Silverpike,
I'll try and answer your questions in order. First, thanks for the smiley face, you're welcome.
As for the transformer, I am not running in bridged mode. I used two transformers to keep a lower chassis profile while keeping above the VA rating that I'd planned (I'm not much of a fan of bridged amps).
I may not truly need to invert to avoid the pumping issue with the approx 40,000 microfards of capacitance that I used, but the idea of alternating the current draw between channels and lowering the overall demand on the power supply, seemed attractive. I would image you could get by without it to save money. In any event, I like input transformers; I feel they sound better and break up ground loops and all sorts of good things. I wasn't so impressed with some of the suggested op-amps. but I do like the new AD8610 by Analog Devices. (I just made up a new 2x3 inch PCB using all SMT devices with 12 AD8610 op-amps for a fourth order Linkwitz-Riley X/O: two channels on one 2x3 inch board, including dual regulated supplies, rectifiers and transformer, along with gain control for bass output using two pots; SMTs are amazing!)
I used the supplied evaluation board because it was easy, and I have so many projects, a busy day job, and so the added expense did not feel so bad (also the price includes not only the parts but the heatsink, which for me is always a pain to source). I certainly would not recommend trying to lay out the board as high speed digital PCBs are not for the beginner. With regards to trying various Mosfets, I did not. Reading the white papers made it seem a fairly trying task.
Finally, as for the tweak in app 15, while I don't have the paper in front of me as I write this, I seem to recall that the schematic for the eval board already had incorportated the tweak. I believe app 15 was written a year or so ago, before the latest board revisions that are now available for the 103 and 104 devices.
Good luck with the project.
I'll try and answer your questions in order. First, thanks for the smiley face, you're welcome.
As for the transformer, I am not running in bridged mode. I used two transformers to keep a lower chassis profile while keeping above the VA rating that I'd planned (I'm not much of a fan of bridged amps).
I may not truly need to invert to avoid the pumping issue with the approx 40,000 microfards of capacitance that I used, but the idea of alternating the current draw between channels and lowering the overall demand on the power supply, seemed attractive. I would image you could get by without it to save money. In any event, I like input transformers; I feel they sound better and break up ground loops and all sorts of good things. I wasn't so impressed with some of the suggested op-amps. but I do like the new AD8610 by Analog Devices. (I just made up a new 2x3 inch PCB using all SMT devices with 12 AD8610 op-amps for a fourth order Linkwitz-Riley X/O: two channels on one 2x3 inch board, including dual regulated supplies, rectifiers and transformer, along with gain control for bass output using two pots; SMTs are amazing!)
I used the supplied evaluation board because it was easy, and I have so many projects, a busy day job, and so the added expense did not feel so bad (also the price includes not only the parts but the heatsink, which for me is always a pain to source). I certainly would not recommend trying to lay out the board as high speed digital PCBs are not for the beginner. With regards to trying various Mosfets, I did not. Reading the white papers made it seem a fairly trying task.
Finally, as for the tweak in app 15, while I don't have the paper in front of me as I write this, I seem to recall that the schematic for the eval board already had incorportated the tweak. I believe app 15 was written a year or so ago, before the latest board revisions that are now available for the 103 and 104 devices.
Good luck with the project.
Any thoughts on swapping output inductors?
I have heard the output inductors on these eval boards are highly critical as to the final sound quality. I wonder if anyone has tried replacing the stock powdered iron toroidal inductors with these: http://www.madisound.com/steellaminate.html
Any comments, for or against?
I have heard the output inductors on these eval boards are highly critical as to the final sound quality. I wonder if anyone has tried replacing the stock powdered iron toroidal inductors with these: http://www.madisound.com/steellaminate.html
Any comments, for or against?
kuribo, I would try experimenting with winding your own air core inductors. So far, I have had some promising results winding magnet wire on a Marks-a-Lot magic marker, taping around the windings, then sliding them off. Then you can run a piece of tape inside the core also to make it more stable. Later, you can even make new ones and use glue instead of tape. Wind two equal cores and physically place them side-by-side. Then place the electrical connections in series so that the current flow is in opposite directions so the magnetic fields cancel at a distance.
I think this idea may be a good possibility for you. Air cores inherently have the lowest distortion because they don't saturate. Also, you need only between 10 and 20 microH, which would require only about 15 or 20 turns in each direction on a 1 inch diameter form. Use at least 18 guage wire.
My need for inductiors involves a Class D project like this one.
http://listen.to/audioexperiment
The Tripath amp seems very good though. I would consider it for myself if I were not in the practice of designing circuits from scratch.
I think this idea may be a good possibility for you. Air cores inherently have the lowest distortion because they don't saturate. Also, you need only between 10 and 20 microH, which would require only about 15 or 20 turns in each direction on a 1 inch diameter form. Use at least 18 guage wire.
My need for inductiors involves a Class D project like this one.
http://listen.to/audioexperiment
The Tripath amp seems very good though. I would consider it for myself if I were not in the practice of designing circuits from scratch.
aircore inductors and tripath amp
Your suggestion to try aircore inductors was my first thought, but when looking at existing aircore 18mH inductors available commercially, they are HUGE. Solen makes aircore inductors up to 5mH and these are already 4 inches in diameter...Do you have the formulas for inductance values for calculating turns at a given diameter, per wire gauge, etc.?
Your suggestion to try aircore inductors was my first thought, but when looking at existing aircore 18mH inductors available commercially, they are HUGE. Solen makes aircore inductors up to 5mH and these are already 4 inches in diameter...Do you have the formulas for inductance values for calculating turns at a given diameter, per wire gauge, etc.?
Air core inductors
Kuribo, I think you've got it wrong: you are looking at 10 to 20 MICRO Henry. That makes the inductor much smaller in size!
On the other hand, Tripath specifically asks for iron core inductors. We're dealing with very high frequencies, so I don't know if you can eliminate the core.
Kuribo, I think you've got it wrong: you are looking at 10 to 20 MICRO Henry. That makes the inductor much smaller in size!
On the other hand, Tripath specifically asks for iron core inductors. We're dealing with very high frequencies, so I don't know if you can eliminate the core.
aircore inductors
Whoops, you are absolutely right, 18uH, not 18mH....I spoke with tripath about the inductors and they said the iron cores were the best solution they found. I still wonder if the laminated steel cores, claimed to be lowest in distortion after aircores, might not be a better choice.....
Whoops, you are absolutely right, 18uH, not 18mH....I spoke with tripath about the inductors and they said the iron cores were the best solution they found. I still wonder if the laminated steel cores, claimed to be lowest in distortion after aircores, might not be a better choice.....
m.parigi, thanks. The iron powder in the cores like the ones Amidon offers does provide specific properties. As I reviewed some of the posts in this thread, I saw that the output filter coil is loaded by a .22microF cap. That combination may make an air core combination too resonant, for example.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.