This is the SKA's measured THD spectrum.
Hi PauloPT,
Was that at 1w 8R?
Member
Joined 2009
Paid Member
This is the SKA's measured THD spectrum.
It looks very good. The dominant H2 no doubt due to the fact that p-type and n-type parts don't match each other perfectly along with other real-life asymmetries.
I also agree that the power supply is another key factor, and furthermore that the simulations are based on a perfect resistive load whilst a real speaker is not. That's why it's great to have so many positive reports from listeners about their experience with the amplifier. I would however be interested in more comparative information, against other amplifiers topologies I have heard - e.g. SKA GB150 vs Aspen Lifeforce ? and SKA vs Bryston BP60 ? (I have seen comparisons against AKSA original already but they are such different amplifiers in terms of their goals that this isn't that informative).
I have no experience using FETs - makes me wonder, did anybody build this amplifier with BJT instead of FET outputs and also find a good result ?
Member
Joined 2009
Paid Member
I could guess it is the difficulty of maintaining an optimal bias for good cross-over performance when using a symmetrical design whereas FETs generally are less-fussy (just give them more current and things keep getting better) - but it's just a guess. I'd like to know what Greg had to say.
I'd also like a summary of the modifications people have made to the GB150 - if I (or somebody more skilled) makes up a pcb I'll want to know what to incorporate. I read about a zener mod but I think this was a passing fad and the latest schematic doesn't benefit from it (besides, zeners are noisy and trying to amerliorate this with large bypass caps on zeners leads to turn on issues).
Edit: I realized that I may not be able to post a pcb design because I'll probably use Greg's layout as a starting point and it is copyright.
I'd also like a summary of the modifications people have made to the GB150 - if I (or somebody more skilled) makes up a pcb I'll want to know what to incorporate. I read about a zener mod but I think this was a passing fad and the latest schematic doesn't benefit from it (besides, zeners are noisy and trying to amerliorate this with large bypass caps on zeners leads to turn on issues).
Edit: I realized that I may not be able to post a pcb design because I'll probably use Greg's layout as a starting point and it is copyright.
Last edited:
VGS threshold
Thanks for confirming my post#67
This is also the same reason why lateral MOSFET can not be used with their lower VGS threshold...
Fab
Hi Bigun; Greg addressed this (the BJT question) to someone else - he says the design requires MOSFETs with a relatively high VGSon threshold circa 3.5V so the driver circuitry can function correctly.
Thanks for confirming my post#67
This is also the same reason why lateral MOSFET can not be used with their lower VGS threshold...
Fab
I don't believe Spice for stability as it knows nothing of track lengths and has model errorsThere are no stability issues with the GB150. I've built or modified over a dozen of these and have never seen an issue.
Don't believe evrything spice tells you.
So register. Greg has no problem approving new people.
I did register. There is still one forum that appears to be locked if you haven't purchased a product. Since Greg released this amp to the public, wouldn't it be OK for you to share your experiences here?
Apart from the biasing problems, BJTs in common emitter outputs are prone to fatal cross conduction.I could guess it is the difficulty of maintaining an optimal bias for good cross-over performance when using a symmetrical design whereas FETs generally are less-fussy (just give them more current and things keep getting better) - but it's just a guess. I'd like to know what Greg had to say.
I'd also like a summary of the modifications people have made to the GB150 - if I (or somebody more skilled) makes up a pcb I'll want to know what to incorporate. I read about a zener mod but I think this was a passing fad and the latest schematic doesn't benefit from it (besides, zeners are noisy and trying to amerliorate this with large bypass caps on zeners leads to turn on issues).
I am thinking about replacing those zeners with transistor clamps sensing the drop across the 0.15R resistors. The usual SOA monitoring of the voltage across the FETs adjusting the clamp current and you allow high currents into low Z but near zero phase loads, while protecting against shorts and high phase loads. Noise should not be an issue when the zeners are (normally) not conducting
Member
Joined 2009
Paid Member
I did register. There is still one forum that appears to be locked if you haven't purchased a product.
That's the case - I've been registered over there for years and there is a sealed forum for owners-only. Fair enough if you ask me and I don't think we should be re-posting from the sealed forum over here without permission.
Hi David - I was just over in KL (mostly Seremban) for a vacation.
Member
Joined 2009
Paid Member
copyright.
But don't worry - there have been some amazingly good pcb designs on this forum and if there's enough interest I expect one will be developed.
fyi - I was thinking about the question from before about supply rails for my TGM6 and hence it's psrr - well I think my TGM6 amplifier has also a very high psrr because of the independent supply for the front end which being low current can be made very quiet.
But don't worry - there have been some amazingly good pcb designs on this forum and if there's enough interest I expect one will be developed.
fyi - I was thinking about the question from before about supply rails for my TGM6 and hence it's psrr - well I think my TGM6 amplifier has also a very high psrr because of the independent supply for the front end which being low current can be made very quiet.
Last edited:
copyright.
But don't worry - there have been some amazingly good pcb designs on this forum and if there's enough interest I expect one will be developed.
fyi - I was thinking about the question from before about supply rails for my TGM6 and hence it's psrr - well I think my TGM6 amplifier has also a very high psrr because of the independent supply for the front end which being low current can be made very quiet.
What you're saying is that the PCB is copyrighted? The schematics are fully open domain though right?
We'll design a new layout no problem, we don't want to infringe on anything here.
Thanks
Do
Last edited:
Member
Joined 2009
Paid Member
What you're saying is that the PCB is copyrighted? The schematics are fully open domain though right?
I believe somebody already asked Greg if he would release the pcb design and he declined.
Actually, what he was asked was if he was planning to release the PCB artwork and his answer was
"Hadn't thought about that...but probably not."
There are plenty of pictures of the forum if someone wants one. Maybe not of the foil, but of the top for sure. I would think he would want folks to do a good layout for the amp so it works properly. Why reinvent the wheel?
"Hadn't thought about that...but probably not."
There are plenty of pictures of the forum if someone wants one. Maybe not of the foil, but of the top for sure. I would think he would want folks to do a good layout for the amp so it works properly. Why reinvent the wheel?
Actually, what he was asked was if he was planning to release the PCB artwork and his answer was
"Hadn't thought about that...but probably not."
There are plenty of pictures of the forum if someone wants one. Maybe not of the foil, but of the top for sure. I would think he would want folks to do a good layout for the amp so it works properly. Why reinvent the wheel?
If Greg is ok with what you said, it would be nice to get the gerber files instead of doing a layout copied from pictures if that is permitted? Anyways, I'm registering on his forum and will ask permissions.
Thanks
Do
- Home
- Amplifiers
- Solid State
- SKA GB150D now public domain...