Amplifiers forum - Recent Threads

AMC0381D: Layout Recommendation

Kazuki Itoh — Tue, 27 Oct 2026 08:33:50 GMT

Part Number: AMC0381D

Hi team,

Could you please provide information regarding the recommended layout for the AMC0381D? Why is the distance from the SNSP pin guard ring to the low-side NC pin not meeting the clearance requirements? What exactly is a "keep-out zone"? In the EVM PCB layout, the GND2 copper appears not to take the keep-out zone into account.

As the datasheet doesn't provide enough information to interpret, I'd appreciate more specific explanation.

Best regards,
Kazuki Itoh

RE: AMC0381D: Layout Recommendation

Jiri Panacek85 — Fri, 13 Nov 2026 10:47:25 GMT

Hi Kazuki,

in fact, I believe that most AMC038x systems use the PCB without the guard ring. It is OK to remove it if the pollution and parasitic resistance of the PCB is not a concern, or if the light accuracy degradation is acceptable.

Regarding the AMC0381D EVM, I confirm that this layout does not follow the keep-out zone.

For the DFJ variant www.ti.com/lit/ds/symlink/amc0381d-q1.pdf the customer has to consider a compromise between the guard ring or the clearance.
Without the guard ring, >8mm is possible

Kind regards, Jiri

RE: AMC23C10: AMC23C10 Trip threshold error

shuhei hayashi — Fri, 13 Nov 2026 07:02:29 GMT

Hi Jiri san,

Thank you for your answer.
I'm sorry but, please kindly tell me the following comments.

Add1) Would you kindkly tell me the factors that can cause the timing of an output signal to change, even when the input signal remains identical?
In a previous inquiry, I was told that output jitter and temperature drift are factors that can alter the timing of the output signal.
Is my understanding correct?
Or is it the case that the only factor causing variations in the detection voltage is minor fluctuations in the frequency of the AC input?

Add 2)Please kindly tell us the VHYS for AMC23C10.
Please see the 6.9 Electrical Characteristics.
VHYS is 25mV(Typ).Would you kindly tell us the MIN and MAX value for VHYS?
Or are there specified minimum and maximum values for VHY?
The customer is particularly concerned about whether the maximum value of VHY might deviate significantly - by a factor of two or more - from the standard value.

Best Regards,
Shuhei Hayashi,

AMC23C10: AMC23C10 Trip threshold error

shuhei hayashi — Mon, 26 Oct 2026 07:12:34 GMT

Part Number: AMC23C10

Hi All,
Would you kindly tell us the meaning of Trip threshold error under COMPARATOR in section 6.9 Electrical Characteristics of the AMC23C10 datasheet.
I understand it as follows. Please kindly tell me whether my understanding is correct.
1)INN=GND1=0V. Is it correct?
2) In the theoretical case, the following threshold formula applies.
VIN+=VINN+VHYS/2 = 0V +25mV/2= 12.5mV
VIN-=VINN+VHYS/2 = 0V - 25mV/2= -12.5mV
The "Trip threshold error" range is from -6mV to 6mV.Therefore, I believe the following formula applies.
VIN+ = 12.5mV ±6mV= 6.5mV to 18.5mV
VIN- = -12.5mV ±6mV= -18.5mV to -6.5mV
The above represents the variation of the AMC23C10 comparator itself.
In other words, VIN+ varies from 6.5mV to 18.5mV, and VIN- varies from -18.5mV to -6.5mV. Is this correct?

Best Regards,
Shuhei Hayashi,

AM62A7: Need more detail on this part number

Shubham Bawner — Thu, 12 Nov 2026 10:39:15 GMT

Part Number: AM62A7

Need more detail on this part number

RE: AM62A7: Need more detail on this part number

Nihar Potturu — Fri, 13 Nov 2026 06:05:53 GMT

Hi Shubham,

Which part number are you referring to? Did you go through the AM62A device data sheet?

https://www.ti.com/lit/ds/symlink/am62a7.pdf

Regards,

Nihar Potturu

AMC23C10: AMC23C10 Switching Characteristics

shuhei hayashi — Mon, 26 Oct 2026 07:15:21 GMT

Part Number: AMC23C10

Hi All,

The voltage VDD2 for each item in section 6.10 Switching Characteristics of the AMC23C10 datasheet is 3.3V.
Please kindly tell us the frequency response values when VDD = 5.0V.
In other words, would you kindly tell us the MIN (ns), TYP (ns), and MAX (ns) values for "PUSH-PULL OUTPUT" and "OPEN-DRAIN OUTPUT"?

Best Regards,
Shuhei Hayashi,

RE: AMC23C10: AMC23C10 Switching Characteristics

shuhei hayashi — Fri, 13 Nov 2026 06:01:00 GMT

Hi Jiri san,

Thank you so much for your answer.
Please tell me the following comment.
The rise time and fall time shall not vary more than +/-20%,
Would you kindly tell me clarify what this ±20% fluctuation range is based on?
In section 6.10, "Switching Characteristics," of the datasheet, a propagation delay (for both rising and falling edges) is specified with a typical value of 230 ns; a ±20% variation would therefore result in a range of 184 ns to 276 ns.
Meanwhile, the datasheet also specifies a maximum value of 320 ns.
To which of these specifications does the ±20% figure you provided refer?
1)Variation range relative to the standard propagation delay time of 230 ns(tpH/tpL)
2)Variation range for rise time and fall time(tr/tf)
3)Variation in switching characteristics when VDD2 is changed from 3.3V to 5V
Furthermore, regarding the propagation delay time when VDD2 = 5V, is it correct to assume that the values specified in the datasheet—a typical value of 230 ns and a maximum value of 320 ns—can be applied as is?

Best Regards,
Shuhei Hayashi,

OPA615: OPA615: Figure 48 Bipolar Peak Detector – Unexpected Behavior with 200 mV / 5 ns Pulses

CHE-MIN LIN — Wed, 11 Nov 2026 08:44:18 GMT

Part Number: OPA615

Hi TI Support Team,

We are evaluating the Fast Bipolar Peak Detector circuit shown in Figure 48 of the OPA615 datasheet for a Partial Discharge (PD) measurement application.

Our goal is to independently detect and hold the positive and negative peak amplitudes of low-amplitude, nanosecond-duration pulses.

We downloaded the original TINA-TI reference circuit provided by TI in the following E2E discussion:

https://e2e.ti.com/support/amplifiers-group/amplifiers/f/amplifiers-forum/986057/opa615-figure-48-in-datasheet-results-not-as-expected/3658421

We kept the original circuit topology, component values, and supply conditions unchanged. Only the input signal amplitude and pulse width were modified for our evaluation.

Simulation observations:
1. Original ±3 V periodic input: Both positive and negative peak detector outputs show the expected response.
2. Reduced input amplitude: When decreasing the input amplitude from ±3 V to ±1 V, ±500 mV, ±200 mV, ±100 mV, and ±50 mV, the negative output operating point and waveform change significantly. The output behavior is not proportional to the input amplitude.
3. Single +200 mV / 5 ns pulse: The positive output (+Vout) responds to the input pulse. However, the negative output (−Vout) also exhibits an unexpected voltage change.
4. Single −200 mV / 5 ns pulse: The negative output (−Vout) responds to the input pulse. However, the positive output (+Vout) also shows a transient response.
We would appreciate your technical advice on the following questions:
1. Is the original Figure 48 circuit suitable for independently detecting and holding positive and negative peaks of low-amplitude pulses (50–200 mV) with pulse widths as short as 5 ns?
2. What could cause the opposite-polarity output to respond when only a single positive or negative pulse is applied? Is this expected behavior of the circuit, or does it indicate a limitation?
3. Could the observed behavior be related to the OPA615/BUF602 operating conditions, circuit biasing, diode characteristics, the 27 pF hold capacitors, or limitations of the TINA-TI simulation models?
4. Are there any recommended circuit modifications, component value adjustments, or alternative TI reference designs to improve peak detection accuracy and minimize interaction between the positive and negative detection paths?
5. Could TI advise on the practical minimum detectable pulse amplitude and pulse width for this Figure 48 configuration, or recommend a suitable method to evaluate these limits?
We have attached the simulation waveforms for your reference.

Any recommendations from TI's applications engineering team would be greatly appreciated.

Thank you for your time and technical support.

Best regards,

Michael Lin

Power Diagnostic Service Co., Ltd.

---------------------------------------------
T +886 3 530 5588 Ext.329
F +886 3 538 7111
E zhe_min_lin@pdservice.com

RE: OPA615: OPA615: Figure 48 Bipolar Peak Detector – Unexpected Behavior with 200 mV / 5 ns Pulses

CHE-MIN LIN — Fri, 13 Nov 2026 05:16:55 GMT

Hi Amber,

Thank you for your clarification and technical guidance.

