• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 48 subscribers
  • Views 238 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

OPA2695: Spice Model

Gordon McKenzie
Intellectual 590 points
Part Number: OPA2695
Other Parts Discussed in Thread: OPA695

Hello,

I notice that the dual OPA2695 is covered by a different datasheet to the single OPA695, with the OPA695 looking to be considerably faster in terms of both small signal bandwidth and slew rate. I have an existing application that uses the 2695 but was wondering if I could extend the frequency range a little by changing to using 2 of the single parts instead so I had hoped to do a quick comparison in Spice. However, the model listed on the OPA2695 product page looks to be the single OPA695 model.

Normally I'd be happy to use the single model and take it to be representative of the dual part, as usually they have the same datasheet and essentially have the same spec. That doesn't seem to be the case here. Can you confirm which of the two parts the model is representative of and, if it's the single OPA695 as I expect, is there anything that can be done to model the dual part?

I'm also curious as to why the specs for single and dual parts are different for this one in the first place, as that's not usually the case.

Thanks,

Gordon.

  • +1
    Guru 48395 points

    Very good question Gordon, 

    The OPA695 single obviously came first, about 2003 where the modeling effort was mainly put into that device. In these very fast CFA devices, literally package delay and bond wire inductance are part of the open loop response, that device is optimized at a gain of 2 with a lower 348ohm Rf to get the best butterworth response. 

    Much later the dual OPA2695 was introduced (many years later), its larger package and bond wire inductance required a higher Rf, more compensation in effect for the CFA, showing a lower F-3dB with the stated 402ohm Rf at gain of 2 100ohm load. That Rf is arrived at empirically on the bench and evm board and should have fed back into a slight mod to the Zol phase essentially in the model, that was not done apparently, 

    But yes, using two singles in the smallest package and lowering the Rf as far as you can for desired flatness should be a little faster nominally. 

  • 0 in reply to Michael Steffes
    Intellectual 590 points

    Thanks, Michael for your quick response, and apologies for not replying until now.

    That's a good explanation, and much appreciated. As far as modelling goes, I'll probably just use the OPA695 single model but use a bit more caution than normal and add some extra margin at the top end of the frequency range for the 2695.

    Gordon.