Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Questions about the XR2206 :-)..

Status
Not open for further replies.

SorensenDK

New Member
Hi there I'm new in this forum and hope for help with my problem :).

I need a 'perfectly symmetrical' and frequency stable triangular signal at 100 kHz. XR2206 allows excellent, but is somewhat temperature sensitive, therefore, .. Is it possible to get XR2206 to work with a 100 kHz X-tal, I only need this one frequency. Has anyone here tried it, there looks like ?

regards Jan with hope :).
 
I think if you add a xtal to the XR2206 the triangle wave will look bad. If you said square wave I would have given that a try.
The XR2206 should work with a PLL. The PLL will compare the XR2206 square wave against a 100khz xtal and then tug on the XR2206 sweep input to pull it over to the right frequency.

If you don't know about PLLs I would start out reading the data sheet on the CD4046 or MC14046.
 
Hi ronsimpson and thank's for answer :)

If I look at the chart of XR2206, then pin 5 and 6 is 2 collectors and pin 7 is a common emitter point, the question is whether I are in some way to get a X-tal to 'swing there'?
I do not think it will affect the triangle curve... If it will lock.
Ok with an external input, here I can also just give it a signal from an external X-tal oscillator on 100 kHz, if I knew what pin on 2206 I should use?

Or use some more temperature stable 'R / C' and put it all in a box also an opportunity perhaps ?

Jan :)
 
Last edited:
Hi,

I would rather use a crystal controlled square wave oscillator and ramp generator. Simpler and you can get a nicer waveform using a good op amp.
 
@MrAl : Tank's for answer :).

Yes also an option, or a 1MHz X-tal square wave oscillator plus a divide-by-10, plus an op-amp as integrator for producing the triangles, perhaps it could also be a solution?

But I have done XR2206 more stable (measured by a SSB beat-tone directly on 100 kHz). Replaced with a 'Micro Poco' capacitor (pin 5/6), resistor replaced with a metal-film type (pin 7)and put it all in a box.
Every time I turn it on, landing it on the same 'beat-tone frequency' and stand there for 'looong time', so it is also a kind of solution (XR2206 can do it)

Note: The installation will not be exposed to extreme temperature fluctuations during operation.
PS: I love XR2206 because it's make some beautiful triangles :D..

Jan :).
 
Hi Jan,


Oh you mean the XR2206 puts out good triangles too? Well that's good then. I wasnt sure if it would be good enough for your application. I know the sine waves are made using a nonlinear shaping circuit so might not be that good at higher frequencies.
 
Pins 7 or 8 can be used as a voltage-controlled VCO input (see Fig. 10 of the EXAR data sheet). You could feed the output of a phase detector to the XR2206 input with one detector input being the EXAR square-wave output and the other from a digital output crystal oscillator.

An XOR gate with a RC filter on its output makes a simple digital phase detector to analog output. The CD4046 PLL circuit also has two different types of phase-detector circuits you can independently use.
 
Last edited:
@MrAl :

My comment about the triangle on 2206 whas most for fun, but remember the 'sine kurve' is generated from the triangle in the 'sine shaper' and I don't use the sine waves, so no prob there..

@crutschow:
Okay and thank's, An interesting solution I also must try.

I would test several solutions against each other and the best wins.

Jan :).
 
Last edited:
Use a PIC with a 4MHz crystal and make a 100kc/s square wave with an infinite loop on one output pin.
Filter that wave with a low pass filter (RC = 200us, R = 4k7 and C = 47nF).
Amplify the resulting signal but do not use any OpAmp. Use a HF OpAmp or make a discrete gain amplifier.
Another solution.
Your square signal can be applied on two transistor current sources. Those current sources, one positive and other negative, have a capacitor as load. The signal obtained will be a good triangular wave with the same precision of the crystal.
You can adjust current sources to generate the amplitude that you wish.
Don’t use capacitors as base time. Its temperature dependency is very bad.
MOR_AL
 
Thank's for info to all :).

Now I have something to do for several hours with testing of options.

I'm not good at / do not have equipment to PIC programming, so that opportunity, I regret to exclude.

I would also think about my next step. A comparator that can handle music to my 100 kHz. PCM project. (hence my desire for neat triangles :D).

Thank's Jan :)..
 
................................
I would also think about my next step. A comparator that can handle music to my 100 kHz. PCM project. (hence my desire for neat triangles :D).
Do you mean PWM for audio? If you had told us that earlier we would have given different advice. ;)

For audio PWM you don't need an accurate frequency -- accurate triangles or sawtooth yes, but not accurate frequency. A slow drift in frequency has no affect on the PWM output. There's no need for it to be crystal controlled.
 
@ crutschow: :)
PCM (Pulse Code Modulation) or PWM (Pulse Width Modulation), what is the ´difference in this case? But okay we call it PWM because this is the most correctly.

You have right, the frequency stability is not importan if I use the 100 kHz. signal directly, but I mix it up to about 1.900 MHz. (160 meter radio amateur band), therefore I also must have good stability on the 100 kHz. signal.
The trick is. Now all my buffers/ drivers and PA can work directly in class 'C' and 'transport' my PWM signal with all the distortion they like forward to the final stage.. An effective LP filter on the signal frequency there make it / transform it to a complete AM sgnal.

I have try to make the PWM directly on 1.900 MHz. but it's requires very special and good / expensive 'OP-AMPs' up to the GHz. area for making a good triangle and comparator, so here I stop :eek:.

This is my plan / project... Jan :).
 
PCM is Pulse Code Modulation which is typically used to convert an analog audio signal into a series of digital words for transmission and/or storage of the signal (used in CDs, DVDs and transmission of digital audio).

PWM is Pulse Width Modulation which is typically used to efficiently control or vary power to devices such as motors and audio speakers. It is not normally used to transmit or store information.
 
...
... if I use the 100 kHz. signal directly, but I mix it up to about 1.900 MHz. (160 meter radio amateur band), therefore I also must have good stability on the 100 kHz. signal.
The trick is. Now all my buffers/ drivers and PA can work directly in class 'C' and 'transport' my PWM signal with all the distortion they like forward to the final stage.. An effective LP filter on the signal frequency there make it / transform it to a complete AM sgnal.
...

If I understand you right, you are making PWM at 100kHz, then "mixing it up" to 1.9MHz, then "transporting" the PWM at 1.9MHz, power amp at 1.9MHz, receiver etc, and after all that you expect you can filter the 1.9MHz PWM to get HiFi quality? :eek:
 
Hi again :).

I will try to explain:
We only talk about AM signal with a highest modulation frequency up to 7.5 kHz. It must match the receiver bandwidth.
We also talk about, we only can use 25 - 50% duty cycle of the square pulses (50% duty cycle gives the highest energy).

The distance between the positive going edges kept constant and only the pulse width varies with the 'music'.
The secret is the LP filter between PA and the antenna, here converted the varying 'pulse width of energy' to an AM information.

If you to day make a modern AM transmitter (rarely happens), so you use PWM because they individual not linear class C RF stage gives a better efficiency of the effect.
Must of all, you save the '50% of the PA effect AF modulator', a god idea especially when we talk about transmitters in the Kilowatt class.


http://www.transmittersrus.com/radica/64/manufacturer

http://hawkins.pair.com/wbbr.shtml

http://www.youtube.com/watch?v=_pz_8cY1f5g

http://www.youtube.com/watch?v=H4ClZHcEWaw

More on Google 'PWM AM on transmitters'... Jan :)
View attachment 65964 View attachment 65965
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top