I made some inductors for the output filtering on a tpa3116 chip. They have only 4 windings but a much larger toroid core. (The toroids are bought). They measure 18uH and the originals were 25uH. The problem now is that the amp won't come out of protection. When I solder the old ones back in its fine again.. Can someone please tell me whats wrong with my coils? Or what causes the chip to shut down?
I'm no expert on coil theory but this sounds like a combination of incorrect value (18 vs 25uH) and probably more likely, a difference in the ferrite and its permeability and saturation characteristics. If the core saturates then the current that the amp has to deliver will rise sharply and that's probably the reason why its going into protect.
I'm only making a wild guess, but could the core go into saturation or could it be too lossy at the switching frequency and its harmonics?
Bet you brought ferrite rather then iron powder cores (They are not the same thing).
Replace the cores with the right ones is about all you can do.
regards, Dan.
Replace the cores with the right ones is about all you can do.
regards, Dan.
I found out that toroids are these http://aranna.altervista.org/dragonsnest/wp-content/uploads/2014/05/1760506.pdf. After reading trough the ebay auction again it seems that they are in fact Ni-Zn alloy. The pdf indicates I should use manganese-zinc(Mn-Zn) instead. It does seem that the impedance of the rings I've bought is too low at 400kHz 🙁
Thanks for helping me figure this one out!
Thanks for helping me figure this one out!
Yeah if it only took 4 windings to reach 18uH they are definitely the wrong material. Depending on the size of the core and material, you will need at least 10x that many turns.
Here's a good online calculator to estimate number of turns for a given size and material...
http://www.66pacific.com/calculators/toroid_calc.aspx
Here's a good online calculator to estimate number of turns for a given size and material...
http://www.66pacific.com/calculators/toroid_calc.aspx
Last edited:
A toroid is the wrong shape for ferrite, and with MnZn, it will be even worse.I found out that toroids are these http://aranna.altervista.org/dragonsnest/wp-content/uploads/2014/05/1760506.pdf. After reading trough the ebay auction again it seems that they are in fact Ni-Zn alloy. The pdf indicates I should use manganese-zinc(Mn-Zn) instead. It does seem that the impedance of the rings I've bought is too low at 400kHz 🙁
Thanks for helping me figure this one out!
As a member said above, you need metal powder if you are going to use toroids, and if you stick with ferrite, you need shapes allowing you to insert a gap
- Status
- Not open for further replies.
- Home
- Amplifiers
- Class D
- Homemade inductors doesn't work..