INA819: Bandwidth is 1.5MHz only

Danilo A.

Part Number: INA819
Other Parts Discussed in Thread: INA826, INA826EVM,

Hi Team,

Our customer replaced the INA826 from the INA826EVM evaluation module with INA819 but is having trouble reproducing the datasheet performance. While the datasheet of INA819 states a -3dB frequency at unity gain of 2.0 MHz, he is able to get around 1.5 MHz only. According to our customer, he is using a signal generator as input and have connected an oscilloscope with 1 megaohm, 15 pF input impedance to the amplifier output. As I understand, aside from INA819 there is no components in the INA826EVM evaluation module that our customer needs to modify to test INA819. Can you please confirm?

Regards,

Danilo

over 2 years ago

0 Tamara M Alani over 2 years ago

TI__Genius 10850 points

Hi Danilo,

Thanks for your post, I can help you. Can you please share more information on how the EVM is set up and which components are populated? For instance, any differential / common mode capacitors C1, C2, C3? What are the supplies? What is the reference voltage and how is it connected (through the OPA)? What does the input signal look like? A schematic outlining each component and the test set-up will help us determine the issue:

0 Danilo A. over 2 years ago in reply to Tamara M Alani

TI__Mastermind 24735 points

Hi Tamara,

According to our customer,

Our EVM has not had any of the input capacitors (C1, C2, C3) fitted, and is not modified except for swapping the INA826 with an INA819. The supplies used were +/- 15V with a 0V REF voltage fed directly (not through the OPA). The circuit shown in the image below was used on the input side. Voltage source V1 is feeding a sine signal at 50 mVpp amplitude.

The transformer on the input side is a Coilcraft PWB2010L, https://www.coilcraft.com/en-us/products/transformers/wideband-rf-transformers/smt/pwb/pwb2010/

Regards,

Danilo

0 Danilo A. over 2 years ago in reply to Danilo A.

TI__Mastermind 24735 points

Hi Tamara,

Correction, when using dual supplies (+/-15 V) the transformer center tap is connected to 0 V (ground). We have also tested without a transformer, ie. connecting V1 directly to the input to feed a 50 mVpp sine wave.

Resistor RG is removed for unity gain measurements.

Regards,

Danilo

0 Tamara M Alani over 2 years ago in reply to Danilo A.

TI__Genius 10850 points

Thanks for the additional information, this is really helpful. The INA819 specs the BW at unity gain to be 2MHz, but it is important to note that this is a "typical" spec. We can expect variance across production lots, and for an AC parameter that can be up to +/- 30%.

If you are in need of 2MHz, I would recommend testing a different INA819 or finding an INA with higher BW. Here are a few to consider: https://www.ti.com/amplifier-circuit/instrumentation/products.html#p358max=15;50&p1181typ=3;28&sort=p1181typ;asc

0 Helge Langen over 2 years ago in reply to Tamara M Alani

Prodigy 60 points

Hello Tamara,

It is I who represent the customer in this case. May I ask what was the test setup used for determining the INA819 datasheet numbers? Was the INA826EVM used, or some other board? What external instruments were used to apply stimulus to the inputs, and how were they connected?

Thanks,

Helge

0 Tamara M Alani over 2 years ago in reply to Helge Langen

TI__Genius 10850 points

Hi Helge,

The bandwidth spec in the datasheet is characterized. It is indirectly "tested" through iq (quiescent current). Since we test iq, we can indirectly verify the bandwidth.

Amplifiers

Amplifiers forum

INA819: Bandwidth is 1.5MHz only