain as I used on previous tests.

The OPA227 (CORRECTED, WAS OPA277) bipolar amplifier was tested. It's low frequency flicker voltage almost reached the performance of the ADA4522 chopper!

...blah, blah, blah...

[3roomlab]

this 2017 cern article says the new adc design uses 4522
I wonder why not 4528 
https://indico.cern.ch/event/725164/contributions/2989640/attachments/1642142/2623032/new_ADC.pdf

[RandallMcRee]

this 2017 cern article says the new adc design uses 4522
I wonder why not 4528 
https://indico.cern.ch/event/725164/contributions/2989640/attachments/1642142/2623032/new_ADC.pdf

The 4528 is a +-2.5v opamp. The 4522 is +-18. Their adc is +-13volts or so.

[chuckb]

Thermal performance summary.
Reference attached pdf.

I did not want the ambient temperature reading influenced by the heated enclosure so the ambient temperature was measured 12" away from the enclosures and 6" above. An HP2804A Quartz Thermometer monitored the ambient temperature and reported it with 0.001C resolution.

Three heaters with integral 0402 sized 100k ohm thermistors monitored the heater temperature. The thermistors were used to form an inner control loop for the heater temperature. The heater temperature was controlled by a PI controller with local, fast thermistor temperature feedback.

The PCB temperature monitoring used an LM335 sensor and a Keithley 2015 6.5 digit DVM. The LM335 provides 10 mV per deg K. At the 30 deg C target temp the sensor provided 3.03V to the K2015 DVM. The DVM was on the 10V range and it used a 40 sample moving average filter (about 40 seconds) to minimize DVM induced noise. Reference the lowest trace on the attached pdf for the DVM noise level with a shorted input on the 10V range. The DVM noise is mostly below 10u deg C.

The three heavy, sealed, metal enclosures (1.5kg total) added several minutes of time delay between external temp changes and surface of the pcb. The LM335 sensor was used to bias the target temp of the external heater controller. The LM335 PI controller was very slow and it had a very limited +- 0.2 C range. There is an offset between the sensor temp and the pcb temp but I'm considering that a constant.

Overall, the temperature control system worked well because the indicated pcb temp stabilty was better than 0.2 mC.

NOTE: The graph indicates a long term (3 days) temperature stability of 20 uC. That may not be reality because the control LM335 is also the monitor LM335. If the sensor drifted I would not be able to tell.

Does anyone have experience with LM335 temp sensor drift at 30 deg C?

[Andreas]

Does anyone have experience with LM335 temp sensor drift at 30 deg C?

Not directly drift but the self heating with 1mA current is quite high (around 0.8 deg C in still air)
I mentioned this when using as outdoor sensor.
Each time when the wind was blowing I had a jump in temperature.

I reduced this jumps by switching on the LM334 current source only during measurement.

with best regards

Andreas

[Edwin G. Pettis]

Regarding the CERN ADC paper, the claim that all of the 1/f noise is due to the LTZ1000 is wrong, those Vishay resistors also generate 1/f noise, if they knew what they were talking about.....

[chuckb]

I tested two of the LT1028A Ultralow Noise Bipolar Op Amps. The power supplies were +- 5Vdc and the chips were in a plastic DIP package in a machined pin gold plated socket. The "A" version of the Op Amp has been tested for better DC performance and lower noise. The SMD version of the Op Amp is not available in the "A" version.

The LT1128A has slightly less speed (20 MHz vs 75 MHz) but it is stable for a +1 buffer application. The voltage noise and current noise should be similar to the LT1028A. There is also a version just for audio, the LT1115. It has relaxed DC specs.

I first tested the Voltage noise with the gain at 80 dB and the input grounded. The results are attached.  I was surprised how well the noise curves matched. The offset voltage was less than 10uV for both Op Amps. The temperature coefficient was +0.3uV / deg C between 31 and 41 deg C.
 
