• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 49 subscribers
  • Views 874 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.

OPA2674: High-output-current opamp that can drive large 300pF load at 60MHz bandwidth

Weiyu Leng
Intellectual 360 points
Part Number: OPA2674
Other Parts Discussed in Thread: LM317, OPA2677, THS3491

To whom it may concern,

I would like a recommendation of an opamp product that can provide 500mA peak current and is stable under capacitive load of 300pF. 

Basically, the load is a current source that can pull a peak current of 500mA in parallel with an equivalent capacitor of 300pF.

The bandwidth should be around 60MHz.

Typically opamp requires a series damping resistor at the output (like OPA2674) for large capacitive load, but if the load needs to pull continuous current at the same time, this series resistor would create a large voltage drop. 

I wonder if there is a product that can drive capacitive load and provide high output current at the same time, without a series resistor.

Thank you.

Weiyu

  • 0
    Guru 140200 points
    Hi Weiyu,

    hmm, I'm not quite clear what you mean. Do you need a current source or a voltage source for another current source? Why not taking a voltage regulator to supply the current source?

    Kai
  • 0 in reply to kai klaas69
    Intellectual 360 points

    Hi Kai,

    Thanks for replying. 

    LDO produces fixed voltage. I need a modulated voltage across a varying current source in parallel with a 300pF capacitor.

    Weiyu

  • 0 in reply to Weiyu Leng
    Guru 48395 points
    what kind of voltage swing are you trying to produce -is it this that is 60MHz or the current source modulation (+/-500mA at 60MHz) Or is the current source fixed and you just trying to put a modulation voltage on it?
  • 0 in reply to Michael Steffes
    Intellectual 360 points
    Voltage swing is from 1V to 5V. Current swing is from 60mA to 500mA. They are strongly correlated, so when current is 500mA, voltage needs to be 5V as well. When current is 60mA, voltage needs to be 1V.

    Thanks.

    Weiyu
  • 0 in reply to Weiyu Leng
    Guru 48395 points

    thanks, but still - are you forcing a voltage at the output pin through 60MHz content - and is it sinusoidal or a waveform that needs to have more harmonics passed. If they are correlated, sound like a load and not a current source connected to the outputs. I have something working pretty well, (200MHz flat, 300Mhz F-3dB) but want to make sure I am looking at the right things. 

  • 0 in reply to Michael Steffes
    Intellectual 360 points
    Hi Michael,

    60MHz is the highest content, no more harmonic is needed. The signal is random and takes bandwidth from 0 to 60MHz.
    The voltage needed across the current source and the current of the current source are correlated, but from small signal point of view (for stability), the opamp needs to drive a capacitor, rather than a resistor.

    For example, OPA2674 cannot drive a large capacitive load, so when I attach a current source (a large capacitor in essence) to its output, the opamp oscillates. When I put a series resistance between the output of OPA2674 and the current source, the opamp is stable, but that renders a big voltage drop when my current source pulls current.

    Thank you.
    Weiyu
  • 0 in reply to Weiyu Leng
    Guru 140200 points

    Hi Weiyu,

    if the bandwidth would be only 1...2MHz, I would recommend to modulate the ADJ pin of LM317...

    Have you thought about using the dual feedback topology, as discussed here?

    Kai

  • 0 in reply to kai klaas69
    Guru 48395 points

    ello Weiyu

    I noticed an earlier question on output drive for the OPA2674 heading towards paralleling op  amps. Some of these dual CFA xDSL line drivers were configured internally for only differential push/pull operation - this might be one, the earlier one, the OPA2677 was independent op amps.

    However, back on this request - yes, Kai's was suggesting the dual feedback approach for directly driving Cload - oddly, that just came up in an article I published where I noted any of those direct Cfeedback circuits you see using VFA can be adapted to CFA by putting the feedback R inside the summing junction to the inverting node.

    https://www.planetanalog.com/author.asp?section_id=3404&doc_id=565097&

    https://www.planetanalog.com/author.asp?section_id=3404&doc_id=565097&

    I started sims with the latest, most capable CFA, the THS3491 - looks kind of promising in sim. The voltage swing at the cap load will be different than the output pin - but that 2ohm isolation in the dual loop approach does not give much rise - you would want to adjust the supplies to minimum operating overhead. Here, I got it flat through 145MHz small signal gain of 2 with 297MHz F-3dB. You can increase the gain here if want - the key is that 750ohm inside the summing junction to the inverting input. I put 500mA DC into the output, might not need that. 

    Have to put these snips into a word file apparently and then attach file. I also checked the phase margin, really good at 56deg. The mapping from LG=0dB to F-3dB is 1.6X in this case - shown in fig. 4 in this article, 

    https://www.planetanalog.com/author.asp?section_id=3404&doc_id=565056&

    here are the sim snips, 

    THS3491 driving 300pF.docx

     and here is the closed loop sim, seems almost too good to be true, but the sims are encouraging for this difficult problem - again you can adjust the gain higher here, it was bandlimiting when I tried gain of 6, but you have a lot of margin. 

    300pF cap load with THS3491.TSC