2
\$\begingroup\$

The output voltage from AD5611 DAC is rail-to-rail, and minimum Vcc is 2.7V - therefore, the signal range is 0-2.7V. I need to attenuate this signal to let's say 0-2V range. The DAC output is very weak, so I need to use an OpAmp to buffer it. Is it a good practice to use G=<1 configuration for decreasing signal's amplitude, or is there a better way to do it?

I saw this question, but I do not think that using voltage divider would be a good idea in my case, since the DAC output is very weak, and using large resistances would introduce noise.

\$\endgroup\$

2 Answers 2

Reset to default
2
\$\begingroup\$

A voltage divider followed by a buffer is not a bad way to approach it. If your supply is 5V you could use something like 30K and 20K resistor divider and a voltage follower as a buffer. The output impedance of the DAC is 0.5 ohm, so 50K won't have much effect.

\$\endgroup\$
6
  • 1
    \$\begingroup\$ Beat me to it by about 10 seconds. Drat! \$\endgroup\$
    – Alfred Centauri
    Commented Jun 5, 2014 at 19:33
  • 1
    \$\begingroup\$ Beat me to it by 1 min and 10 secs +1 \$\endgroup\$
    – Andy aka
    Commented Jun 5, 2014 at 19:38
  • \$\begingroup\$ For this particular case, is it better to provide higher voltage for DAC (possibly expending its dynamic range), and then use appropriate resistors, or use smaller voltage? Say 5V-30k-20k OR 3V-10k-20k ? \$\endgroup\$
    – Nazar
    Commented Jun 5, 2014 at 20:26
  • 1
    \$\begingroup\$ It's better to use a clean and stable power supply since that's the reference. If you can derive a lower voltage supply that's better than the 5V supply, then that probably trumps any difference in the other factors. You could use two 5V regulators too. All other things being equal, it's probably better to use the higher voltage supply. \$\endgroup\$
    – Spehro 'speff' Pefhany
    Commented Jun 5, 2014 at 20:32
  • \$\begingroup\$ @Andyaka, you're not getting my star! :) \$\endgroup\$
    – Alfred Centauri
    Commented Jun 5, 2014 at 22:28
0
\$\begingroup\$

If you can afford to use two opamps you can use a normal non-inverting configuration with another gain-of-two opamp (example value) in the feedback path. Hence, the overall gain is 0.5.

\$\endgroup\$
3
  • \$\begingroup\$ Do you mean the first one will be a buffer, but instead of shorting Vin- with Vout, place another stage of G=+2? Since they say that for this particular case using a simple voltage divider is not such a bad idea, would the two-opAmp approach be more advantageous? \$\endgroup\$
    – Nazar
    Commented Jun 5, 2014 at 20:18
  • \$\begingroup\$ The gain of two amplifier has to swing to both rails, so it would have to be RRIO. Stability (phase margin) might be an issue. \$\endgroup\$
    – Spehro 'speff' Pefhany
    Commented Jun 5, 2014 at 21:55
  • \$\begingroup\$ @Naz, yes - that´s what I mean. However - as mentioned in the comment above, stability will be degraded and some additional stabilizing means are perhaps necessary. Hence, it is not clear which of the two methods will be advantageous. But that is a normal situation in case of sveral solution alternatives, because very often a trade-off is necessary because of conflicting requirements. \$\endgroup\$
    – LvW
    Commented Jun 6, 2014 at 7:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.