THS4302: High speed DAC buffer amplifier selection

Hi Jordan,

You are correct that the THS4302 is a good candidate for your application. Devices in the bandwidth range and slew rate performance of the THS4302 are very limited. Please let me look into your application to make the best recommendation possible.

Best Regards,

Ignacio

Hi Jordan,

I was able to look deeper into our portfolio for amplifiers that would work for your application. Although the THS4302 is one of the better performing amplifiers in this range, implementing it into a differential to single ended design that has good matching will be difficult. Like was mentioned the performance requirements for this application limits what is available. We have a decompensated amplifier that would work in this frequency range, however implementing a difference configuration while keeping the gain to an appropriate level is only adding a layer of difficulty to the design. One potential option is the LMH6703. The datasheet can be found here.

Best Regards,

Ignacio

Hi Ignacio

Thanks for that input. It seems we may not be able to find an amplifier that can do it all. One option would be for us to have a diff-to-SE amplifier first, then have the THS4302 as the output stage amplifier.

Are there any preferred TI options for a diff-to-SE capable amplifier that can do DC-2GHz? Something like the OPA855 or similar?

Hi Jordan,

You are correct that a solution to make a difference amplifier with discrete components in this bandwidth range will be difficult to find. The only devices that are capable of this bandwidth are the OPA858 and the OPA855 you highlighted; the only difficulty/concern is that these devices are decompensated which means they require a minimum gain to remain stable (7V/V in this case). Something close to this that is stable at 1V/V is the device I highlighted previously but it does not cover the full bandwidth range you require.

Best Regards,

Ignacio

Hi Ignacio

We also found the THS4304, which may be suitable as a diff-to-SE buffer, then use the THS4302 as our output driver.

Hi Jordan,

That might be a good amplifier to experiment with, the only reason I didn't recommend it was because the slew rate performance is so much lower than the THS4302, and you mentioned the slew rate performance as being a highlight of this part. The THS4304 does have the bandwidth requirement so it might be a good option for this application.

Best Regards,

Ignacio

Hi Ignacio

Our slew rate requirement can be eased if we accept using the THS4302 as a second stage buffer to a diff-to-SE stage.

We have tried the THS4304 but when configured as a unity difference amplifier, it's bandwidth is reduced to <1 GHz, so it seems not suitable. The OPA855/OPA858 could be used but it means we have to deal with the high minimum stable gain of 7V/V.

If there are any other amplifiers you can suggest for a diff-to-SE stage with ~2GHz of bandwidth, please let me know.

Thanks

Jordan

Hi Jordan,

In a gain of 7V/V, the OPA855/OPA858 should perform in the bandwidth region you are looking for. These are the fastest amplifiers we have and are not fixed gain devices.

Best Regards,

Ignacio

Hi Ignacio

Yes we have found the same for what TI can provide. The main downside is using the OPA855 as a DAC buffer is that a large attenuation is needed at the DAC output to make sure that the output of the OPA855 (in a gain of 7V/V) is within spec. 

Coming back to using the THS4302 as a DAC buffer. The main issue I can imagine is that the internal 200R/50R feedback resistors can't be exactly matched when setting up as a difference amplifier, which would cause CMRR degradation. Is there any info on the tolerance of the 200R and 50R internal feedback of the THS4302?

Hi Jordan,

I will look into this detail, and I will get back to you as soon as I have an update.

Best Regards,

Ignacio