AMC3330-Q1: Surge protection for IC

Kunal Pathak

Part Number: AMC3330-Q1
Other Parts Discussed in Thread: AMC3302, AMC3330, TINA-TI, AMC1300

Hi,

For Line-to-Line voltage measurement, we are using AMC3330-Q1 IC. We have referred a TI application note, where the low voltage end IC is used for voltage measurement and their diodes are used for surge protection. Similarly we have used diodes for surge protection in our application where we have used AMC3330-Q1 IC (please refer below image). Please confirm if we can use diodes for surge protection for AMC3300-Q1 IC and also please inform if AMC3330-Q1 IC has any internal surge protection feature.

over 3 years ago

+1 Alexander Smith over 3 years ago

TI__Guru 50561 points

Hi Kunal,

The AMC3330-Q1 has internal ESD diodes that allow for the device to withstand up to 6V below HGND or 500mV above V_HLDOOut as per the Absolute Maximum Ratings table in the datasheet. Depending on what type of transients you expect to see on the input, these external diodes may or may not be necessary. Most important is to limit the input current to be less than 10mA.

0 kai klaas69 over 3 years ago

Guru 140200 points

Hi Kunal,

voltage measurement? Then RL11 and RL12 limit the input currents sufficiently. See figure 8-1 and table 8-2 of datasheet.

And don't forget to connect INN to HGND.

Kai

0 Kunal Pathak over 3 years ago in reply to Alexander Smith

Intellectual 375 points

Hi Alex,

Please confirm if AMC3302 IC has internal surge protection.

Regards,

Kunal Pathak

0 Kunal Pathak over 3 years ago in reply to kai klaas69

Intellectual 375 points

Thanks Kai!

+1 Alexander Smith over 3 years ago in reply to Kunal Pathak

TI__Guru 50561 points

Hi Kunal,

The device is tested during production to ensure that the isolation barrier can withstand a surge voltage above 6kV, as noted in the datasheet. The AMC3302 has the same internal ESD protection diodes as the AMC3330-Q1. This however does not apply directly to the inputs, they can not withstand 6kV and table 6.1 in the datasheet should be observed.

As Kai mentioned, I would recommend installing 0ohm resistor R1116 to ensure that the inputs stay within the common-mode input range.

0 Kunal Pathak over 3 years ago in reply to Alexander Smith

Intellectual 375 points

Thank you Alex!

0 Alexander Smith over 3 years ago in reply to Kunal Pathak

TI__Guru 50561 points

Hi Kunal,

I should reiterate that these internal diodes are for ESD, not surge protection. The energy for ESD is very small compared to surge, so I recommend leaving the external diodes in the design to protect the device from surge voltages.

0 Kunal Pathak over 3 years ago in reply to Alexander Smith

Intellectual 375 points

Hi Alex,

We are seeing a clamp in input voltage when we are using the external diode. So, is there any recommended diodes for surge protection.

0 Alexander Smith over 3 years ago in reply to Kunal Pathak

TI__Guru 50561 points

Hi Kunal,

That diode should not start conducting significant current when less than 4V is applied. Given that the linear input range of the AMC3330 is +/-1V, this breakdown voltage should not interfere significantly during normal operation.

Can you please describe your issue in greater detail?

This document may help: https://www.ti.com/lit/an/slvae37/slvae37.pdf

This video series may help as well: https://training.ti.com/eos-and-esd-adc

0 Kunal Pathak over 3 years ago in reply to Alexander Smith

Intellectual 375 points

Hi Alex,

Could you please suggest any suitable diode part for AMC3330-Q1, since we have both positive and negative voltages.

0 kai klaas69 over 3 years ago in reply to Kunal Pathak

Guru 140200 points

Hi Kunal,

I'm a bit confused. You say you want to make a voltage measurement. In this case you would use a high ohmic voltage divider according to figure 8-1 and table 8.2 of datasheet. But with this high ohmic voltage divider you would not need a surge protection because the high ohmic voltage divider would succiently limit the input current of AMC3330.

Or do you want to perform a current measurement and sense the voltage across a shunt resistor? In this case you again would not need a surge protection because then you would connect HGND to one side of the shunt and the AMC3330 would no longer see any surge at its inputs. The surge would drop across the barrier between input and output side.

So, please explain in detail what you are planning to do.

Kai

0 Kunal Pathak over 3 years ago in reply to kai klaas69

Intellectual 375 points

Hi Kai,

We are using AMC3330-Q1 for AC voltage measurement. Even though, we have current limiting resistor divider network but in case of surge voltage in AC line, we will be seeing high voltage across input pins of AMC3330-Q1. Do you have any suggestions to protect the IC from such scenario.

0 Alexander Smith over 3 years ago in reply to Kunal Pathak

TI__Guru 50561 points

Hi Kunal,

I cannot suggest parts from other manufacturers. The diode you selected and the configuration looks fine to me, assuming that you have populated R1116 with a 0ohm resistor to keep the negative input tied to HGND. Additional details on the issue would be helpful.

