THP210: FDA attenuator configuration stability analysis

Briquel Florian

Part Number: THP210
Other Parts Discussed in Thread: THS4541, TINA-TI

Hello,

I plan to use THP210 for a energy power metering application. Sensor connected to THP210 present a very high output impedance (sensor are capacitors divider for kV voltage application).

Then I should use very large value resistor for input and feedback. In the range of mega ohm.

The THP210 will be used to made a single ended to differential conversion with a negative gain of approx -4.4dB.

Rin resistors about 3.3mega ohm, Rfb about 2 mega ohm and a small capacitor of 20pF in // with Rfb.

I will use a specific layout to avoid parasitic stray capacitance, especially in the area of inputs of THP210, I don't know if I should take another consideration to achieve a such design, but when I read following statement on datasheet I try to get the TI training but can't find it:

High feedback resistor values (RF > 100 kΩ) interact with the amplifier input capacitance to create a zero in
the feedback network. Compensation must be added to account for potential source of instability; see the TI
Precision Labs FDA Stability Training for guidance on designing an appropriate compensation network.

I can't find FDA stability training on TI website, the only video I found is relative to gain and phase margin, no mention of large value feedback resistor and stability.

How we can define criterion for stability in terms of phase margin/gain margin in a attenuator configuration ?

Example (I use LTSpice simulator) Bode diagram of my circuit:

over 1 year ago

0 Michael Steffes over 1 year ago

Guru 48395 points

This comes up occasionally, here is a discussion of stability and and fixes,

https://www.planetanalog.com/are-inverting-attenuators-with-voltage-feedback-op-amps-inherently-unstable-and-why-might-you-want-to-apply-them-insight-11/

Also, in the THS4541 datasheet this is a discussion of designing attenuators for FDA's I put in there.

0 Briquel Florian over 1 year ago in reply to Michael Steffes

Prodigy 40 points

Hi Michael,

Thanks for sharing this resource. This make me confident about the understanding I have from TI Precision labs serie FAD stability, maybe you can confirm:

For attenuator configuration the way to go to be confident about stability is to look at NG instead of LG and to work with AOL bode diagrams instead of ACL.

Phase margin is the amount of phase resulting of subtraction of phase value just before first pole (AOL bode) and phase value at intersect point between NG line and and gain value still in AOL bode.

Ex:

Example above show a NG value of 2MR/3.3MR +1 = 4dB, I add L1 and L2 to plot AOL bode instead of ACL, I don't care of phase value near DC point, I took phase value before first pole of amp. This show me a phase value of 180-(93+14)=73C, do you agree ??

Best regards,

Florian

0 Michael Steffes over 1 year ago in reply to Briquel Florian

Guru 48395 points

You know I can run that LT plot in LTSpice if you attached it, hard to see and respond

0 Briquel Florian over 1 year ago in reply to Michael Steffes

Prodigy 40 points

Off course, here the ltspice I use.
LPVT_stability_spice.zip LPVT_stability_spice.zip

0 Michael Steffes over 1 year ago in reply to Briquel Florian

Guru 48395 points

Thanks, I got real busy today, will take a look at tomorrow.

+1 Michael Steffes over 1 year ago in reply to Michael Steffes

Guru 48395 points

Ok Briquel, I redrew you schematic in more classic LG analysis where I opened the loop at DC with a direct L feedback 1GH, then tapped off the differential feedback at the summing nodes with the input 1pF added and sensed the gain and phase around the loop injecting a differential test signal at the FDA inputs through those 99.97MF caps - the 1mOhm series helps with numerical chatter. Lot going on in the phase, but yes, in this sim (if I have the phase polarity right) you have 64deg phase margin, you might check that with a closed loop sim of the output spot noise to high F looking for sharp spikes.

Well it is not letting me upload this modified .asc, you can contact me offline for that if you want, or perhaps the TI guys could redo this schematic in TINA.

0 Luis Chioye over 1 year ago in reply to Michael Steffes

TI__Guru 55466 points

Hello Michael,

Thank you. Using the same open-loop THP210 circuit on the post above in TINA-TI, yields a phase margin of 64-degrees.

TINA file: THP210_stability_5-1-24_method-b.TSC

The THP210 has a input common mode voltage (VICM) range requirement where the voltage right at the FDA input terminals requires at least 1V headroom from the VS+ and VS- voltage supplies. The attenuator circuit above will accommodate a single ended input signal up to VIN= ±5Vp without issue, but will be outside of input common mode range at larger amplitude input signals, since the device requires a 1V input common-mode headroom above GND at the input terminals. Let me know if you have questions.

Thank you and Regards,

Luis

0 Briquel Florian over 1 year ago in reply to Luis Chioye

Prodigy 40 points

Hi Luis,

Indeed the THP210 has a 1V headroom for input. Our sensor goes up to +/-7.6V but has an output impedance about 50kR so we design our circuitry to get 1V headroom for +/-7.6V p.

Attached our ltspice files.

best Regards,THP210.LIB

0 Luis Chioye over 1 year ago in reply to Briquel Florian

TI__Guru 55466 points

HI Briquel,

If the circuit design/signal chain design already scaled the sensor signal +/-7.6Vp to about +/-4.9Vp prior to the 47nF DC blocking capacitor at the THP210 attenuator circuit input, and confirmed the VICM has enough headroom in simulation, please disregard my comment. Otherwise, the input voltage divider, or THP210 attenuation may need to be adjusted. If you wish me to review the complete schematic, let me know, I can always contact you via direct email, if this is your preference. The post above includes a THP210.LIB model file attachment, but not the signal chain schematic showing the +/-7.6V sensor simulation.

Thank you very much,

Kind Regards,

Luis