I didn't find a lot of information out there on this topic. How can I switch out a 30uH inductor for a ferrite bead? I did a little searching and it looks like the low pass capacitor going to ground becomes quite a lot smaller (instead of around 1uF after an inductor, with the ferrite bead that value becomes 0.001uF). But is that all? Are there any other considerations (for it to function)? I did see in the datasheet that layout runs must be compact to the chip. I also saw that it must be able to carry enough current and must go to around 120 ohms at 100MHz.
Ferrite beads are usually the lowest cost option (well, no filtering at all is even cheaper) for short wiring in boom boxes not for HiFi applications with loudspeaker cabling. Please look up the documents by the class D chip manufacturers. Like these:
https://www.infineon.com/dgdl/Infin...N.pdf?fileId=5546d46264a8de7e0164b359d78016f6
https://www.ti.com.cn/cn/lit/an/sloa119b/sloa119b.pdf?ts=1745650749281
https://www.ti.com/lit/an/slyt198/slyt198.pdf?ts=1745666376122
If you now have inductors it makes little sense to use ferrite beads instead of inductors. Unless you want extra EMC/RF to filter out in other devices that is.
https://www.infineon.com/dgdl/Infin...N.pdf?fileId=5546d46264a8de7e0164b359d78016f6
https://www.ti.com.cn/cn/lit/an/sloa119b/sloa119b.pdf?ts=1745650749281
https://www.ti.com/lit/an/slyt198/slyt198.pdf?ts=1745666376122
If you now have inductors it makes little sense to use ferrite beads instead of inductors. Unless you want extra EMC/RF to filter out in other devices that is.
Last edited:
My goal was to hear the difference. I noticed ferrite beads in a Bose device that used a TPA3118 and thought the sound was good in the format. I would like to A/B with real speakers using the bead and an inductor.
Maybe the Bose device had an adequate PSU. I have built many linear LDO and uLDO PSUs for class D amplifiers during COVID. Nearly all of them (13 pieces) made the amplifiers sound better than with their original SMPS.
Do what you like but make sure not to build an RF transmitter. The output filtering is meant to filter that out. Judging that by ear…
The beloved MA12070 (better sounding than TPA3118) is an exception in a few ways. When its power supply and outputs are not filtered well enough combined with less optimal PCB layout it may distort itself noticable by ear.
Do what you like but make sure not to build an RF transmitter. The output filtering is meant to filter that out. Judging that by ear…
The beloved MA12070 (better sounding than TPA3118) is an exception in a few ways. When its power supply and outputs are not filtered well enough combined with less optimal PCB layout it may distort itself noticable by ear.
Last edited:
In the Bose case, it used a 12V supply of 3 18650s built into the speaker (model name Revolve).
I need to hear the MA12070. Interesting on the power supply note. For the TPA3118, I usually use a laptop brick. I need to test with a standard vintage style power supply (Transformer, FWB rectifier, and caps) to compare the difference. I suppose a good battery supply (like what the Bose uses) may be an even better power supply to use as the benchmark of what a good supply should sound like.
I need to hear the MA12070. Interesting on the power supply note. For the TPA3118, I usually use a laptop brick. I need to test with a standard vintage style power supply (Transformer, FWB rectifier, and caps) to compare the difference. I suppose a good battery supply (like what the Bose uses) may be an even better power supply to use as the benchmark of what a good supply should sound like.
They seem to love very low Zout over a wide bandwidth regulated linear PSUs. The crème de la crème of linear power supplies.
Like Tombo56’s R25 or simple LDO regulator based PSUs. Overdimensioned transformers, ideal rectifiers, CLC, RF filtering. The works. Stuff done well in detail.
Batteries are not making this planet any better.
Like Tombo56’s R25 or simple LDO regulator based PSUs. Overdimensioned transformers, ideal rectifiers, CLC, RF filtering. The works. Stuff done well in detail.
Batteries are not making this planet any better.
Last edited:
You cannot directly replace an 30uH inductor by some ferrite beads. In that case output caps have to be changed as well. You should refer to TI applications.
And I do not see any reason to do so. There is no point in doing DIY aiming at the cheapest solution. You will not compete with chinese nit pickers. DIY only makes sense for me looking for the best (not the most expensive) solution.
And I do not see any reason to do so. There is no point in doing DIY aiming at the cheapest solution. You will not compete with chinese nit pickers. DIY only makes sense for me looking for the best (not the most expensive) solution.
Yes, that's what I had mentioned in the original post. But I wasn't sure if it was as easy as changing the inductor and cap. I'm hoping that if I go from the uF (maybe 1uF can't remember) to 0.001uF that it will be sufficient to at least allow it to function.
It sounds like quality will not be improved. That's good to know. I'm still curious to hear it since I haven't tried the experiment.