The video series and paper I linked should provide you with the information necessary to select a diode for surge protection.

0 Kunal Pathak over 3 years ago in reply to Alexander Smith

Intellectual 375 points

Hi Alex,

Please find the below schematic for AC line to line voltage measurement using AMC3330-Q1.

We are measuring the input voltage to the IC across the marked points in schematics i.e. before the CMC choke and after the input filter. Below are the waveform, the positive peak is fine but we are seeing the negative peak is reduced.

As the recommended CMC part went obsolete, so we have used different CMC part. Below is the comparison of both the CMC parts.

Is the issue happening due to change in CMC part (change in impedance or DC resistance)?

If so, can you please provide some references for CMC choke parameter calculation/ selection or please suggest alternate part.

0 kai klaas69 over 3 years ago in reply to Kunal Pathak

Guru 140200 points

Hi Kunal,

your schematic is highly confusing

Could you please show a schematic which only contains those components that are actually mounted?

Kai

0 Kunal Pathak over 3 years ago in reply to kai klaas69

Intellectual 375 points

Hi Kai,

PFB schematic FYR. I have crossed the components in schematics which are not populated (DNP) on PCB.

We are measuring the voltage before the CMC (CH1) and after the input filter (CH2). We are seeing difference in negative peak value of output voltage (CH2), compared to input voltage (CH1). Please refer below image.

As the recommended CMC part went obsolete, so we have used different CMC part. Below is the comparison of both the CMC parts.

Is the issue happening due to change in CMC part?

If so, can you please provide some references for CMC parameter calculation/ selection or please suggest alternate CMC part.

0 Alexander Smith over 3 years ago in reply to Kunal Pathak

TI__Guru 50561 points

Hi Kunal,

It is difficult to read the schematic that you most recently posted. Is there a voltage drop across a resistor in the negative input line that is not present in the positive input line?

I would think that the drop across the inductor would be the same. Are you seeing this behavior on multiple boards or just one?

0 kai klaas69 over 3 years ago in reply to Kunal Pathak

Guru 140200 points

Like this?

Remove the common mode choke and try again.

Swap the scope channels and try again.

Kai

0 Kunal Pathak over 3 years ago in reply to kai klaas69

Intellectual 375 points

Hi Alex/ Kai,

We are seeing the issue in multiple boards.

We have removed the CMC, but the issue is still present. We are seeing the difference in positive and negative peaks in waveform after the input filter.

We have not tried swapping the channels, but will try that and let you know.

0 Kunal Pathak over 3 years ago in reply to Kunal Pathak

Intellectual 375 points

Hi Alex,

In continuation to above mail.

In AMC1300B application note (below is the link of application note), it's recommended to use R3' resistor to cancel out Gain error introduced by the bias current flowing through R3 (please refer below image).

https://www.ti.com/lit/an/sbaa350/sbaa350.pdf?ts=1631112360284&ref_url=https%253A%252F%252Fwww.ti.com%252Fproduct%252FAMC3301

Please confirm for AMC3330-Q1 IC also R3' resistor should be used or not.

0 Alexander Smith over 3 years ago in reply to Kunal Pathak

TI__Guru 50561 points

Hi Kunal,

This R3' resistor should not be necessary for the AMC3330-Q1 since the input impedance of the device is high.

0 kai klaas69 over 3 years ago in reply to Kunal Pathak

Guru 140200 points

Why not running a TINA-TI simulation?

kunal_amc3330_1.TSC

Keep an eye on the input bias currents of AMCs and note that the input bias current of AMC3330 is considerably smaller than the input bias current of AMC1300.

Kai

0 Kunal Pathak over 3 years ago in reply to kai klaas69

Intellectual 375 points

Hi Alex/ Kai,

Thanks for the confirmation regarding R3' resistor.

Please provide suggestions on how the positive and negative peak voltage difference issue can be resolved. We tried interchanging the channels but the issue remains the same. We have checked on other boards also but we are seeing the difference.

0 kai klaas69 over 3 years ago in reply to Kunal Pathak

Guru 140200 points

Hi Kunal,

why don't you connect the ground pin of scope probe to HGND of AMC3330?

Are you sure C1108 is actually 8n2?

Kai

0 Alexander Smith over 3 years ago in reply to kai klaas69

TI__Guru 50561 points

Couple other things to confirm, are both channels set at 0V or is there an offset?

Is it possible that the filter resistors, R1111, R1115 are larger than 10ohm?

I suggest removing the CMC and C1108 then replacing R1111, and R1115 with 0 ohm resistors

0 kai klaas69 over 3 years ago in reply to Alexander Smith

Guru 140200 points

Connecting both scope probes to the same signal would show whether there's an issue with the scope.

Kai.

Amplifiers

Amplifiers forum

AMC3330-Q1: Surge protection for IC