Hello, I'm searching for a high quality TPA3255 output filter inductor as classic, non SMD type.
Wurths 7443631000 and 7443630700 are SMD only as far I can see.
Are there any good alternatives in the 7-10 uH range?
Wurths 7443631000 and 7443630700 are SMD only as far I can see.
Are there any good alternatives in the 7-10 uH range?
Bending? Sorry, that's not working for me.
I found the Sagami 7G14J 100M.
20A rated and good measurements ... but no shop/seller.
I found the Sagami 7G14J 100M.
20A rated and good measurements ... but no shop/seller.
https://www.mouser.de/ProductDetail/BI-Technologies-TT-Electronics/HA55-2223070LF?qs=qjs3DF31iJwSGpht6k%2Bn9w==&srsltid=AfmBOooU7GWjN0ifTCVqq8pyX_Ar0PdG0HdzLBtGTco1d072Ovf0Xefa
BTW things depend on the PCB. If you design a PCB better add a few footprints of various coils over each other to make trying out various coils the neat & tidy way possible. There are practically no TH coils anymore and if you would find them you likely will find not fitting ones and by using these you possibly could have too long lead wires for optimal performance. Medicine would be worse than the disease aka antennas making stuff worse and not better. Also extension wires are a real no no.
I found these to be pretty OK as not all parameters are stellar and of course they will need to fit physically:
https://www.ebay.de/itm/325753580138?_skw=epcos+drossel&itmmeta=01JDHA7TR1XN0RST058S11BBJS&hash=item4bd86d2a6a:g:zfQAAOSwcJhkjs6v&itmprp=enc:AQAJAAAA0HoV3kP08IDx+KZ9MfhVJKnXzNVvUvY/mzVqDAX9wTOjAWwfWs9+NPMbhbBzsCsrKBJUopm2P7IhNOvZphlVQWCCrMFuIGpMgQVyDvj99MkIaNbosnsug1eMz5LO5G1Gy0tsYfHMC6Rarb30x1Gw2st41VHBt6MGba/HFXtzyoVS5gtPXCDCiEjr4+XU3BhuFOwPqw5XjkEoF7be2hAerJN11kPDom2VPmhS2aLzyn8D8wognjAdN2oKSCV41bt0Rp46HfIZ/oQNC+Kp6/aDwZY=|tkp:Bk9SR5asn6rsZA
Bending copper is one of the basic skills of any person working or hobbying in electric/electronic matters. Copper is weak and can be bent easily without penalty (unless you do it 20 times on the same spot). So with that particular coil it works for literally everybody.
BTW things depend on the PCB. If you design a PCB better add a few footprints of various coils over each other to make trying out various coils the neat & tidy way possible. There are practically no TH coils anymore and if you would find them you likely will find not fitting ones and by using these you possibly could have too long lead wires for optimal performance. Medicine would be worse than the disease aka antennas making stuff worse and not better. Also extension wires are a real no no.
I found these to be pretty OK as not all parameters are stellar and of course they will need to fit physically:
https://www.ebay.de/itm/325753580138?_skw=epcos+drossel&itmmeta=01JDHA7TR1XN0RST058S11BBJS&hash=item4bd86d2a6a:g:zfQAAOSwcJhkjs6v&itmprp=enc:AQAJAAAA0HoV3kP08IDx+KZ9MfhVJKnXzNVvUvY/mzVqDAX9wTOjAWwfWs9+NPMbhbBzsCsrKBJUopm2P7IhNOvZphlVQWCCrMFuIGpMgQVyDvj99MkIaNbosnsug1eMz5LO5G1Gy0tsYfHMC6Rarb30x1Gw2st41VHBt6MGba/HFXtzyoVS5gtPXCDCiEjr4+XU3BhuFOwPqw5XjkEoF7be2hAerJN11kPDom2VPmhS2aLzyn8D8wognjAdN2oKSCV41bt0Rp46HfIZ/oQNC+Kp6/aDwZY=|tkp:Bk9SR5asn6rsZA
Attachments
Last edited:
The HA55-2223070LF are 40A and 2 milliOhm (1.7 milliOhm typically) but they're unshielded.