It sounds like quality will not be improved. That's good to know. I'm still curious to hear it since I haven't tried the experiment.
Experiments trying to improve matters as described in the manufacturers documents possibly may give more satisfaction.
Or is it not about quality but lowest cost? There are still many experiments possible in that direction. Omitting stuff like the Chinese no name manufacturers do for example. Laptop bricks can be replaced for lower cost stuff, coils replaced for beads, film caps for ceramic caps, ceramic in anything below X7R, electrolytic caps for lower quality etc. You could even try to have no output filtering at all so not even beads.
Or is it not about quality but lowest cost? There are still many experiments possible in that direction. Omitting stuff like the Chinese no name manufacturers do for example. Laptop bricks can be replaced for lower cost stuff, coils replaced for beads, film caps for ceramic caps, ceramic in anything below X7R, electrolytic caps for lower quality etc. You could even try to have no output filtering at all so not even beads.
Last edited:
I wouldn't put Bose in the Chinese junk category. Sure, it's probably made in China, but at least here in the US, it's considered pretty decent consumer quality audio gear. I remember the first time I heard the Revolve, I was pretty impressed for something that was smaller than a bowling pin and ran on a built in battery.
Seeing the guts of the Bose Revolve and no inductors got me wondering. Then I saw in the datasheet a mention of ferrite beads as an alternative. I wanted to hear the difference for myself.
I've heard the difference between electrolytics in the signal path vs good film caps. That's an audible difference and no way I'd chose electrolytics in most cases. For example, in a preamp where signal is passed one stage to the next. There are also places where I don't think it makes a difference. For example on a guitar amplifier on a 12ax7 cathode, I don't hear any difference going between a 4.7uF electrolytic and the same in a film cap. In my opinion, the more expensive film cap wouldn't be worth the cost in this location for this application.
So at least to me, this type of experimentation is valuable.
It was what I consider to be a decent product (the Bose Revolve powered speaker) that piqued my interest to hear the ferrite bead vs an inductor. Maybe it will be similar to going from a good film cap to a bad electroyitic in a hifi scenario where the difference is very audible.
Seeing the guts of the Bose Revolve and no inductors got me wondering. Then I saw in the datasheet a mention of ferrite beads as an alternative. I wanted to hear the difference for myself.
I've heard the difference between electrolytics in the signal path vs good film caps. That's an audible difference and no way I'd chose electrolytics in most cases. For example, in a preamp where signal is passed one stage to the next. There are also places where I don't think it makes a difference. For example on a guitar amplifier on a 12ax7 cathode, I don't hear any difference going between a 4.7uF electrolytic and the same in a film cap. In my opinion, the more expensive film cap wouldn't be worth the cost in this location for this application.
So at least to me, this type of experimentation is valuable.
It was what I consider to be a decent product (the Bose Revolve powered speaker) that piqued my interest to hear the ferrite bead vs an inductor. Maybe it will be similar to going from a good film cap to a bad electroyitic in a hifi scenario where the difference is very audible.
Ferrite beads may be used when length of wire between amplifier and loudspeaker is very short - as in Bose Revolve.
Probably the portable BT loudspeaker has DSP onboard or such digital trickery correcting the cheapness. The JBL stuff also sounds nice at first glance. It is not of HiFi quality but portable plastic stuff for the beach. Normally they also excel in boominess and listening fatigue. They are usually not the devices one sits down for to listen to an album like with HiFi stuff.
The experimentation is good, I started with that when I was 8 years old and knew nothing of what I was doing. Then opamps and (electrolytic) caps were the things that changed the sound quality considerably. It was pretty clear that for instance ceramic caps would not cut the cake when wanting best quality. It also became clear that wanting 5% improvement costs twice the price and triple the time 🙂
The experimentation is good, I started with that when I was 8 years old and knew nothing of what I was doing. Then opamps and (electrolytic) caps were the things that changed the sound quality considerably. It was pretty clear that for instance ceramic caps would not cut the cake when wanting best quality. It also became clear that wanting 5% improvement costs twice the price and triple the time 🙂
Last edited:
And 10x the space. For example Solen caps. And yes you're right about it (the Bose) being portable and not necessarily ideal. I do think the TPA3118 shines especially well in the portable category. 12-24VDC, and very efficient, means it is very portable.
When posting it could be a good idea to mention the desired application as that is pretty much determining how stuff is made. Portability is not a standard parameter in HiFi. Many here will react in the field of best performance.
BTW good stuff does not need to be extremely large. The excellent Wima MKS2 in 5 mm are about as large as electrolytic caps. The recent news is that these will be discontinued mid 2025. That will hurt many DIYers.
BTW good stuff does not need to be extremely large. The excellent Wima MKS2 in 5 mm are about as large as electrolytic caps. The recent news is that these will be discontinued mid 2025. That will hurt many DIYers.
Last edited: