Just came across this site because of the information below. Found some valuable info and leads to help me. Thanks.

Recently smoked this chip in an analog/Digital Trainer and ordered some replacements on line, none of them worked in the existing circuit. I downloaded the data sheet for the chip and breadboarded the sample circuits and tried the chips in them, with little or no success. Eventually I found by adjusting the supply voltage, and the values of the external resistors and caps I could get them to a least partially function. I noticed that the original chip had a batch number of F0918 and the ones I received had a batch number of F0919 and were from china. I had a second Function Generator that used the same chip, so I pulled it and tried it in the unit I smoked the chip in and it works just fine. I than tried it in the test circuits from the data sheet and it also worked fine in them. I found a source for the chips stateside, called and customer service checked stock and confirmed their stock had batch number F0918. Ordered and received them they work just fine.

Bottom line here is my experience with the china version of XR2206 is the external components must be custom selected for the particular chip and will not work as a suitable substitute for XR2206cp F0918. Also, I was never able to get the FSK circuit on the china version to work no matter what I did.

I know the previous posts regarding this chip are old but since china started making this chip maybe this will save some people some time and aggravation.

 

I recall having similar problems years ago with XR2206 chips I'd bought at Radio Shack; some worked in the recommended circuits and some did not, and required certain component values to be "tweaked."

My guess is that in both cases, the chips are rejects that failed production testing-- not badly enough to be called "dead" but which nevertheless failed to meet certain spec requirements.

 

Is this a Nida trainer by any chance?? The ones we had you could adjust the trim pot on the circuit board and all oscillations would stop...

 

Is this a Nida trainer by any chance?? The ones we had you could adjust the trim pot on the circuit board and all oscillations would stop...

There is only one board mounted trim pot and is a course adj for DC balance all others are angel mounted. None have any affect on the operation of suspect chips. I duplicated the circuit on the breadboard with the same results as the sample test circuits. Problem is just with the chips.

 

Even the old Exar parts (I still have two somewhere) were a bit spotty on their operation. They were in fact part of a baseband FSK modem chipset.

https://archive.org/details/ExarFunctionGeneratorDataBook