We will proceed with the OPA860 reference circuit evaluation and develop separate positive and negative peak detection paths as suggested.

We appreciate your support.

Best regards,

Michael Lin

Power Diagnostic Service Co., Ltd.

RE: OPA593: When the EN pin is turned off, a high-voltage pulse is present at the output.

stephen li — Fri, 13 Nov 2026 03:47:53 GMT

In my application, the OPA593 outputs a square wave signal, and there are requirements on the slew rate of the signal's transition edges. When I added a 1nF capacitor at the output, it indeed filtered out more than 80% of this pulse, but the output signal's slew rate could not meet the specification. In addition, my application sometimes outputs a high voltage of 24V as the high level, and sometimes only needs 5V, so a single TVS diode cannot solve the problem in all scenarios. If this problem cannot be avoided when using the OPA593 chip, are there alternative solutions? Does the OPA598 also have this problem?

OPA593: When the EN pin is turned off, a high-voltage pulse is present at the output.

stephen li — Thu, 12 Nov 2026 10:39:28 GMT

Part Number: OPA593

I am using the OPA593 as an output power amplifier, with VCC supplied at 30 V and a 100 pF capacitor connected from the output to GND. When the EN pin switches from high level to high level, an unexpected high-voltage pulse appears at the OUTPUT pin.

CH4=EN

CH2=OUT

TAS2120EVM: TAS2120 EVM can't work when 1st power up

Will Wang47 — Fri, 13 Nov 2026 03:13:48 GMT

Part Number: TAS2120EVM

Hi team,

I'm test TAS2120 EVM, while I find when load the saved project, every time, the 1st power up, the board can't work, and I find there is error. If I clear the error, the board still can't work to output voltage.

I have to reset at the motherboard side and then reload the configuration to the chip, it then could look like work as normal.

Could you help check why this board get this kind of issue?

Regards,

Will

LMH5401EVM: How to see input signal

ttd — Thu, 05 Nov 2026 18:39:17 GMT

Part Number: LMH5401EVM

Dear Technical Support Team,

I'd like to see input signal with oscilloscope.

When I input 100mVDC input from SG(50ohm) and probe input such as IN- SMA(1pin) or R1,

then I see about 380mV. However differential ouput is expected like 400mVpp(G=4, OUT+ =1.2V and OUT- =1.6V).

How do I see input signal correctly on EVM?

If you have any method for LMH5401, please let me know.

Is that because the LMH5401 is current feedback? If it used a voltage feedback High speed amp, would it be possible to see this using an oscilloscope probe?

■Configuration

single end input to differential output.

VS+ = 3.9V

VS- = -1.1V

VCM = 1.4V for ADC3669

Best Regards,

ttd

RE: LMH5401EVM: How to see input signal

ttd — Thu, 12 Nov 2026 22:06:58 GMT

Hi Team,

Do you have any update so far.

If you have EVM, could you check it?

Best Regards,

ttd

RE: INA200: INA200 RESET threshold min max

Mallika Senthil — Thu, 12 Nov 2026 21:09:11 GMT

Hi Yunjing,

Unfortunately we do not publish min or max limits for the RESET Threshold, I would recommend sizing the divider so the LOW level is well below and the HIGH level well above 1.1 V, rather than targeting it.

[quote userid="408808" url="~/support/amplifiers-group/amplifiers/f/amplifiers-forum/1697317/ina200-ina200-reset-threshold-min-max"]What is the voltage here for "power supply is below the voltage show here"? What is this "voltage show here" spec?[/quote]

If the power supply is below 1.5V, then comparator will be in reset mode.

Best,
Mallika Senthil

INA200: INA200 RESET threshold min max

Yunjing Wang1 — Thu, 12 Nov 2026 10:29:03 GMT

Part Number: INA200

Hi team,

We need more data to design INA200 RESET circuit, use divider to provide RESET.

May I know the min max value of INA200 RESET threshold?

What is the voltage here for "power supply is below the voltage show here"? What is this "voltage show here" spec?

Thank you!

Yunjing

JFE150: JFE150 – Transition Frequency (fT) and RF Performance Data

Kalle Kersten — Thu, 12 Nov 2026 13:03:31 GMT

Part Number: JFE150

Hello TI Support Team,

I'm currently working on a low-field MRI receiver design at the Technical University of Munich, using the JFE150 in a low-noise amplifier operating around 4.66 MHz.

