Code Oscillator using a (nowadays fake/counterfeit) XR-2206 (EXAR) function generator chip

104 views
Skip to first unread message

df7t...@gmail.com

unread,
Jul 26, 2022, 2:07:54 PM7/26/22
to iCW - internet CW
Hello ... just to not invent the wheel a second time:

-->  Has someone already applied/published a code oscillator using  the XR-2206 chip?

The XR-2206 has an Amplitude Modulation Input (PIN 1) which may be (...well acording to my thoughts) used as a keying input and which would allow to have a user defined rise time of a keyed sine wave Morse Code signal.

The fake chips available today (e.g. from HLF, China) have some limitations compared to the original EXAR part, but should work in such an application.

Anyway – I ordered the components and will keep you informed about any progress on this project.

TNX
CU 73
Tom DF7TV


HLF-XR2206CP.jpg

S. Steltzer

unread,
Jul 26, 2022, 7:33:09 PM7/26/22
to 'joe living' via iCW - internet CW
Yes, keep us informed by all means.

73 Tom, cul.
Steve, WF3T


Message has been deleted

df7t...@gmail.com

unread,
Jul 27, 2022, 4:02:33 AM7/27/22
to iCW - internet CW
...it seems that I inadvertently  deleted parts of my original message:

      --> So – I just asked if someone already applied the XR-2206 chip as a Code Oscillator.


As you may see from the data sheet
 

there is PIN 1,  the Amplitude Modulating Signal Input.

Figure 6 shows the Normalized Output Amplitude
versus DC Bias at AM Input (Pin 1)

The idea is to use this input as a keying input. Rise (Fall) time of the output signal may be influenced by suitable
components of an RC-network via which the voltage at Pin 1 is changed (key on / off).

The available fake chips of the XR-2206 (e.g. by HLF, China) have a supply voltage range of only 9V to 12V.
The original (EXAR) part allowed 10V to 26V as supply voltage.

There is also a nice XR-2206 application note available:
  


CU 73
Tom

df7t...@gmail.com

unread,
Jul 31, 2022, 5:25:11 AM7/31/22
to iCW - internet CW
Hello,

here are some informations on a first (ugly built) test circuit:



...and here you may find a recording of a test message received from the Mumble Test Server:



As I continue to work on this project, I may add a section to my home page at QSL.net ( https://www.qsl.net/df7tv/morse-code-circuits.html ).

The information in that section will be more up-to-date and
files mentioned in this message
may be no longer be available.

CU 73
Tom

df7t...@gmail.com

unread,
Jul 31, 2022, 5:56:24 AM7/31/22
to iCW - internet CW
As you may remark if analyzing or listening to the .wav file there is a remaining amplitude between the code elements.

I did not have available the 100 Ohm trimmer-R originally foreseen between the two 820 Ohm resistors at the right side of the schematic.

Using such a trimmer should allow  to reduce the amplitude between code elements to zero or at least to a much lower level.

Such a trimmer connected to the Amplitude Modulating Signal input (PIN 1) of the XR2206CP will be used to reduce the amplitude for the key "OFF" state.

The present circuit shows the following characteristics:

pitch about 650 Hz
rise time about 5 ms
fall time about 7 ms

CU 73
Tom

df7t...@gmail.com

unread,
Jul 31, 2022, 9:01:35 PM7/31/22
to iCW - internet CW
UPDATE: I finally found a trimmer-R to be connected to PIN1 and have been able to reduce the audio output at key "OFF" substantially (near zero).

I now don't hear any oscillator signal between the code elements.

Attached is a present recording, made locally using Voicemeeter Potato.


CU 73
Tom
CO2206-DF7TV-without-audio-between-code-elements.mp3

S. Steltzer

unread,
Jul 31, 2022, 9:11:27 PM7/31/22
to 'joe living' via iCW - internet CW
Sounds good Tom!

Chuck Vaughn

unread,
Aug 1, 2022, 6:29:31 AM8/1/22
to iCW - internet CW
Hi Tom
thanks for all your work on this circuit !

i have recorded your file and added a SCOPE and SPEC-AN
looks like about -54 db of signal is leaking through...  
(see attached mp4)

thanks again Tom for your efforts to help us all out on this..


--

---
You received this message because you are subscribed to the Google Groups "iCW - internet CW" group.
To unsubscribe from this group and stop receiving emails from it, send an email to i_cw+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/i_cw/f2cd8a73-278a-4f08-859b-adcc19c9f6d6n%40googlegroups.com.
2206test1-2022-08-01_05.24.25.mp4

df7t...@gmail.com

unread,
Aug 1, 2022, 6:51:14 AM8/1/22
to iCW - internet CW
Hello Chuck,

Thank you very much for the analysis done!

Leakage of -54 dB sounds okay to me; seen the data sheet of the original chip that's about the lowest achievable value.

So far, I did not work on the distortion; but there are simple measures (trimmer-Rs at PIN13/14 and at PIN 15/16 to opitimize THD.
They are described in the data sheet in detail.

