xr2206 problems

[sakis]

i worked with this schematic and the square wave output comes out not symmetric .... the existing trimmers can adjust the sine vave curve but not the square wave .....

anybody has an idea or a better circuit based on this ic ????

thanks

[Juergen Knoop]

my ELV XR2206 function generator does a acceptable squarewave.
What is your problem with symmetry?
regards

[sakis]

1-the upper part is rounded ....
2-and then again there is no way to adjust this ( at least on this circuit )
3- scuare wave doesnt function throuthe all spectrum of freq ( sine wave was perfect )

the circuit is done as is in the book

[sakis]

that i built this for the square wave mostly ..... sine wave i have a very nice device ......

[Juergen Knoop]

it has a high output impedance, have you tried a 10:1 probe?
regards

[sakis]

this has nothing to do with it i think ..... i ve tryied more than one scope with several probes ....there is got to be something else

[Elvee]

Hi,

The problem comes from the open collector output: the high level is created passively, with the 10K pull-up.
The current in this resistor is insufficient to overcome the capacitive parasitics and the stored charge of the transistor. You have to use a resistor as low as practical, and add a totem-pole buffer to remove the effects of loading.
But in my opinion, it is rather silly to use a full-blown function generator just to generate square waves.
If you want clean, frequency-variable square waves, you can use a schmitt trigger oscillator (CD40106) followed by a flip-flop to get a symetrical waveform (CD4013) and an output buffer made 8 bus drivers paralleled (74C244). This is just an example, there are many others options that are simpler, cleaner and cheaper than the 2206.
LV

[sakis]

Originally posted by Elvee
Hi,

The problem comes from the open collector output: the high level is created passively, with the 10K pull-up.
The current in this resistor is insufficient to overcome the capacitive parasitics and the stored charge of the transistor. You have to use a resistor as low as practical, and add a totem-pole buffer to remove the effects of loading.
But in my opinion, it is rather silly to use a full-blown function generator just to generate square waves.
If you want clean, frequency-variable square waves, you can use a schmitt trigger oscillator (CD40106) followed by a flip-flop to get a symetrical waveform (CD4013) and an output buffer made 8 bus drivers paralleled (74C244). This is just an example, there are many others options that are simpler, cleaner and cheaper than the 2206.
LV

but really dont know how to do this ....is there a schematic that you know that is tetsted and working that you can point to me ????

[Elvee]

Make an oscillator like this one:
http://www.interq.or.jp/japan/se-inoue/e_ckt13.htm
eventually followed by a 4013 in divide by 2 mode if you need a perfect symetry, followed by some kind of buffer (you can use the five remaining 40106 or a supplementary IC).
Alternatively, you can use a ready made oscillator like the 4046 or 4047; with the latter you have the advantage of having the flip-flop already included.
Try to find information by yourself, it's so elementary and there are huge amounts of info on the web.

[sakis]

old thread but i am back to this cause now i need it again ...

fixed the square wave issue adopting the output scheme of this circuit http://electronics-diy.com/electroni...tic.php?id=818
then the problems are
1) i get spikes on both up or low sine ( will post pics tomorrow)

2)also it is impossible to adjust the lower wave form correctly , the lower part of the sine is out of center

any ideas ?