I couldn't find a specified transition frequency (fT) in the JFE150 datasheet.

Could you please clarify:

Is the JFE150's unity-current-gain transition frequency (fT) characterized, and is a typical value available?
How does fT depend on the drain current (ID) and drain-source voltage (VDS)?
Are there any high-frequency characterization results, such as S-parameters or gain-versus-frequency measurements, available for this device?

As a rough estimate, we obtained approximately 119 MHz using gm ≈ 18 mS and Ciss ≈ 24 pF. However, these parameters aren't necessarily specified at identical bias conditions, so we'd appreciate a more reliable figure.

Thank you for your help!

Best regards

RE: JFE150: JFE150 – Transition Frequency (fT) and RF Performance Data

Chris Featherstone — Thu, 12 Nov 2026 20:03:20 GMT

Hello,

We do not have data for this however the unity gain frequency is strongly dependent on the application circuit. Have you tried simulating your circuit in Tina Ti? We offer the Tina models which I have attached below. There is a strong and weak model included. Tina is a free download (link below). Can you share your schematic for us to look at it as well?

e2e.ti.com/.../slpm351b-_2800_2_2900_.zip

I have made Gain vs Frequency measurements of the JFE150EVM I created, however this circuit would likely be different from your circuit.

Best Regards,
Chris Featherstone

RE: AMC3302: Part Failure Modes

Chris Ulmer — Thu, 12 Nov 2026 19:40:32 GMT

Hi Alexander,

The first failed unit I believe was damaged in re-work (hence the failure no longer being intermittent) so I won't return that one. Instead, I'll return one of the units that's failing intermittently (fails to start up about 5% of the time at -35C) and which has undergone another A-B-A swap test (with more careful hot-air removal.) The other one that's failing intermittently I'll keep for now and will move to a TI evaluation board just to further rule out anything circuit-related. In the mean time, I'd like to proceed with an NDA so we can understand the nature of the Q1 testing. If we can be assured that the Q1 part should rule out this issue, we'll be able to communicate that to our customer (Lockheed) with confidence. Please send an NDA regarding this issue and I'll forward that to my contracts team.

Thanks,
Chris

AMC3302: Part Failure Modes

Chris Ulmer — Sat, 07 Nov 2026 00:38:18 GMT

Part Number: AMC3302

I have an AMC3302DWE in a shipping product that seems faulty. The initial failure mode was that it went into "failsafe" mode intermittently at -30C or below. To further diagnose it, I moved that part to a different board location (swapping with another AMC3302) and now it stays in failsafe mode constantly at room temperature. Pin 13 and 16 (tied together) are at 3.1V instead of the 2.4V that I would expect and seem to indicate that the DC-DC is not turning on. Please let me know if there's any other tests to try or if I should sent that part to TI for failure diagnosis.

LM7322: Marking Date Code

Brandon Xiong — Thu, 12 Nov 2026 17:20:54 GMT

Part Number: LM7322

Mouser has Top marking-83XY, AZ4A as Date Code 1736 on device 1?

Top marking 79FX, AZA4 as Date Code 1736 on device 2? What is the actual Date Code? I'm expecting "83" as date code and "XY" as lot code? AZ4A is part marking.

RE: LM7322: Marking Date Code

Carrie — Thu, 12 Nov 2026 17:58:38 GMT

Hi Brandon,

The 83XY and 79FX date codes follow the format "YMLL".

Y = year the device assembly lot was produced

M = month the device assembly lot was produced

LL = internal lot tracking information

Therefore, "83XY" corresponds to the device being manufactured in March of 2018, and "79XY" corresponds to the device being manufactured in September of 2017.

It is possible to receive multiple lots within one shipment as long as the date codes are not more than 52 weeks apart. This is why the date code does not match the top side marking of every device received. The date code given (1736) matches the oldest device in the shipment.

Please let me know if you have further questions.

Best regards,
Carrie

INA236: Effect of PWM Frequency on INA236 Current Measurement Accuracy

Fumiya Michishita — Tue, 10 Nov 2026 11:58:43 GMT

Part Number: INA236

We are measuring the stepper motor current (approximately 40 mA to 150 mA) of a TMC5271 operating in StealthChop2 mode using an INA236. Since the measurement is performed while the motor is holding position, the current polarity is constant.

We compared the average current measured with a current probe to the current calculated from the INA236 CURRENT register and VSHUNT register. Under some conditions the values match well; however, under other conditions the current reported by the INA236 is lower than the value measured by the current probe.

