DC on output

[G4ME]

Okay so i have a working standard triangle wave/comparator amp with an IR2010, and some IRF540's.

The only problem is I notice DC on my output. I have looked at my output of the comparator with no input signal, and my understanding is the duty cycle should be 50%. Right now mine is 45% and does not change much more then the Oscop can measure. Thinking this through it must mean my triangle wave is not perfectly balanced and has a DC offset on it as well and i measure +6.875V to -5.813V. I am using Exars XR-2209 VCO. My voltage rails are not balanced as i am using LM7812 and LM7912 rectifiers, they are 12.31V and -11.60V.

http://www.datasheetcatalog.com/data...2/XR2209.shtml
link to the VCO

Is there anything I can do to remove that dc on the triangle wave and thus remove my output DC, as i kinda don't want to fry my voice coils.

Am I screwed till i get better rails?

Thanks

[mag]

Looking at simplified schematics on the XR2209 datasheet seems that pin 5 (bias) sets the offset of the output op-amp for the triangle wave output. The offset on your triangle wave
How did you connect pin 5 on your design?. The power supply imbalance should not affect the triangle wave offset.

Try to change the voltage on pin 5 trough a resistive divider (or a trimmer) and see if the offset on the triangle disappears.

I have never used that chip so I cannot tell you if it works...

[G4ME]

as of right now I have it tied to ground, i will see if i can get a pot in there to see if it changes anything.

[G4ME]

well i got some 10K multi turn pots and using a unity gain inverting op amp with the pot rigged up as a voltage divider with the rail i am able to take that dc offset to pretty much nothing, and the resulting PWM out of the comparator with no input is 50%

[mag]

good,
I don't think you need an op-amp after the pot.
Try to connect the pot direcly to pin 5. Connect also a 100nF capacitor between pin 5 and GND as a decoupling.

BTW which voltage have you forced on that pin to reach 0VDC offset? Is that something around 0.6-0.7V??

[G4ME]

I initially tried to connect the pot to pin 5, but seeing I needed to add a negative DC offset, somehow when pin 5 became negative it forced the output to get all funky and messed up. I think pin 5 is really only there to add DC bias not subtract it.

So I had to sit and stew about it for a little while and then came up with the op amp and pot configuration. Yeah its one more thing in the signal but I am using a OPA4227 for the input buffer and it had three free op amps left.