Some of the component values of my first test circuit were just chosen because I had them available.
I have a lot of the (blue) 10 uF electrolytic capacitors :)

As time allows, the circuit will be further optimized.


CU 73
Tom

Chuck Vaughn

unread,
Aug 1, 2022, 7:13:37 AM8/1/22
to iCW - internet CW
Thanks Tom for your comments on the video demo...
i wonder,  if by adding an   "EXPANDER" circuit,  it would ZERO out the bleed through totally ...

complicated, but in theory,  i am curious how it might function & perform,  for that purpose

df7t...@gmail.com

unread,
Aug 1, 2022, 8:34:25 AM8/1/22
to iCW - internet CW
Thanks Chuck for the link to the  ...Noise-Gate... article.

I am sure that some kind of SQUELCH (MUTING/QUIETING) circuit could be added to achieve a "real zero" ouput amplitude between code elements.
Seen the XR2206, a squelch threshold in the range of about 2 mV_pp may make sense.

To avoid sharp edges of the audio output signal when the gate circuit (SQUELCH) switches on/off, an RC-network may be used to get smooth transitions to/from zero output signal.

When using the Amplitude Modulating Input (PIN 1) for keying, like I do, a very well regulated and stabilized power supply for V_CC is needed.
A change of V_CC (without readjusting the trimmer-R at PIN 1) will unavoidably increase the output amplitude in the key "OFF" state.

CU 73
Tom
Message has been deleted

df7t...@gmail.com

unread,
Aug 2, 2022, 10:38:29 AM8/2/22
to iCW - internet CW
Ideas for a second test circuit using the HLF XR2206CP

  1. [PIN 2] Change the output voltage divider to 2k2 / 2k2 instead of 4k7 / 4k7, change the decoupling capacitors from 10uF to 1uF (film-type). This will lead to a shorter time needed to get the capacitors to the DC voltage present at PIN 2.
  2. [PIN 3] Add an resistor in parallel to the amplitude potentiometer to avoid clipping of the output signal when approaching the full turning range; or use of a  25k potentiometer instead of the 50k one.
  3. [PIN 4] Add a quality 10 V or (preferred, if the circuit behaves well) 9 V  voltage regulator for V_CC.
  4. [PIN 5, PIN 6] change the timing capacitor to 22 nF.
  5. [PIN 7] replace the fixed timing resistor by a series combination of a fixed resistor of 36K and an audio-taper (type "A") logarithmic potentiometer of 100k. In combination with the new timing capacitor this will allow to set an output frequency from about 400 Hz to about 1200 Hz.
  6. [PIN 13, PIN 14] Adjustment of THD: Replace the 220 Ohm resistor by 300 Ohm and add a trimmer-R of 1K in parallel (see APP-note https://www.qsl.net/df7tv/datasheets/xr-2206-app-note-tan-005.pdf , Figure 1.)
  7. [PIN 15, PIN 16] Adjustment of wave symmetry/THD: Insert a trimmer-R of 25k and a fixed resistor of 1k from the wiper to ground. (see APP-note https://www.qsl.net/df7tv/datasheets/xr-2206-app-note-tan-005.pdf , Figure 1.)
  8. Use of 1% metal-film resistors throughout the whole circuit.

I do not know when I will be able to do that, but I'll post a message as soon as done.

CU 73
Tom

S. Steltzer

unread,
Aug 2, 2022, 3:17:55 PM8/2/22
to 'joe living' via iCW - internet CW
Sounds like you are going to have a perfect unit there when you get done. 


df7t...@gmail.com

unread,
Oct 7, 2022, 3:51:42 PM10/7/22
to iCW - internet CW
Hello all,

Since my last message, I have been improving the XR-2206 setup and it worked nicely for some iCW QSOs.

Please find the present schematic at:

                       https://www.qsl.net/df7tv/download/DF7TV-XR-2206-Second-Test.jpg


Some remarks:

1.) The supply voltage V_CC has been reduced to 9 V in preparing the addition of a standard voltage regulator, which may be driven by 12 to 15 V.

2.) The circuit for keying the Amplitude Modulation Input Signal (PIN 1) has been changed.

3.) The rise/fall times of the circuit shown are at about 8 ms each.

4.) Trimmer resistors R_S (PIN 16) and R_D (PIN 14) have been added and are adjusted for minimum THD.

5.) The output to Sound Card 2 (in fact the input to Mumble) now is by an audio transformer

6.) The harmonics of the output signal are down by about 55 db. That's not much but may be improved
by a low pass filter in future. The capacitor C_R parallel to the secondary winding (inductivity about 290 mH) of the audio transformer already achieves a slight improvement in getting harmonics attenuated.

7.) The output "To Audio AMP (M.T.) -- for the monitor tone -- is fed to a TEA2025B audio amplifier not shown in the present schematic.

8.) The simple circuit for keeping the carrier down (down to about - 55 dB) at "KEY OFF" around R_CS works fine.


CU 73
Tom DF7TV
Reply all
Reply to author
Forward
0 new messages