Not actually possible if all else is equal
As I read the same posts DUGs PBTL board has slightly better bass... not more.. The differences become increasingly subtle as resistances decrease though. The ICE component inductors have 9mOhm DC resistance compared to a standard ebay TPA3116 amp which typically has 33mOhm (or worse with saturation). The output transitors in the chip already contribute 120mOhm so they're already the dominant factor. Decreasing it further under the 9mOhm of the ICE component inductors will naturally see less and less return. However, as DUGs board is PBTL the output transitors are parallel so the effective output resistance is half. That is actually the crucial thing here. The total series resistance is halved (instead of just being modified by a few percent) and that actually does mean the resulting bass has slightly more volume.
But I started this and the other post by saying: "if all else is equal". That means if you have to compare it to DUGs PBTL board you also have to set up the Weiner in PBTL mode. Otherwise, it is far from an equal comparison.
Last edited:
The 120mohm Rds(on) of the TPA3116 output transistors is within the feedback loop. At bass frequencies, feedback will correct for it.
Not actually possible if all else is equal
As I read the same posts DUGs PBTL board has slightly better bass... not more.. The differences become increasingly subtle as resistances decrease though. The ICE component inductors have 9mOhm DC resistance compared to a standard ebay TPA3116 amp which typically has 33mOhm (or worse with saturation). The output transitors in the chip already contribute 120mOhm so they're already the dominant factor. Decreasing it further under the 9mOhm of the ICE component inductors will naturally see less and less return. However, as DUGs board is PBTL the output transitors are parallel so the effective output resistance is half. That is actually the crucial thing here. The total series resistance is halved (instead of just being modified by a few percent) and that actually does mean the resulting bass has slightly more volume.
But I started this and the other post by saying: "if all else is equal". That means if you have to compare it to DUGs PBTL board you also have to set up the Weiner in PBTL mode. Otherwise, it is far from an equal comparison.
That's practically impossible unless the amp runs in 1SPW modulation mode. As all other BLT or push-pull amplifiers even order harmonics are canceled out. In 1SPW mode the amp operates basically the same as a single ended output with an infinite coupling capacitor. Most of the rise in THD in 1SPW mode compared to standard BD mode is from this simple fact. Even ordered harmonics, especially the 2st order, however, are essential for making an amp sound more "natural" or "warm". Without even order harmonics there can be no psychoacoustic missing fundamental effect.
Somewhat OT - Buck Converter
Later this month I will embark on my first buck converter build and the chip choices are over whelming. Can any of you suggest an easy to use chip series that tends to be fairly fool proof for audio design if you follow the data sheet guidelines?
What I want to do is leverage 6 cell LiPo packs that I already own for use with the Wiener. The voltage could be as high as 25.2 which is higher than optimal so I would like to regulate that to 20V. Additionally I need a 5v rail for wireless communications.
Thanks in advance.
Later this month I will embark on my first buck converter build and the chip choices are over whelming. Can any of you suggest an easy to use chip series that tends to be fairly fool proof for audio design if you follow the data sheet guidelines?
What I want to do is leverage 6 cell LiPo packs that I already own for use with the Wiener. The voltage could be as high as 25.2 which is higher than optimal so I would like to regulate that to 20V. Additionally I need a 5v rail for wireless communications.
Thanks in advance.
For 25V -> 5V conversion, I'd go with a synchronous buck from TI or Linear, lots of options available. If you want absolutely foolproof, TI sells integrated modules that just need capacitors placed on the input/output.
To run the Wiener card itself, I'd convert to 24V with a LDO linear regulator. Converting 25.2V->24V is >95% efficient, and that efficiency # will only get better as the battery voltage drops. You'd be hard pressed to beat that with a switcher.
To run the Wiener card itself, I'd convert to 24V with a LDO linear regulator. Converting 25.2V->24V is >95% efficient, and that efficiency # will only get better as the battery voltage drops. You'd be hard pressed to beat that with a switcher.
Excellent - may you keep on enjoying good company, and more audio adventures!
Hear hear!
Yes Dr that is a concern. The batteries will happily go down to 3.25V cell (19.5v for the pack). I don't want to be in drop out before the pack hits 20V. That means a lower operating voltage. Maybe I should just run unregulated. The 25.2 is less than the TPA 26v max recommended and well below the 30v absolute max.
A LDO with a PMOS pass transistor should turn completely on in dropout, and not draw excessive current or anything like a PNP based one.
Probably wouldn't be too hard to roll your own using a PFET, a TL431 and a handful of discrete parts. Simulate the thing up in LTspice and go from there.
So, big question time... I'm gauging interest for another round of Wiener builds.
Who wants a monoblock amp?
With DUG's PBTL group buy ending this month, I might make a monoblock "Big Wiener" to replace it on the market. The card will be the same size/dimensions as the existing Wiener card and share most of the same design/parts. The left audio input will be hauled out, and the current 1D14A/7G14D based stereo output filter will be changed to use larger ICE 1D17A inductors. These inductors are also available at 4.7uH, so amps can be built for 2 ohm operation as well as the current 4/6/8/16 ohm configurations.
Everything else in the design (broadband decoupling, uC based startup circuit, DIP switches/pots to configure everything, etc) will be kept.
I'll keep selling the stereo Wiener of course. I might add a couple of .100" jumpers which connect the negative audio inputs to ground, as the fully differential inputs have been giving people some grief, but other than that the design will remain unchanged.
Who wants a monoblock amp?
With DUG's PBTL group buy ending this month, I might make a monoblock "Big Wiener" to replace it on the market. The card will be the same size/dimensions as the existing Wiener card and share most of the same design/parts. The left audio input will be hauled out, and the current 1D14A/7G14D based stereo output filter will be changed to use larger ICE 1D17A inductors. These inductors are also available at 4.7uH, so amps can be built for 2 ohm operation as well as the current 4/6/8/16 ohm configurations.
Everything else in the design (broadband decoupling, uC based startup circuit, DIP switches/pots to configure everything, etc) will be kept.
I'll keep selling the stereo Wiener of course. I might add a couple of .100" jumpers which connect the negative audio inputs to ground, as the fully differential inputs have been giving people some grief, but other than that the design will remain unchanged.
Regarding Icecomponents and Saturnus and Gmarsh bass remarks: march 2014 I described comparing the 9mohm Icecomponents to the Wurths I believe 11mohm smd square inductor as follows:
"
icecomponent/wurth inductor2
wurth it is!
icecomponent bass sounded louder, total mono sound more dynamic...."
I chose the Wurths on that set of ampboards, but bass was less loud, not only heard by me, but after listening with friend. In light of Saturnus and Gmarsh on bass, datasheet Wurths might be more accurate or Wurths distort less.
"
icecomponent/wurth inductor2
wurth it is!
icecomponent bass sounded louder, total mono sound more dynamic...."
I chose the Wurths on that set of ampboards, but bass was less loud, not only heard by me, but after listening with friend. In light of Saturnus and Gmarsh on bass, datasheet Wurths might be more accurate or Wurths distort less.
Does going mono/PBTL make enough room for the Coilcraft SER29xx series inductors? Maybe if they stand upright instead of flat on the PCB?... and not available in 4.7uH. That's what I get for talking off the top of my head, I've got ICE and Sagami parts confused.
In any case, bigger inductors will be used