Last edited:
🙂 HA55-2223070LF bit unlucky long to get it on the board.
Simplest would be classic toroidal MA5173-AE with 1 mm wire, not too bad in distortion, not the very best either (8A)... on the board is a lot of space between the coils, so not sure if shielding is this important.
Simplest would be classic toroidal MA5173-AE with 1 mm wire, not too bad in distortion, not the very best either (8A)... on the board is a lot of space between the coils, so not sure if shielding is this important.
You did not specify a board. If the board already has SMD coils things are easier.
8A is too light and the toroid types are also unshielded FWIW. A modification usually is meant as an improvement.
8A is too light and the toroid types are also unshielded FWIW. A modification usually is meant as an improvement.
Last edited:
OK thanks, the Epcos are 5 uH only. But the price seems on the right side.
Why are those simple inductors so expensive, is there no DIY coil winding tutorial here? 🙂
Ah, and just plain simple BPTL 2xTPA3255 boards, like sold from aiyima for a cheap trick 6 channel mono TPA setup. As long as it's active and PEQed anyway I don't care for PFFB. I just want to mod them slightly in terms of caps and coils. Not too much hassle, low budget stuff. Driven with 46 Volts.
Why are those simple inductors so expensive, is there no DIY coil winding tutorial here? 🙂
Ah, and just plain simple BPTL 2xTPA3255 boards, like sold from aiyima for a cheap trick 6 channel mono TPA setup. As long as it's active and PEQed anyway I don't care for PFFB. I just want to mod them slightly in terms of caps and coils. Not too much hassle, low budget stuff. Driven with 46 Volts.
Oh DIY is possible but it is hard to be as precise as a machine and it will take time. DIYing shielded coils will be tough. Best is ready made shielded ones I guess.
If you care to make a picture or mention the SMD footprint the alternatives present themselves I think.
I used those Epcos in an Aiyima and a Sabaj A10a amplifier with good results. They can be mount vertically but if that really will work out OK.....
If you care to make a picture or mention the SMD footprint the alternatives present themselves I think.
I used those Epcos in an Aiyima and a Sabaj A10a amplifier with good results. They can be mount vertically but if that really will work out OK.....
Last edited:
In one instead of some low value air coils and in the other instead of nothing. Both not optimal (they needed a network) but better than before.
OK, ordered some Wurth 10 uH 19 A from Digikey for 2 amplifier modules. Leaving the low freq. amp on toroidal.
8 pcs. 7443631000 are 60 € inc. shipping and tax. Not cheap but fair.
Not much alternatives out there and these seems to be reference in function and quality.
8 pcs. 7443631000 are 60 € inc. shipping and tax. Not cheap but fair.
Not much alternatives out there and these seems to be reference in function and quality.
SMD!
What footprint do the boards have for the coils?
Just not solder extension antennas to those new coils. Then the original ones can be better. And … maybe first do all mechanical en electrical other work before starting on the coils. Just to avoid having erroneous boards with shiny expensive coils.
What footprint do the boards have for the coils?
Just not solder extension antennas to those new coils. Then the original ones can be better. And … maybe first do all mechanical en electrical other work before starting on the coils. Just to avoid having erroneous boards with shiny expensive coils.
Last edited:
So ... experience added. Some information for searchers:
I started with the TPA3255 PBTL stereo modules, bought it from Aiyima over ali but you find plenty offers from other shops too. I guess these boards are well known and absolutely tempting pricewise.
To good to be true? Yepp, some drawbacks are integrated. First the 8x 470 uF @50V are no mm higher than the board itself. This reduction is maybe made for cheap shipping in first place?!
However, 8x470 uF is not what you want to see on a 2x 3255 board, 50V neither.
Therefor I switched them for 8x1000 uF KYP Chemicon low ESR types, they fit perfect and the capacity is spot on.
Next thing was the output filter design, like most ti based designs, 10 uH and 1uF. Not practical in most cases. A 0.68 100V Wima instead the 1uF 75V foil lowering the Q and is a better compromise.
Biggest concern and reason for this thread were the coils. The classical ironpowder 1mm toroidals. had to leave for the Würth 7443782012100 with round wires for easy install.
