1
\$\begingroup\$

Intro: I'm designing a makeshift oscilloscope and now need an input cascade for it (it is more fun project for me, rather than industrial-grade tool).

The matter is I want to get something faster compared to popular 200 kSps things based on arduino (atmega). My current approach utilizes TLC5540 adc with stm32f030 controller, giving 8 MSps - quite enough for now.

The problem is: that ADC seem to have input not well suited for using directly:

  • it is low-impedance, hundreds Ohms, if I got it correctly
  • it is about 2V range.

Thus I want input cascade with higher input impedance (at least 10-100 kOhms) and either unity gain (with resistor divider) or adjustable 0.5-0.2 gain.

OpAmp voltage follower - usually I use something like LM358 for such case - but it won't suit the intended frequency; there are others, like TL084 - but they seem to need more than 5V power supply... I can resort to this option if no better available... MC3317x and MC3407 may suit marginally...

NPN voltage follower - no problem with frequency, but it won't work linearly below ~0.7V input as I understand.

I think, I can manage with some shift and gain - e.g. if input 0..5 V range would be mapped to 0.8..2.2 V - but the mapping needs to be linear. E.g. stupid and straightforward approach could look like this:

Linear voltage follower/divider

As about accuracy - it is not an extreme-precision tool of course, so 3-5% error is not critical...

So what better options may I miss? - probably some JFET/MOSFET follower? Thank you in advance for suggestions!

\$\endgroup\$
3
  • 2
    \$\begingroup\$ Go to the main opamp manufacturer sites, TI, analog, maxim, and use their parametric search tool to look for 5v, rail to rail, opamps with adequate bandwidth, then buy one from digikey. Sometimes, you just have to accept that the old faithful LM358 won't cut it, although the same unity gain voltage follower principle will. \$\endgroup\$
    – Neil_UK
    Jun 2 '18 at 7:09
  • \$\begingroup\$ @Neil_UK as long as they are unity-gain stable. More of a thing to think about if you go for high slew devices. \$\endgroup\$
    – Asmyldof
    Jun 2 '18 at 8:21
  • \$\begingroup\$ @Neil_UK - thanks, I've browsed shops more thoroughly and have found certain LMH66xx with 130 MHz (not extremely cheap) and MC33xx with 24 MHz... Now I can at least weigh options of using any of these specific OpAmp versus general transistor solution (especially since I've found ADC can be set to 0.6...2.6 V input range and resistors could be chosen to match this snugly)... Unless more answers will follow :) \$\endgroup\$ Jun 2 '18 at 9:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.