Hey,
I'm currently designing a CODEC board for the TLC320ad77 from TI. That is my first project of this kind, so I'm having quite a hard time.
This is the datasheet:
http://www.ti.com/lit/ds/slas194/slas194.pdf
At the application section, page 25, there is an example of analog front-end circuit. In the negative feedback of both opamps, there is polarized caps of really low values (10pF). What kind of cap should it be? And is C4 electrolytic?
Thank you for your help!
I'm currently designing a CODEC board for the TLC320ad77 from TI. That is my first project of this kind, so I'm having quite a hard time.
This is the datasheet:
http://www.ti.com/lit/ds/slas194/slas194.pdf
At the application section, page 25, there is an example of analog front-end circuit. In the negative feedback of both opamps, there is polarized caps of really low values (10pF). What kind of cap should it be? And is C4 electrolytic?
Thank you for your help!
Thanks for answering.
oshifis, this is exactly what I thougth ^^.
I don't really understand the purpose of having polarized caps here (talking about C2 and C3).
Do you think I can use non-polarized caps?
yes, C4 seems electrolytic, but I can't find any electrolytic cap of 1nF
It seems hard to find so small values with tantalum caps too. Farnell min is 0.1uF.
oshifis, this is exactly what I thougth ^^.
I don't really understand the purpose of having polarized caps here (talking about C2 and C3).
Do you think I can use non-polarized caps?
tbh i didnt look at the schema, but rather just went by your description of where it was. My guess is oshifis is dead on with that, yes you can use a nonpolar cap, thats why i recommended the quality ceramic in np0/c0g type. for highest performance an RF grade mica, teflon or thin film cap should be used, but good ceramics will do just fine.
Figure 5-2 shows the correct symbol for a feedback cap. My guess would be student co-op term or hours of work directly adjacent a long weekend.
The meaning is that the capacitor with the curved plate MIGHT be or not polarized. For small ones of course they will be non-polarized, but in case that they have a polarity or a metallic case, TI showed the correct mounting of that case (usually the side with larger signal or lower impedance).
For exemple C1 and C20 MIGHT be polarized, with ground terminal (negative) at position #2. You COULD use a non-polarized one if you find one, but would be an increased cost.
The ones that don't have rounded plate, usually means that they NEED to be symetrical and non-polarized. Like a ceramic or mica one, not a rolled one...
For exemple C1 and C20 MIGHT be polarized, with ground terminal (negative) at position #2. You COULD use a non-polarized one if you find one, but would be an increased cost.
The ones that don't have rounded plate, usually means that they NEED to be symetrical and non-polarized. Like a ceramic or mica one, not a rolled one...
Last edited:
Hey again,
Maybe I should start another thread, but since it is about the same datasheet I go on here.
I'm now wondering about powering and grounds. Maybe you've seen it coming
.
http://www.ti.com/lit/ds/slas194/slas194.pdf
Page 25 and 26, opamps are powered by 5v, and wired to ground. I believe this is a "mistake". Opamps process analog signals, so they have to be wired to AGND. And their positive inputs are biaised to AVdd/2, so they must be powered by AVdd, which is 3.3v, not 5v. Am I right?
This is what TI did page 26 of this application note:
http://www.ti.com.cn/cn/lit/ug/slau057/slau057.pdf
Isolating analog and digital sections now:
I use an arduino mega, so power supply is regulated.The arduino provides me with 5v and 3.3v power supplies. 5v is my digital supply, and 3.3v the analog one.
I think I will only use one ferrite bead, placed between the two grounds. DGND will be directly wired to power supply ground, and will physically be close to it (to prevent high frequencies currents from returning into the analog part). AGND will be isolated by the ferrite bead.
What do you think?
Knowing that what I'm designing is an audio effect. So this is NOT hifi. And it will be my first pcb design with more than 10 components. So I'm not looking for perfection.
Sorry for english mistakes, everyone noticed I'm not a native ;+).
Maybe I should start another thread, but since it is about the same datasheet I go on here.
I'm now wondering about powering and grounds. Maybe you've seen it coming
.http://www.ti.com/lit/ds/slas194/slas194.pdf
Page 25 and 26, opamps are powered by 5v, and wired to ground. I believe this is a "mistake". Opamps process analog signals, so they have to be wired to AGND. And their positive inputs are biaised to AVdd/2, so they must be powered by AVdd, which is 3.3v, not 5v. Am I right?
This is what TI did page 26 of this application note:
http://www.ti.com.cn/cn/lit/ug/slau057/slau057.pdf
Isolating analog and digital sections now:
I use an arduino mega, so power supply is regulated.The arduino provides me with 5v and 3.3v power supplies. 5v is my digital supply, and 3.3v the analog one.
I think I will only use one ferrite bead, placed between the two grounds. DGND will be directly wired to power supply ground, and will physically be close to it (to prevent high frequencies currents from returning into the analog part). AGND will be isolated by the ferrite bead.
What do you think?
Knowing that what I'm designing is an audio effect. So this is NOT hifi. And it will be my first pcb design with more than 10 components. So I'm not looking for perfection.
Sorry for english mistakes, everyone noticed I'm not a native ;+).
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.