Here're some impressions, as we all know, these are subjective but there's no doubt about the changes.
When I first tested the freshly deliverd module for 2 days, I made my regular setup, measured it with REW and a calibrated mic, EQed everything via PEQ in SoundDesk to my prefered housecurve. So amplitude was spot on and no unlinearity or room mode annoyed anything.
Power supply for the TPAs is a MW 48V 20A, that I really like 🙂
Even the unmodified modules offer a relative good amount of power, more than enough for private use and lower efficiency speakers like my nubert or Charios, this is a good thing. The problem was a slight harshness in womens voices i.e., far from what I would tolerate nowadays. No way that I went on with this crappy modules. But the parts from digi already shipped out and arrived yesterday. Heck, why do we need to import parts from the US ... but that's another topic.
After bringing the modules up to a higher quality, the sound changed in a rather dramatic fashion that I hoped for, but not really expected. Again did my sound setup 1:1 than before.
Absolutely stunning, how clear and pleasant womens voices are now, thats from fastfood to gourmet restaurant. Still, it's clearly a type of Class D sound but all the harshness went away, and no matter how hard I drive it, it stays pleasant to listen. What a change! Now I can order the amplifier case for the project 🙂
How good is it really? I just can compare it to the relative popular A07 Max, and the DIY board is definetly superior in every aspect.
So I need to strongly recommend swapping those parts if you're in the same boat with this PBTL board.
My guessing is, that most part of the change is caused by the MnZn coils and for louder levels the doubled capacity surely helps.
It's a pity, that I have no sophisticated equipment to present more detailed information by measurements. However, I hope this report helps a bit.
I started with the TPA3255 PBTL stereo modules, bought it from Aiyima over ali but you find plenty offers from other shops too. I guess these boards are well known and absolutely tempting pricewise.
To good to be true? Yepp, some drawbacks are integrated. First the 8x 470 uF @50V are no mm higher than the board itself. This reduction is maybe made for cheap shipping in first place?!
However, 8x470 uF is not what you want to see on a 2x 3255 board, 50V neither.
Therefor I switched them for 8x1000 uF KYP Chemicon low ESR types, they fit perfect and the capacity is spot on.
Next thing was the output filter design, like most ti based designs, 10 uH and 1uF. Not practical in most cases. A 0.68 100V Wima instead the 1uF 75V foil lowering the Q and is a better compromise.
Biggest concern and reason for this thread were the coils. The classical ironpowder 1mm toroidals. had to leave for the Würth 7443782012100 with round wires for easy install.
Here're some impressions, as we all know, these are subjective but there's no doubt about the changes.
When I first tested the freshly deliverd module for 2 days, I made my regular setup, measured it with REW and a calibrated mic, EQed everything via PEQ in SoundDesk to my prefered housecurve. So amplitude was spot on and no unlinearity or room mode annoyed anything.
Power supply for the TPAs is a MW 48V 20A, that I really like 🙂
Even the unmodified modules offer a relative good amount of power, more than enough for private use and lower efficiency speakers like my nubert or Charios, this is a good thing. The problem was a slight harshness in womens voices i.e., far from what I would tolerate nowadays. No way that I went on with this crappy modules. But the parts from digi already shipped out and arrived yesterday. Heck, why do we need to import parts from the US ... but that's another topic.
After bringing the modules up to a higher quality, the sound changed in a rather dramatic fashion that I hoped for, but not really expected. Again did my sound setup 1:1 than before.
Absolutely stunning, how clear and pleasant womens voices are now, thats from fastfood to gourmet restaurant. Still, it's clearly a type of Class D sound but all the harshness went away, and no matter how hard I drive it, it stays pleasant to listen. What a change! Now I can order the amplifier case for the project 🙂
How good is it really? I just can compare it to the relative popular A07 Max, and the DIY board is definetly superior in every aspect.
So I need to strongly recommend swapping those parts if you're in the same boat with this PBTL board.
My guessing is, that most part of the change is caused by the MnZn coils and for louder levels the doubled capacity surely helps.
It's a pity, that I have no sophisticated equipment to present more detailed information by measurements. However, I hope this report helps a bit.