Then I tested the Current noise of both Op Amps with a 1000 ohm shunt on just the non-inverting input. The bias current was around 30na. The final test was with a 1000ufd Aluminum Electrolytic cap connected across the 1000 ohm current shunt. This lowered the current noise down to the level of the voltage noise for frequencies higher than 10Hz.

The next results will be for the MAX426. It has several operating modes so it takes longer to gather the data.

[chuckb]

The first three LT1028A Op Amps that I tested were manufactured in 2000-2001. So I picked up another one with 1835 date code (35th week of 2018). The voltage noise was the same as the other three. Looks like their process is not changing over the years.

I wanted to see where the LT1028A flicker noise floor changed to white noise. So I did an extra test at a much higher sampling speed of 20 Hz. As you can see in the second graph if you are averaging the signal for longer than 2 seconds the OPA189 or ADA4528 have a lower noise than the LT1028A.

I tested the OPA2189 (Dual version of the OPA189) at 30V in a SOIC-8 package. The voltage and current noise was the same as the single version.

[maxwell3e10]

It looks like OPA2189 is an all-around winner, with a higher voltage range and a little smaller input current noise than ADA4528.  I was also surprised that it has  a much larger slew rate and gain bandwidth than other zero-drift op-amps. So got some and tested them briefly:
https://www.eevblog.com/forum/projects/what-is-your-favorite-most-versatile-op-amp/msg2457006/#msg2457006
I was only able to find one peculiar behavior at 67.5 kHz that is probably associated with the chopper.

[alex-sh]

Did you test OPA388 or ADA4528?

[chuckb]

Did you test OPA388 or ADA4528?

I have not tested the OPA388 but it looks like a very modern design (2016).
The ADA4528 was one of the first designs tested. The results can be found in the first post in this topic.

[chuckb]

Here are two more low noise Op Amps in this never ending story.
The ADA4523 is a new very low noise chopper from Analog Devices. The datasheet was released April 20, 2020. I had to get samples from the factory because the distributor did not have any stock yet.

https://www.eevblog.com/forum/metrology/new-chopper-opamp-from-ad-ada4523-1/msg3076193/#msg3076193

The LT6018 is a very low noise bipolar Op Amp. It was introduced in 2017 time frame. It has a 33V power supply limit and high bias currents. It has been on my list of Op Amps to test.

Today I just have the voltage noise spectrum for you. I collected current noise data also and will present that when I get some time.
I used +-5V power for both Op Amps. The enclosure was not temperature controlled this time. However the lab had less than +-0.3 deg C variation and the heavy insulated enclosures provided several hours of thermal lag. Each amplifier was configured for a gain of 80dB using 10 ohm and 100k ohm metal film SMD resistors. The amplifier outputs were measured with an HP35665A Low Frequency Spectrum Analyzer.

The new ADA4523 chopper has 20% less flicker noise than the OPA189. The bias current of my sample was 50pa.

The LT6018 is an exceptional bipolar opamp for white and flicker noise. It has roughly 1/3 the flicker noise of the LT1028A. The bias current of my sample was 61na.
More results to follow.

[Bud]

The vertical scale (or Y axis label) is incorrect.

[chuckb]

I corrected the Spectrum Analyzer model number and the label on the Y axis of the graph in my previous post.

[exe]

Wow, those modern auto-zero opamps seems to be quite good. I'm talking about opa189.

[chuckb]

Attached are the bias current noise plots for the ADA4523 and the LT6018.

The non-inverting input has the indicated resistance and capacitance to ground. The amplifier is configured for a gain of 80dB with 10 ohms on the Inverting input.

The ADA4523 bias current noise is much improved compared the the previous ADA4522.

The LT6018 bias current noise is very similar to the LT1028A noise spectrum. For this test I measured unbalanced current noise. The input resistances are not balanced (1 kohm and 10ohm). My results match the datasheet unbalanced data within 10%. The datasheet shows a 6x current noise reduction with balanced input resistances in the flicker noise region, I have no reason to doubt that. I will not be testing that.

