I just put together this signal generator with a xr2206 DIY kit, single supply, tried 9,6V smps first (too low!), then w/ checked speced minimum 12V DC Li-Ion Vcc, same behaviour, distortion/unstable. It drew about 30ma current with 12V, even though in the other Datasheet of Exar it was said to consume a maximum of 17-20mA (respective of version of chip), mine took about even 32mA-ish, I don't know whether my two multimeters are too unprecise, but one is a decades old digital one and one is a cheapo modern digtial one. Initially it came ootb with an unmarked/unbranded xr2206, but because of it suddenly putting out huge distortion and noise and weird unregular hum, i thought I destroyed it with back emf and/or static discharge. Even though I'm very careful of that, I thought I did, as I have little experience to be honest in designing audio circuits.

So, I then replaced it with a HLF XR2206 with markings on it, order from spain, and it turned out, both are unstable in this circuit, they just run, kind of fine, ( I guess) for like 2 minutes or so (ircc), but soon after that, it distorts, or even earlier, in three of my "analyzers" in the DAC USB interface (192khz samplingrate ac coupled 4,7uf 400V MKP) as well as in a seperate test with 3,5mm jack input to an old stereo (2x mono) loudspeaker (it has internal 50hz hum/noise -regardless I could hear/adjust the xr2206's sqr/sine/tri signals-, ac coupled w/ 4,7uf 400V MKP) just to hear what it sounds like. Then I went on to search for an oscilloscope, found one, but located too far away for me to be comfortable to pick up, but then the LORD helped me to get my hands on a used 465B oscilloscope, in my hometown, basically unused for the past 20 years, mostly in tiptop shape from outside and working, and just one of the two X10 (original with BNC pin) probes was faulty, and the sellers weren't even able to turn it on. Nice. So I do have mixed oscilloscope snapshots, of some very good looking sinewaves and most of the bad I missed to take a picture of, but I do have something.

Tldr my Q. is:

  1. Does anybody have any Idea why this circuit is unstable? Huge noise, intermittently no regular recognisable signal. Got three chips, two different brands, 2 breadboards, we can build different circuits.
  2. Is this even worth to find out, or do you guys use a proper bench FG? I really like the concept, because it's small and simple parts.

If I'm allowed to here, here are pics from the assembled circuitenter image description here. enter image description here

  • \$\begingroup\$ Idea why this circuit is unstable? ... what circuit? \$\endgroup\$
    – jsotola
    Dec 30, 2023 at 19:57
  • \$\begingroup\$ Oh, I'm sorry for not mentioning it, it's the assembled physical printed circuit board, as shown in the added two images above. It's parts came almsot entirely except one nut, from a common commercially available DIY kit, that one can solder together, and it's quite easy. And so I did, but it's unstable. Does anybody have any idea why the xr2206 doesn't run stable? I just use it for like max 100khz, mostly (if it worked) I'd use it for 20hz-20khz. \$\endgroup\$
    – rantoniusw
    Dec 30, 2023 at 20:14
  • \$\begingroup\$ electronics.stackexchange.com/questions/485625/… \$\endgroup\$
    – Mattman944
    Dec 31, 2023 at 1:39
  • 1
    \$\begingroup\$ You might find the YouTube video Fixing the issues with the XR2206 Function-Generator from EBay useful. \$\endgroup\$ Dec 31, 2023 at 16:50

3 Answers 3


I built one of these 5 years ago. It is poor quality, so I never use it. Do function generators produce AC signals in the sense we know?

Your parts seem to be installed correctly. The electrolytic caps are oriented correctly. The value on R4 is hard to read, it could be wrong, but that won't affect the triangle or square wave outputs.

Current draw with 9 Volts is 12 mA, with 12 Volts it is 16 mA. Your current draw is higher, this could indicate that your ICs are bad.

I changed R4 to a trimpot to match the IC vendor recommendation.

The amplitude pot is backwards, clockwise decreases the amplitude. Be sure to move this pot away from clockwise, clockwise is minimum!

The coarse and fine adjustment pot values are only different by a factor of 2, 5X or 10X would make more sense.

enter image description here

This is the schematic I received with it. It is basically the same as recommended by the IC vendor.

enter image description here


The sine and triangle waves are offset from zero.

enter image description here

  • \$\begingroup\$ Yes, ICs got corrupted, probably by me, I ordered a new one marked HLF brand, but one additional changes was needed to get it right. \$\endgroup\$
    – rantoniusw
    Apr 3 at 8:32

You’ve bought a cheap kit based on a long obsolete chip, pretty much any of which you find today will be counterfeit. The kits themselves are generally poor quality as well, some even have pots that work backwards. There are numerous posts online with people having problems with these kits. You might just take it as a ‘buyer beware’ learning experience.

  • \$\begingroup\$ Is there even a pot "backwards" and "forward"? Does this description even exist for potentiometers? \$\endgroup\$
    – rantoniusw
    Dec 30, 2023 at 20:50
  • \$\begingroup\$ @rantoniusw When you turn a level control clockwise do you expect the level to increase or decrease? \$\endgroup\$
    – GodJihyo
    Dec 30, 2023 at 21:34
  • \$\begingroup\$ Don't they go backward in the UK? :) \$\endgroup\$
    – QTX
    Dec 30, 2023 at 22:28
  • \$\begingroup\$ It's irrelevant, of what I expect of potentiometers, even if I did expect sth., still I don't do determine it's specification. And therefore also this misses the original q and the additional. \$\endgroup\$
    – rantoniusw
    Dec 30, 2023 at 23:02
  • \$\begingroup\$ I have an idea, kinda obvious, but will try to post the results later here for you, perhaps a week or so, on the other side, Happy New Year in the LORD. Since I'm tidying up and rearranging my place atm and much is in the basement already. \$\endgroup\$
    – rantoniusw
    Dec 30, 2023 at 23:49

It works now but, to tell you shortly what got it working.

Some pointed out that it is not worthy because of poor quality, well, I disagree strongly. The built quality seems excellent, with minor "inconveniences" so to speak, but for that it was dirt cheap.

I was able to get 4th chip to work. All I had to do was using a regulated power supply. But the SMPS didn't work out, it doesn't like those supplies.

In the attached image you can see, that the sine wave looks like a smoothed triangular wave. That's the signal quality of the IC, not the overall quality of the physical built. Yes I would like to have a smoother more curved sine wave, but this should work too for problem solving and testing.

This was a follow up for other people who might struggle with this cheap intro beginner OX and want to see it run.

enter image description here


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.