We have also observed that the chopping frequency of the TMC5271 changes when the motor current setting is changed.

My understanding is that the INA236 calculates current by measuring the voltage across the shunt resistor. Could changes in the chopping frequency affect the sampling or averaging of the shunt voltage waveform and therefore cause differences in the measured current?

I have previously seen recommendations that the INA240 or INA241 are preferred for motor current sensing because the INA236 does not include PWM rejection.

Could the discrepancy observed in our measurements be related to the lack of PWM rejection in the INA236?

Additionally, if the PWM (chopping) frequency is fixed, would you expect the average current measurement accuracy or measurement repeatability of the INA236 to improve?

RE: INA236: Effect of PWM Frequency on INA236 Current Measurement Accuracy

Louis L — Thu, 12 Nov 2026 17:56:35 GMT

Fumiya,

Thanks for letting me know you've already run those tests. That result is informative and changes my ranking of the likely cause.

[quote userid="678926" url="~/support/amplifiers-group/amplifiers/f/amplifiers-forum/1696669/ina236-effect-of-pwm-frequency-on-ina236-current-measurement-accuracy/6537105"]Given these results, do you still believe the sampling-window/chop-period interaction is the most likely cause, or would you recommend investigating other factors?[/quote]

Sampling-window/chop-period interaction is no longer my top suspect. If partial-chop-cycle capture were dominant, increasing VSHCT × AVG should have driven the error toward zero, because longer conversion time simply averages more of the same fixed 2 µs samples (70 samples at 140 µs, 4,122 samples at 8.244 ms). Since it didn't, and bus/shunt multiplexing is already eliminated by continuous shunt-only mode, the error is almost certainly not a sample-count or window-alignment problem. I believe it is a front-end signal-fidelity problem, which averaging cannot fix.

The mechanism I'd now put first: the INA236 input frequency response is flat only to roughly 100 Hz before rolling off. A band-limited front end does not reproduce the true waveform before averaging, so a chopped, asymmetric current waveform can average low no matter how long the window is which is what I think you're seeing. This is the practical consequence of the INA236 having no PWM/chopper rejection, which is the reason the INA240/INA241 are the recommended parts for chopped-current inline sensing.

The next things I'd rule out would be:

Factor	Verdict	Basis
Shunt resistor tolerance	Check this	As I mentioned in my previous reply, we have seen shunt tolerance cause similar errors here.
ADCRANGE vs. actual swing	Check this	±20.48 mV / 625 nV/LSB vs ±81.92 mV / 2.5 µV/LSB — a mismatch against actual peak shunt swing matters at your 40–150 mA level

The most valuable next test I'd suggest: disconnect the shunt and apply a stable external DC voltage directly to the INA236 inputs at a level equivalent to your 40–150 mA readings. If the DC reading is accurate, the device and your configuration are clean and the error is entirely waveform-fidelity-related — which points directly towards the INA240/INA241 path. If it still reads low, the problem is in configuration, LSB scaling, or CAL/shunt-value programming.

[quote userid="678926" url="~/support/amplifiers-group/amplifiers/f/amplifiers-forum/1696669/ina236-effect-of-pwm-frequency-on-ina236-current-measurement-accuracy/6537105"]Also, how many TMC5271 chopping cycles would you recommend including within a single INA236 averaging window to accurately measure the true average current?[/quote]

What I can give you is the sizing basis and my engineering recommendation:

Target ≥ 100 chop cycles per single conversion, and ideally ≥ 1000 across the full AVG-accumulated window. Below ~100 cycles, a one-cycle boundary mismatch is a ≥1% error term; above it, partial-cycle contribution becomes negligible relative to your other error sources.
Sizing math: the longest single conversion is ~8.244 ms (CT = 111). At a 20 kHz chop rate that single conversion already spans ~165 chop cycles; with AVG = 1024 the accumulated window spans ~169,000 cycles. You are very likely already far past any reasonable cycle-count threshold, which is further confirmation that cycle count is not your limiting factor.

Ultimately, I no longer think window/chop-period interaction is your root cause because you've tested it out. Run the external-DC input test to separate device/configuration error from waveform-fidelity error, and verify your shunt tolerance. If the DC test is clean, the chopped waveform is exceeding what the INA236 front end can faithfully average, and the INA240 or INA241 driving an external ADC is the correct architecture for this measurement.

Louis