[macaba]

Thank you for all your hard work. The ADA4523 is an impressive device.

I have seen references to putting a small capacitor across the inputs of zero drift devices to reduce noise, I would be interested to see an investigation into this for the leaders (ADA4523, OPA189).

[chuckb]

I noticed early on that the bias current appeared to change with different input capacitance.
I decided to plot offset voltage vs input impedance for the ADA4523. The Op Amp has a 330kHz input switching freq so that's what I used to compute input Z vs offset. There are usually input current spikes at 10x this freq also.

The DC input resistance is 100 kohms, that causes the 5uV offset from the 50pa of input bias current. The extra input capacitance is added to reduce the input impedance at the switching frequency. This seems to stabilize the operation of the Op Amp. Reference the attached graph. In this setup the inverting input has 10 ohms to ground so it is not disturbed by the input switching spikes.

The OPA189 needed the full 100nF to stabilize it's operation also.
The OPA187 (Low bias current, low power) only needed 1nF to achieve stable voltage.
 
There is much more to research and experiment on this topic.

[SilverSolder]

I noticed early on that the bias current appeared to change with different input capacitance.
I decided to plot offset voltage vs input impedance for the ADA4523. The Op Amp has a 330kHz input switching freq so that's what I used to compute input Z vs offset. There are usually input current spikes at 10x this freq also.

The DC input resistance is 100 kohms, that causes the 5uV offset from the 50pa of input bias current. The extra input capacitance is added to reduce the input impedance at the switching frequency. This seems to stabilize the operation of the Op Amp. Reference the attached graph. In this setup the inverting input has 10 ohms to ground so it is not disturbed by the input switching spikes.

The OPA189 needed the full 100nF to stabilize it's operation also.
The OPA187 (Low bias current, low power) only needed 1nF to achieve stable voltage.
 
There is much more to research and experiment on this topic.

Are you placing the 100nF between non-inverting and ground, after a 100K resistor?  (like a ~15Hz filter?)

[Kleinstein]

The impedance / capacitance at the input effects the charge injection of the chopper switches. The switching should be really fast (e.g. more like 10 ns range) and thus really high frequencies (e.g. 100 MHz to a few GHz) that matter.
The effect of capacitance on the charge injection at CMOS switches is known.
At the high frequencies it may be more than just the capacitance, but also ESR, ESL that can have an effect.

It is a little surprising the it takes so much capacitance to saturate the effect.
It could be interesting to have the same curve with a smaller resistor (e.g. 1 K), so that one also has more of the range where  R*C is less than some 3 µs.

[SilverSolder]

The LT6018 looks pretty amazing as well,  but it isn't exactly cheap...

[SilverSolder]

The impedance / capacitance at the input effects the charge injection of the chopper switches. The switching should be really fast (e.g. more like 10 ns range) and thus really high frequencies (e.g. 100 MHz to a few GHz) that matter.
The effect of capacitance on the charge injection at CMOS switches is known.
At the high frequencies it may be more than just the capacitance, but also ESR, ESL that can have an effect.

It is a little surprising the it takes so much capacitance to saturate the effect.
It could be interesting to have the same curve with a smaller resistor (e.g. 1 K), so that one also has more of the range where  R*C is less than some 3 µs.

If the impedance seen by the + and - inputs are different, perhaps the charge injection behaves differently between the two inputs, and that difference ends up being amplified by the gain of the op amp?  Perhaps there are slight differences between the internal switches on the two inputs as well, which also end up being amplified?  (Just thinking out loud...)

[chuckb]

I noticed early on that the bias current appeared to change with different input capacitance.
I decided to plot offset voltage vs input impedance for the ADA4523. The Op Amp has a 330kHz input switching freq so that's what I used to compute input Z vs offset. There are usually input current spikes at 10x this freq also.

