0
\$\begingroup\$

I'm using a LM4565FV-GE2 opamp to amplify the sine signal generated by an AD9837 DDS chip, but the opamp output is stuck at around ~11.5V. I'm using a DS3502 digipot to control the gain, represented in the schematic as a pot.

schematic

simulate this circuit – Schematic created using CircuitLab

I've checked the sine output at the non-inverting input and it's good. The digipot appears to be working correctly as I'm able to change the voltage on the inverting input by changing the wiper position. Power rails are good and nothing's shorted on the board.

Is the opamp busted? Did I screw up somewhere?

\$\endgroup\$
5
  • \$\begingroup\$ You're running single supply, where's the virtual earth? \$\endgroup\$
    – Kartman
    Commented Jan 6, 2022 at 5:02
  • \$\begingroup\$ @Kartman I was not aware that was a thing. After some reading I believe that means I need to replace the ground on my pot with 12V/2? \$\endgroup\$
    – Christopher Moore
    Commented Jan 6, 2022 at 5:10
  • \$\begingroup\$ @Kartman Would just using a split supply with the -12V instead of ground to the opamp work? \$\endgroup\$
    – Christopher Moore
    Commented Jan 6, 2022 at 5:58
  • \$\begingroup\$ I don’t think the digipot would appreciate a negative voltage - check the datasheet. I’d go with the 12V/2 for the virtual earth. \$\endgroup\$
    – Kartman
    Commented Jan 6, 2022 at 8:55
  • \$\begingroup\$ @Kartman If my input to the amplifier never goes below 0V would that be an issue? Wouldn't the 12V/2 virtual earth turn my 0-0.6V input into 12V/2-gain*0.6V? \$\endgroup\$
    – Christopher Moore
    Commented Jan 6, 2022 at 14:59

1 Answer 1

Reset to default
2
\$\begingroup\$

The op-amp you are using is not a single-supply type. It can only reliably deal with voltages in the range of about Vee+3 to Vcc-3, with a single 12V supply that is about 3V to 9V. The output range is similar (the output cannot go all the way to ground or all the way to Vcc).

The output of the DDS chip is typically 0.037V to 0.645V, so it is way outside the allowable common mode input range for that particular op-amp when it is operated from a single 12V supply.

If you are okay with having a bias of G\$\cdot\$ 304mV (where G is the gain) in the output then you can use the digipot as shown, however you will need to use an op-amp that has an input common-mode range that includes the negative supply (and you also probably want one that has an output that can swing close to the negative supply rail).

So, look for a "single supply" op-amp or a RRIO (rail-to-rail input-output) op-amp that can work from 12V and meets your other requirements. RRIO will definitely work however performance may be inferior in subtle ways (such as distortion) compared to a single supply. Or not, since your application will likely leave half of the complex RRIO front end unused due to the low maximum input voltage.

\$\endgroup\$
9
  • \$\begingroup\$ I've found this part. It's rail-to-rail and I think it meets my other requirements. Do I also need to fix the issue with the lack of a virtual ground someone mentioned in the comments? Or should I use a split supply? \$\endgroup\$
    – Christopher Moore
    Commented Jan 6, 2022 at 15:07
  • \$\begingroup\$ Could you also help me understand where the bias is coming from using the digipot? \$\endgroup\$
    – Christopher Moore
    Commented Jan 6, 2022 at 15:37
  • \$\begingroup\$ You don't need a virtual ground (or AC coupling)if you are happy with the output bias. The output of the DAC is already biased above ground by about 300mV. For example, with a gain of 10, the output would go from 0.37V to 6.45V). I would pick an amplifier with a bit more supply voltage tolerance for a 12V nominal supply- Abs max is only 600mV higher than that. You are exceeding the input CM limits of the amplifier, so anything can happen at the output, including what you are seeing. There may still be something else wrong in practice now (damaged part, wiring issue etc.) but fix that first. \$\endgroup\$
    – Spehro 'speff' Pefhany
    Commented Jan 6, 2022 at 16:14
  • \$\begingroup\$ So just to clarify, are you saying that the circuit show in the schematic of my question is fine, it's just the opamp I'm using is inappropriate? And I don't need a split +12V -12V supply? The output bias is fine as long as it really means the output will go from 0.37V to 6.45V at a gain of 10. \$\endgroup\$
    – Christopher Moore
    Commented Jan 6, 2022 at 16:19
  • \$\begingroup\$ Pretty much.We don't really know if it's fine or not since your intentions are not clear. Yes, it should do as you said with an appropriate op-amp and the present single supply and the way the digipot is connected in your schematic. \$\endgroup\$
    – Spehro 'speff' Pefhany
    Commented Jan 6, 2022 at 16:32

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.