Dear all, I am designing a PBTL amplifier circuit around the TDA3220 chip. The datasheet gives this example circuit bellow. I have chosen for a post LC filter PBTL because it says that the inductors can be smaller. The impedance of the speaker I will drive is 2 ohms (quite low). Since I have no experience in power amplifier design my question is:
Thanks a lot!
- Are there components in this circuit that require to have some minimum specs. Like the components in the filtering circuit after the 10uH coils towards the speaker, do they need to be able to withstand the power (~60watts continuous). And if so, what specs would I need to look into and what would be safe margins to use based upon your experience.
- I have chosen post filter PBTL to reduce needed component size, but are there any hard pro's for going pre filter PBTL?
Thanks a lot!
Last edited:
Every component in every circuit has to meet some minimum spec. Any capacitor must be able to withstand the voltage across it, and every series element needs to handle the current through it. For example the input 1uF capacitors also have a voltage across them, usually need to be rated for higher than you might think.
For series inductor you can find the required current with P = ((V^2)/R) where P is rated power output and R is your load impedance. In reality the speaker impedance is not constant so you have to live with some peaking, hence practical output filters add a small zobel to the output (not very effective, but it helps a little). Typically inductors should be rated close to worst case situation, where 4 ohm nominal speakers can dip as low as 2.5 ohms.
@Sangram Thank you for the tip, didnt know that equation for the rated power of inductor. I assume this equation would be for if I use 1 coil? So that would mean I could divide this rated power by 4 since I will be using 4 inductors.
What minimal wattage rating would you advise for the 3R3 resistors in the endstage circuit? The supply voltage is 16.2V so:
16.2/3.3 = 4.1[A] -> 4.1*16.2 = 66.42[W]
In my experience this is quite a lot
What minimal wattage rating would you advise for the 3R3 resistors in the endstage circuit? The supply voltage is 16.2V so:
16.2/3.3 = 4.1[A] -> 4.1*16.2 = 66.42[W]
In my experience this is quite a lot
Well no, current remains the same as it is flows in series. Page 23 of the EVM manual says it uses 4.6A inductors, and the chip itself is rated for over 100W in PBTL mode.
Obviously, you control how much power you ask the device for (your right foot, so to speak). You can use as small an inductor as you want, with the only restriction being that if the inductor hits saturation you will hear many unpleasant things. Cheap Class D amps sound bad for a reason.
Obviously, you control how much power you ask the device for (your right foot, so to speak). You can use as small an inductor as you want, with the only restriction being that if the inductor hits saturation you will hear many unpleasant things. Cheap Class D amps sound bad for a reason.
Not to hijack, but I'm looking for guidance selecting components for a passive crossover network. If I'm working with the output of an amplifier that will never exceed 50w with a 4o nominal load (<15v, ~3.5a) are 16v capacitors and 4a inductors adequate, or should I build in a safety factor?
Capacitors need to have sufficient ripple current rating to withstand the current through them. Your sim will be able to tell you what the current through the component is. You will find that 16V capacitors will not have sufficient current ratings to pass this hurdle, and have ESR high enough to affect the crossover function. I typically target 200-250V for tweeter circuits (higher is usually better), and 100-200V for woofer shunts.