The DC input resistance is 100 kohms, that causes the 5uV offset from the 50pa of input bias current. The extra input capacitance is added to reduce the input impedance at the switching frequency. This seems to stabilize the operation of the Op Amp. Reference the attached graph. In this setup the inverting input has 10 ohms to ground so it is not disturbed by the input switching spikes.

The OPA189 needed the full 100nF to stabilize it's operation also.
The OPA187 (Low bias current, low power) only needed 1nF to achieve stable voltage.
 
There is much more to research and experiment on this topic.

Are you placing the 100nF between non-inverting and ground, after a 100K resistor?  (like a ~15Hz filter?)

That is correct. Ref this link for the schematic and pictures of the PCB layout with the DIP switches selecting the capacitance.

https://www.eevblog.com/forum/metrology/low-frequency-noise-of-zero-drift-amplifiers/msg2206434/#msg2206434

[souldevelop]

I used ADA4523-1 to make a low noise amplifier which is ULNLFA-100, and its performance is as follows:

Input noise voltage: 40nVp-p in the range of 0.1 Hz to 10 Hz
Spectral density voltage: 2nV/√Hz at 1kHz
Maximum voltage gain: 1000000 or 120dB
Gain control: 4 levels (60dB 80dB 100dB 120dB)
Low-pass filter: 3 gears (<10Hz, <100Hz, Full)
Power supply: 2 pieces of 103450 4.2V 2000mAH lithium battery
Power supply voltage: ±4.2Vdc, maximum power supply voltage ±15Vdc
Charging interface: miniUSB 5V 300mA
External dimensions: (length/width/height) 13CM/7.65CM/3.5CM
Total weight: 308 grams (including battery)

ULNLFA-100 uses 8 pieces of ADA4523-1 in parallel to achieve low noise performance of 40nVpp.
The total cost of making it is about 140$. I still have some remaining machines here but the number is not much. If you need it, plz contact me.

[Andreas]

Input noise voltage: 40nVp-p in the range of 0.1 Hz to 10 Hz

Hello,

how is the input impedance?

Is there any protection of the source against excessive charge current of the input capacitor?

with best regards

Andreas

[souldevelop]

1.input impedance is about 4.7k.
2.Inside are two 4.2v lithium batteries, which are connected in series when charging,
   the maximum charging current protection limit is 500ma, and the working current limit is 100ma.

[TiN]

Andreas was asking about input signal current pulse, when you connect e.g. 10V source to preamp, while input AC capacitor is not charged (at 0V).

[souldevelop]

Sorry, I haven't considered this issue yet. Any good suggestions?

[macaba]

Your board layout looks nice but potentially a few issues:

- 4.7k resistor is an unusually large value that generates noise.
- Input current noise of ADA4523 is high... 8 in parallel? 
- Your Rigol scope traces have AC-coupling enabled (which is a HPF of something like 5Hz/10Hz)

[Kleinstein]

For AC coupling the resistor to ground adds to the noise, but not in the normal sense. It adds to the current noise and thus a smaller resistor adds more noise. So the 4.7 K to ground are more an the low side, but it depends on the amplifier.

The input current noise of the AZ OPs is relatively high and with more units in parallel this adds up. So the amplifier is only good for a low impedance source and a large input cap is needed.  The current noise of the amplifier together with the capacitor set the useful lower frequency limit. For the lowest end, it may have actually be better to have only 1 amplifier active.

Noise wise it would make sense to set the lower frequency limit not by the RC at the input, but have a larger resistor and set the lower frequency limit later (e.g. in software).

[souldevelop]

Thank you for your reply. Indeed, as you mentioned, I experimented with more than 8 4523-1. The noise reduction is not so obvious. I will publish the circuit diagram as follows to see if I can improve it.

[souldevelop]

thk! There will be a small offset when using DC coupling. I did this to facilitate observation.

