PPA Project Announcement
Apr 10, 2003 at 8:30 PM Post #227 of 340
The gain differences between heaphone amps and preamps aren't really that great, all things considered. 6 to 12dB is common for preamps, but realize that they're dealing with line-level signals. Headphone amps often go for gains of 20dB or so, but that's because they're often used with portable sources that have less than line level output voltages. (My Panasonic PCDP's "line out" is something like 300mV as I recall, but true line level is 1 to 2V.) If you're making a headphone amp designed for being driven from a true line-level source, a gain of 6 to 12dB will be fine with most headphones.

Output current is indeed given more attention in headphone amps, since preamps don't have to drive low-impedance sources. However, having high output drive on a preamp can't hurt, and it can help a lot with some power amps.

But the biggest single difference is that most preamps have multiple inputs and one or two line level outputs plus a headphone output.

Bottom line, it's not too hard to convert most headphone amp circuits into a preamp. It's mostly a physical issue: a larger case for holding all those extra jacks and switches and such.

Morsel's point is that the differential drive circuit in the PPA is optimized for headphones. While it should work if you plug the PPA's output into a power amp, the ground channel is likely to be useless in that kind of setup because the ground channel is either tied to earth ground or is some kind of active ground. Headphones' ground channels are totally passive, connected to nothing external. That allows the PPA's rule of the ground channel to be absolute.
 
Apr 10, 2003 at 11:58 PM Post #228 of 340
Yes as stated above I would not use the PPA as a preamp or connected in a tape monitor Loop as tangent pointed out this will put the ground channel and the input ground at the same point. now considering the ground channel has Open loop Voltage gain and this is reduced by Negitive feedback. connecting the output ground and the input ground at the same point is connecting the input to the output of the Ground channel. This is a great way to create an oscillator.
 
Apr 11, 2003 at 9:03 PM Post #232 of 340
Judging by the quiet reception of v0020 it sounds like most of the major issues have been addressed.

Squawk now or hold your peace, as we will be having a second set of prototypes made soon.
 
Apr 21, 2003 at 5:51 AM Post #234 of 340
Quote:

Originally posted by morsel
Squawk now or hold your peace, as we will be having a second set of prototypes made soon.


My only comment is that there are a lot of places where traces could be made bigger without any problems.

Possible candidates for fatter traces would be the rows of caps along the left, the input signal traces to and from the pot, power rails to the opamps (especially the bits around the transistors, which are really skinny), etc, etc.

Is there any reason not to have wider traces?
 
Apr 21, 2003 at 6:15 AM Post #235 of 340
I guess that wasn't my only comment.

It seems like there would be room to add an input protection section on the right. If you chose not to use it you'd just have to jumper it, while still allowing the board to be used with outputs that have DC on them.

Also, from a purely aesthetic point of view, has anyone considered putting the pot dead center rather than off to the side? I think a pot in the middle with a jack on one side and a LED on the other has a nice symmetry to it.
 
Apr 21, 2003 at 6:51 AM Post #236 of 340
Quote:

Updates?


The second prototype is being finalized now.

Quote:

My only comment is that there are a lot of places where traces could be made bigger without any problems.


I guess along the power rails would be helpful, but larger traces don't seem to make sense anywhere else. Keep in mind, we've segregated the high-current sections of the amp from the low-current sections, and the circuit paths are all quite short. Also, fatter traces aren't without cost: space between traces can be as helpful as thick traces.

Quote:

the bits around the transistors, which are really skinny


These are low-current paths that are nearly static -- only microamps flow here.

Quote:

It seems like there would be room to add an input protection section on the right.


The PPA is not a general-purpose amp. It's for higher-end systems, which are assumed to have their DC offset situation under control already. Therefore, the cap is of no help, and of some disadvantage.

My other thought on this is, if you want a cap, you can certainly put it in between the input jack and the input pads if you have to have it. The main downside of this is that it will have to be quite a large cap since the pot that follows it will be ~50K. That's not a big enough issue to justify trying to stick the input cap in after the pot, since that would require giving up precious space in the actual amp section.

Quote:

I think a pot in the middle with a jack on one side and a LED on the other has a nice symmetry to it.


Not when the jack is a Neutrik NJ3FP6C jack, which is seriously being considered due to the buffer protection issues we're facing.
 
Apr 21, 2003 at 6:52 AM Post #237 of 340
No news atm. I need to synch up the v0021 layout and schematic and make some modifications for prototype testing. I've been preoccupied lately but will try to get to it soon. After the new prototypes are ordered we will turn our attention to the power supply, probably in a new thread.

I did widen the power traces to the output buffers. The other traces are probably wide enough that changing them is not going to improve anything. The pot is not going dead center, sorry. Air wire it if you want it centered. Given the limited space, having the pot and input leads against the edge of the board keeps noise and hum to a minimum.
 
Apr 21, 2003 at 3:16 PM Post #238 of 340
Quote:

Originally posted by tangent Not when the jack is a Neutrik NJ3FP6C jack, which is seriously being considered due to the buffer protection issues we're facing.


You're right. I didn't consider the size of that jack.

Thanks for the comments from both of you. This is shaping up to be a very sweet-looking amp. I've been thinking about doing a better one than what I've got, but I might just hold off until this is ready...
 
Apr 21, 2003 at 6:02 PM Post #239 of 340
Quote:

Originally posted by tangent
That's not a big enough issue to justify trying to stick the input cap in after the pot, since that would require giving up precious space in the actual amp section.


Put two holes in the input trace, and if someone needs/wants a cap, they can solder the leads into the holes and cut the trace between them...
 
You must log in or register to reply here.

Users who are viewing this thread

Total: 2 (members: 0, guests: 2)
Back
Top