OPA615: Figure 48 in datasheet, results not as expected

Larry Rose

Part Number: OPA615
Other Parts Discussed in Thread: BUF602, , TINA-TI, OPA860

Following the datasheet figure 48 bipolar peak detector circuit using the OPA615 in TINA the negative Vout- doesn't work as expected. The positive Vout+ will track the Vpositive peak but the Vout- vpeak is about 1.8V more negative than the input. If I do two circuits, neither with the buf602, one for Vout+ and one for Vout- and drive them with the same input I get the expect Vpeaks for each circuit difference being the diode direction.

I'd like to capture the Vpeak values of a bipolar signal between -1V to 1V. Has this circuit been modeled in TINA, can a working schematic file be shared to use in the TINA tool?

over 3 years ago

0 Sima Jalaleddine over 3 years ago

TI__Mastermind 25295 points

Hello,

Are you testing Figure 48 with the OPA615 and the BUF602 Tina-TI model? Can you share the Tina-TI file with us? This specific circuit was not modeled in Tina-TI, but we can help you get a working solution. To share a file over E2E: press insert, then image/video/file and upload file, or you can drag and drop file to upload.

Thank you,

Sima

0 Larry Rose over 3 years ago in reply to Sima Jalaleddine

Prodigy 10 points

OPA615_FIG48.TSC

0 Sima Jalaleddine over 3 years ago in reply to Larry Rose

TI__Mastermind 25295 points

Hello,

Thank you for the file, I will get back to you by Thursday.

Thank you,

Sima

0 kai klaas69 over 3 years ago

Guru 140200 points

Me thinks the circuit from figure 48 is crap

I would do it this way:

https://e2e.ti.com/support/amplifiers/f/amplifiers-forum/763240/opa842-non-inverting-resistor-calculation

Kai

0 Larry Rose over 3 years ago in reply to kai klaas69

Prodigy 10 points

Thanks Kai. Using two OPA615 is what I shared to confirm the circuit. I didn't want to consume the extra PCB space and was hoping that FIG 48 had missing circuity or the TINA sim for the OPA615 and/or BUF602 was incorrect.

Identifying the root problem will provide the confidence to design and to use these parts on my application.

Hi Sima, have you had any luck getting to the bottom of this datasheet reference circuit versus TINA results?

Larry

0 kai klaas69 over 3 years ago in reply to Larry Rose

Guru 140200 points

Hi Larry,

this question comes up again and again. So it would be really nice from Sima to build up the questionable circuit from figure 48 and do some testing. If the circuit is really not working, TI could eventually carry out a revision of datasheet then

As it turned out during the simulations that the diodes play an important role, the test could be performed with several different diodes, like BAT17, BAV99, HSMS2812 (or a suited follower), or else.

I had to do with the OPA660 in the 90' while studying physics and can say that all these measurements and circuit modifications are truly challenging as this chip is really really fast

Even the layout is critical...

Kai

0 Larry Rose over 3 years ago in reply to kai klaas69

Prodigy 10 points

Thanks again Kai, I appreciate the follow up and comments.

Hi Sima, can you share update on your findings? If possible I'd like to wrap up my design this week so your help is critical, thanks in advance.

Regards,

Larry

0 Sima Jalaleddine over 3 years ago in reply to Larry Rose

TI__Mastermind 25295 points

Hello Kai,

Thank you Kai for the help on this!

Hello Larry,

Sorry for the very long delay on this. Here are two versions of the peak detector that could work. The first is from the figure 48 in the datasheet, and the second is Kai's circuit from the linked above thread. The first one does not have a long ramp up time for the initial peaks compared to the second one but does not compensate for the diode voltage drop. Which type are you looking for, for your application?

PeakDetectorFig48.TSC

6622.duc.TSC

Thank you,

Sima

0 Larry Rose over 3 years ago in reply to Sima Jalaleddine

Prodigy 10 points

Hi Sima,

I've included a sample peak to peak waveform that I will be trying capture the peaks from. I've went with a modified version of the 6622.duc. Please take a look and advise there is anything else I can change to more accurately store the peak values.

This is in layout now so I've got a little time to tweak the design.

Regards,

Larry

PeakDet_OPA615.TSC

0 Sima Jalaleddine over 3 years ago in reply to Larry Rose

TI__Mastermind 25295 points

Hello Larry,

I reviewed your file, and I would replace the diodes with Schottky diodes. The transient show consistent values in the hold stage. If you would like, we can also review the layout portion of your design.

Thank you,

Sima

PeakDet_OPA615_Updated.TSC

0 kai klaas69 over 3 years ago in reply to Larry Rose

Guru 140200 points

Hi Larry,

another option is to use the faster OPA860:

larry_opa860.TSC

Kai

Because of the Thanksgiving holiday in the U.S., TI E2E™ design support forum responses may be delayed from November 25 through December 2. Thank you for your patience.

Amplifiers

Amplifiers forum

OPA615: Figure 48 in datasheet, results not as expected