[Electrole]

The test suite by chuckb of the various Zero-Drift Amplifiers is highly interesting.
Any ideas how a chopper amplifier like the ICL7650S from Renesas performs in comparison??
The data sheets states a typical "Change in Input Offset with Time" of 100 nV/sqrt(month), a typical input noise voltage of 2 µVpp from DC to 10 Hz, and a typical input current noise density of 10 fA/sqrt(Hz) at 10 Hz, but I have not found any noise density specifications as a function of the frequency.
BR

[chuckb]

Check the data sheet for the Texas Instrument ICL7652. It looks like a similar architecture with external capacitors. It has 90nV / rt Hz Voltage noise density and 4fa / rt Hz Current noise density at low frequencies.

The newest chopper stabilized amplifiers have roughly 20x less voltage noise with a higher bias current. I have not researched low bias current choppers.

[Kleinstein]

The ICL7650 is an old classic auto zero OP.  AFAIR I saw some noise spectrums and it follows the usual way for an AZ OP:  relative high noise for low frequenices and lower noise (like 1/2 - 1/3)  from some 5 kHz on.
The noise current looks good, but maybe too good - some of the very low numbres for the current noise are not real - more like an lower limit derived from the bias.

I just tested the microchip  MCP6V51 , which has good voltage noise specs at some 11 nV/sqrt(Hz).  I had not expected to get current noise as low as the data-seet number, but I had some hope i might be better than the otherwise comparable LTC2057. However the test showed way more current noise - more like 700-800fA/sqrt(Hz)  - so something like 5 x the LTC2057.

[Electrole]

The Texas Instruments ICL7652 looks quite much like the TLC2652A / TLC2652, and the noise density at low frequencies is certainly not impressive. I suspect the ICL7650s would show similar performance, given the age and topology of the device, but I do not know. For this reason, I was thinking about measuring the ICL7650S, but I'm finding the OPA189ID increasingly interesting. The wide supply voltage span is also a bonus, which makes it easier to use in some designs. One thing I have in mind is a buffer for a 10 V reference, for which the OPA189ID seems quite appropriate...
BR

[RandallMcRee]

Have you looked at the ADA4523-1?
https://www.analog.com/en/products/ada4523-1.html

For my 7v-10v transfer a few years ago I used the OPA189, today I would use the above.

[Electrole]

I made a parametric search at the AD web site the other day, but I somehow did not spot the ADA4523-1. Thanks!
Looking at the specifications the ADA4523-1 is definitely a device that targets some of the same applications as the OPA189, such as a voltage reference buffer. The two devices also seem to have a comparable price tag.
May I ask why you came to prefer the ADA4523-1 over the OPA189?
BR

[RandallMcRee]

I made a parametric search at the AD web site the other day, but I somehow did not spot the ADA4523-1. Thanks!
Looking at the specifications the ADA4523-1 is definitely a device that targets some of the same applications as the OPA189, such as a voltage reference buffer. The two devices also seem to have a comparable price tag.
May I ask why you came to prefer the ADA4523-1 over the OPA189?
BR

I wanted the lowest possible flicker noise. The input is several (eight) averaged LTZ1000 circuits so *their* noise is approx. 0.05 uV. So that opamp, in that position, is the major noise contributor.

[macaba]

May I ask why you came to prefer the ADA4523-1 over the OPA189?

I'm not Randall, but as someone who picked the ADA4523-1 for use in ultra precision metrology circuits, I have some input to make.

Have a look at Figure 7-11 of OPA189 datasheet, and Figure 14 of ADA4523-1 datasheet.

Essentially; the ADA4523-1 has input bias compensation that makes it extremely well behaved over temp range. This makes up for the fact that it has 1x - 2x the average input bias current of other devices. In practice, this means that if you have a 10k resistor on the ADA4523 input, the input bias current will only cause a 0.75uV drop over that resistor, and only vary by 0.25uV over the temp range from 25 to 100 degrees.

[Kleinstein]

I would not count too much on the figues give an the input current. Most curves are for typical units. The input current of chopper OPs in general does not change much with temperature, but it can be different a lot between individal units.

The AD4523 is very low current noise, but relatively high bias and high current noise.  There are very few referene that really need voltage noise that low.