I decided to quickly put together a circuit today but found an issue that has sometimes happened when making circuits with TL072.

Briefly, the circuit is a non-inverting amp with about 34dB gain in the audio region, including some switchable clipping diodes in the feedback loop. Since supply is 9V, the input is biased with Vref.

enter image description here

Vref is set to be roughly 6V. This is because of the common mode input range of the TL072 being 4V to 9V when the supply rails are 0 and 9V. However, when I power it on and input a signal (active guitar, <500mV peak), the output comes much weaker and heavily distorted, the gain knob starts working in reverse and the bias voltage drops to roughly 1.6V. If I increase the gain to max, it is very quiet until the input is very high where it suddenly outputs a burst. There was a brief moment when I had disconnected everything and reconnected with low gain that it worked fine and had the correct bias, but increasing the gain suddenly pulled down the bias and the problem returned.

I am completely at a loss as to what could be causing this, what am I missing? Thanks in advance!

Edit: To add, also when the circuit was reset and had the correct bias, when i then measuring the voltage with multimeter at the non-inverting input, it suddenly dropped down to 1.6V and exhibited all the above mentioned issues.

  • \$\begingroup\$ What's the input bias current for this opamp, worst case? Just a sanity check about your 1 Meg resistor to (+)... \$\endgroup\$ Apr 2 at 20:59
  • \$\begingroup\$ @periblepsis According to the datasheet it seems to be 200pA worst case, so at worst the voltage drop from the bias current is 2•10⁻¹⁰• 10⁶ = 0.2mV, right? Edit: The input bias is the average of both so it would really be larger but it doesn't seem nearly large enough to warrant a drop of 4.6V! \$\endgroup\$ Apr 2 at 21:24
  • \$\begingroup\$ I just wanted you to do a quick check. I would have figured a higher value, just off-hand, over its full temperature range. That seems a bit low. But since you checked, I'm fine. Just wanted that removed from the table. Do you really have a 100k output load??? (R7) (I've yet to look at the topology... it's kind of messy as you show it there.) \$\endgroup\$ Apr 2 at 21:32
  • \$\begingroup\$ @periblepsis Thank you for the reminder to check that! Guitars are high impedance so generally you'll find something on order of 100k on the outputs of mainy pedals, although some pedals go lower for around 10k. Quite a few have a 100k pot on the output as a volume knob! \$\endgroup\$ Apr 2 at 21:41
  • \$\begingroup\$ How do you figure there is 34db gain in the circuit? \$\endgroup\$
    – Marla
    Apr 2 at 21:48

3 Answers 3


Table 6.8 of the TI TL072 datasheet shows that maximum output swing can be as low as ±12 V for a ±15 V supply. That implies that on a 9 V supply you might only be able to swing between 3 and 6 V.

I suspect that it's the wrong op-amp for the job.

  • \$\begingroup\$ The TL072 is probably the most common op-amp used in the guitar pedal industry, often running on 9V with a 4.5V bias so I don't think it's that! Even so, shouldn't it still produce a good signal at low gain? \$\endgroup\$ Apr 2 at 19:30
  • \$\begingroup\$ @CouchAnalysis The TL072 is an op amp from 1978; there are much better ones for that kind of application these days! I can't imagine why they're still using it like that, unless they're just using copies of copies of copies of the same exact circuits from the 80s. Which I suppose is also why hobbyists end up using the 741 so much... \$\endgroup\$
    – Hearth
    Apr 3 at 2:18
  • \$\begingroup\$ @Hearth I've actually been looking for a modern JFET replacement for the TL072! I mainly took it because it's "tried and tested" as they say but really it's been begrudgingly due to its limitations. \$\endgroup\$ Apr 3 at 5:15
  • \$\begingroup\$ @CouchAnalysis If price is no object, the ADA4625 might be worth a look. Costs about $12 each, but the specs are pretty damn impressive. \$\endgroup\$
    – Hearth
    Apr 3 at 13:18
  • \$\begingroup\$ @Hearth As we say in Sweden, "That's where the shoe pinches", more than 10 times the cost seems steep for tinkering! \$\endgroup\$ Apr 3 at 17:12

22uF bypass capacitor has 362ohm impedance at 20Hz. You can test if your Vref change a lot. If yes, maybe try bigger capacitor or add another opamp in voltage follower configuration as Vref?


I sounds like there is an intermittant open somewhere. I suspect R3 is not properly connected so that there is no (intermittant) bias on the non inverting input. When an ac input is applied of a significant amplitude then it is rectified by the non-inverting input(perhaps the internal rotection diodes). This may be enough to provide the necessary bias for operation, thus resulting in the loud burst.

This may also explain why the volume control works backwards. The signal is coupling from in+ to in- directly, then through the volume control to the output. The op-amp output resistance is high now because of the incorrect bias. This further suggests there is a problem with R3. (Or perhaps the Vref circuitry) Make certain there is no interrmintant short between Vref and ground.

  • \$\begingroup\$ Thanks for having me check that! I measured the resistance between the non-inverting input and Vref and sure enough things aren't right, I got 20.4k where it should be 1M! I then measured the resistance between the inverting and non-inverting inputs and got 19.5k. It seems my TL072 is busted :( \$\endgroup\$ Apr 3 at 5:45

Your Answer

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

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