[error401]

I've been very swamped with work the past few weeks, but the bug that TzeYang noticed has so far been defeating me. It seriously complicates the layout since the (now) two sets of power rails cross in several places. I'm not sure I have the motivation to completely ripup the board and redo it, and I'm not sure that would help either.

So for now it's on hold until I have more time to put into redesigning the PCB to fix this error. Or I can toss the JFET isolation entirely (I'm not sure it buys anything anyway, TBH) and do a even more stripped down PPA-like amp instead.

 

[NelsonVandal]

Quote:

Originally Posted by error401 /img/forum/go_quote.gif I've been very swamped with work the past few weeks, but the bug that TzeYang noticed has so far been defeating me. It seriously complicates the layout since the (now) two sets of power rails cross in several places. I'm not sure I have the motivation to completely ripup the board and redo it, and I'm not sure that would help either. So for now it's on hold until I have more time to put into redesigning the PCB to fix this error. Or I can toss the JFET isolation entirely (I'm not sure it buys anything anyway, TBH) and do a even more stripped down PPA-like amp instead.

Can't you hardwire resistors, something like 10R, instead of the JFET's. I'll eat my hat if someone can hear the difference

 

[error401]

Removing the JFETs isn't an issue, in fact I think that'd be preferable really. But to keep the original PPAS design I'd have to keep the JFETs and use them only for the opamp supplies. That's where I'm having layout difficulties.

But if everyone is in favour I'd rather just rip them out. The power layout on this board is much better than the original anyway with proper bypass.

 

[pinkfloyd4ever]

I wish I knew jack about eagle (or even circuit design for that matter), since I'm on spring break this week and would actually have some time to play around and see what kind of a layout I could come up with.

I don't even really understand the difference in the original PPAS design and error's layout. I think the schematic being in separate sections is messing with me.

 

[error401]

Does anyone have any input on the proposal to remove the JFET isolation? Any prototypers willing to try jumpering the JFETs and reporting on sound quality changes if any?

 

[tcpoint]

I'll try jumpering the jfets. I'll probably have some time this weekend.

 

[TzeYang]

If it's especially battery operated, I don't really think there's a humongous difference. Plus it's only really "significant" if the current draw is matched to the IDSS of the JFETS.

 

[tcpoint]

I bypassed the jfets and it still sounds great. I couldn't tell the difference (before and after). I don't have magic ears, either

 

[TzeYang]

That's not the proper way to test it because you're still feeding the buffers with the same supply line. The JFETs will also limit the current of the buffers.

I'd say removing the fets and jumpering them counts as a temporary "fix" until the entire board layout is changed again.

 

[FallenAngel]

Knowing the crappy economy, I still would love to see this project move forward quickly

TzeYang - Yes, the PSU drives everything if you bypass the JFETs, but it shouldn't be a big concern, most amps don't even have these. We are of course aiming for best portable here, but until then, we should at least know that the amp works perfectly without rail isolation for the opamps.

error401 - If you have the time to change the board, I'll be happy to pay for the full prototype PCB run of the revised boards. Just send me a PM. You know I love this amp, I want to see more of them build!

 

[TzeYang]

I meant it is not a valid comparison because it is not divided into two situations where they are both working properly. JFET feeding the buffers and the OPAMPs at the same time might be erroneous during high current transients.

If we wish to see whether it's a significant different in sound quality, we must test them both with the JFET situation working properly.

Personally, I could careless because they don't usually make a huge difference in my past experience. I'd suggest dropping the JFET scheme and make them as compact as possible, to even consider putting the I/O jacks on board? That'll make building so much easier.

 

[tcpoint]

I agree with TzeYang that w/o FET isolation it technically isn't a PPAS. I think that may be the last 1% or something like that. I think this is a great sounding amp in small package. There really is a need for an amp like this.

 

[error401]

Quote:

I agree with TzeYang that w/o FET isolation it technically isn't a PPAS. I think that may be the last 1% or something like that. I think this is a great sounding amp in small package. There really is a need for an amp like this.

Exactly my dilemma. I haven't found a way to layout the board in as small a package while maintaining the original PPAS JFET isolation. So our choices are to keep the current board with the JFETs or remove them entirely. I'm on the side of removing them entirely, but it's up to you guys. And if someone wants to try a layout that does have them wired up the way the original was that's fine with me too, but I can't do it on 2 layers at this form factor.

If nobody is going 'pfft thats a stupid idea' I can go ahead and work on removing them and another prototype run. It'll certainly make for a simpler amp.

 

[linuxworks]

remove them. if you are tight for space, yank them.

(did I say that?)

hey, don't listen to me.

 

[cobaltmute]

If you remove the JFETs entirely, you've just made a Pimeta.