OPA4354: OPA4354 Bandwidth

Morning Eddie, 

Actually kind of hard to know exactly what is going on here, 

1. The Spec table clearly shows 250Mhz for gain of 1, which is far beyond what you would expect for a 100Mhz GainBandwidth product, 

The measured plot shows this as well - with no peaking?? That is quoting a 1kohm load which does make a difference. One possiblity here is that bandwidth extension is simple feedthru for this gain of 1 case. We used to do an added cal of feedthrough with the device powered down and simply removed that from the measured data as being fictional BW extension. The test there is the small step response. A feedthru will show a sharp spike then recovery to the actual SSBW set rise time. (I first got into this measuring the CLC110 S21 which far exceeded my design sims, feedthru was the culprit there). 

The 2011 TINA model running gain of 1, 1kohm load shows about 160Mhz which makes more sense for a 100MHz GBP.. I did not run a LG sim to find phase margin, but that 1.6X factor is  pretty common (Figure 4 in this article), 

https://www.planetanalog.com/stability-issues-for-high-speed-amplifiers-introductory-background-and-improved-analysis-insight-5/

And finally, there is a humorous error in the OPA4354 web folder, somehow a 100V part description got attached here - you wish on a CMOS part??

So Eddie, I probably did not answer your questions, but 

1. Unity gain BW above 100Mhz can be influenced by many factors (here, you could also extend the bandwidth with a little more peaking by increasing the feedback R introducing more phase shift around the loop with the pole back to the inverting input parasitic C). 

2. Frankly, that 250Mhz looks like a marketing number, not a particularly reliable physical number. 

3. The slew rate is very low on this device for a >100Mhz part. So those BW are only even approachable at low test swing levels. 

High speed quads are problematic - isolation is an issue. 

1. Is this just a datasheet validation exercise - if so, consider lowering test level and/or increasing feedback R if you want to measure higher SSBW. 

2. If you really need a quad >200MHz, find out the actual Vpp needs and load and if unity gain is a constraint. 

3. SIngle channel parts are easier to use >200MHz, look at the BUF602 if you need unity gain.