Further details published :
http://www.diyaudio.com/forums/pass-labs/206240-f5-turbo-posted.html#post2897587
Patrick
There are some discussions about the level of Vgs match between P type & N type output devices at the GB thread :
http://www.diyaudio.com/forums/grou...e-user-distribution-check-17.html#post2903207
While the discussion over there is whether the supplier has delivered what he promised, most of you would be interested in how well the P-N matching has to be. After all, perfectly P-N match is an absolutely rarity.
http://www.diyaudio.com/forums/grou...e-user-distribution-check-17.html#post2903371
The consequence of a slight mismatch between P & N devices is an increase in second harmonics. While this is more important for the single ended F5 ( and hence Nelson's modified version to include a trimpot across the two feedback shunt resistors), it is much less important for the F5X, as even harmonics will cancel out between the left and the right halves, provided they behaves identically. And the latter can be achieved through good N-N and P-P match, which is much easier to obtain. I consider a 20dB cancellation between left & right halves to be achievable.
Further fine tuning of the even harmonic contents of each (L or R) half is still possible, either by trimming the feedback shunt resistor, as Nelson proposed, or by trimming the output MOSFET source resistor, as he did earlier with his First Watt F5's using Fairchild devices. Note than the Fairchild devices has no P-N match at all. The Vgs at 1.3A (for single ended F5) differs by more than 1V !!
All these issue would be dealt with once fitzfish finds time to start the Build thread.
Right now he does not have enough time to sleep.
One should not forget that this is hobby, and we all have a living to make in the first place.
Patrick
http://www.diyaudio.com/forums/grou...e-user-distribution-check-17.html#post2903207
While the discussion over there is whether the supplier has delivered what he promised, most of you would be interested in how well the P-N matching has to be. After all, perfectly P-N match is an absolutely rarity.
http://www.diyaudio.com/forums/grou...e-user-distribution-check-17.html#post2903371
The consequence of a slight mismatch between P & N devices is an increase in second harmonics. While this is more important for the single ended F5 ( and hence Nelson's modified version to include a trimpot across the two feedback shunt resistors), it is much less important for the F5X, as even harmonics will cancel out between the left and the right halves, provided they behaves identically. And the latter can be achieved through good N-N and P-P match, which is much easier to obtain. I consider a 20dB cancellation between left & right halves to be achievable.
Further fine tuning of the even harmonic contents of each (L or R) half is still possible, either by trimming the feedback shunt resistor, as Nelson proposed, or by trimming the output MOSFET source resistor, as he did earlier with his First Watt F5's using Fairchild devices. Note than the Fairchild devices has no P-N match at all. The Vgs at 1.3A (for single ended F5) differs by more than 1V !!
All these issue would be dealt with once fitzfish finds time to start the Build thread.
Right now he does not have enough time to sleep.
One should not forget that this is hobby, and we all have a living to make in the first place.
Patrick
For more info on P-N match we published in 2008, see :
http://www.diyaudio.com/forums/pass-labs/121228-f5-power-amplifier-14.html#post1507044
followed by the same info for Fairchild MOSFETs as used by Nelson :
http://www.diyaudio.com/forums/pass-labs/121228-f5-power-amplifier-30.html#post1567443
Patrick
http://www.diyaudio.com/forums/pass-labs/121228-f5-power-amplifier-14.html#post1507044
followed by the same info for Fairchild MOSFETs as used by Nelson :
http://www.diyaudio.com/forums/pass-labs/121228-f5-power-amplifier-30.html#post1567443
Patrick
Last edited:
Patrick.... I appreciate you taking the time to "dig out" the relevant info, and for posting the links. It really helps us who are still attempting to come up to speed on F5 and F5X topologies. These "mega-threads" have a wealth of good info in them--digging them out can be a real chore, so my thanks for doing that legwork. I know you're busy--I appreciate you taking the time.
There is no consensus. This is how I would build a balanced F5.
As said in post#1, only one of the many ways to skin a cat.
And as already mentioned, in simulations and measurements, the floating and grounded X behave similarly.
But the floating X is capable of taking single ended input. The grounded X needs balanced input signals.
With proper matching between left and right halves, all even harmonics will cancel out in the F5X.
There is no need to introduce the P3 adjustment. In particular not with Toshiba Mosfets.
Patrick
As said in post#1, only one of the many ways to skin a cat.
And as already mentioned, in simulations and measurements, the floating and grounded X behave similarly.
But the floating X is capable of taking single ended input. The grounded X needs balanced input signals.
With proper matching between left and right halves, all even harmonics will cancel out in the F5X.
There is no need to introduce the P3 adjustment. In particular not with Toshiba Mosfets.
Patrick
I am not sure what the grounded-X topology is. How does floating-X compare to what Nelson shows in the F5 Turbo Balanced diagram? What terminology do you use for that topology?
I found your reference to gounded-X here: http://www.diyaudio.com/forums/pass-labs/183362-f5x-euvl-approach-63.html#post2837133. So grounded-X is the same as a pair of bridged amplifiers.
Apparently Nelson's balanced F5 Turbo isn't an X, just floating balanced that should accept unbalanced inputs, but doesn't get any advantage from the X cancellation. The P3 pots are needed to minimize (or adjust to taste) the 2nd harmonic.
I have been planning to build a balanced F5 for quite some time. There are many ways to do this. Even more so after the publication of the Turbo version.
EUVL: you refer to "them" above, who will decide whether the F5X boards will be available at DIYA forum. I am not sure of whom you refer, but I would like to know if these boards as such will be available. Making a set on my own seems somewhat uneccessary.
Chalk
> who will decide whether the F5X boards will be available at DIYA forum.
Variac and Jason, i.e. the Forum.
We made an offer to give them the Gerber files and the right to sell exclusively, right at the beginning.
Whether they consider it is worth their while is entirely their decision.
If they were to decide not to, then xen audio will make them available in limited batches of Group Buys.
But in principle we have no real interest to do so. Too much work for little return.
Patrick
Variac and Jason, i.e. the Forum.
We made an offer to give them the Gerber files and the right to sell exclusively, right at the beginning.
Whether they consider it is worth their while is entirely their decision.
If they were to decide not to, then xen audio will make them available in limited batches of Group Buys.
But in principle we have no real interest to do so. Too much work for little return.
Patrick
The F5X build thread is up. We will post more information there as it progresses.
http://www.diyaudio.com/forums/pass-labs/208880-f5x-euvl-approach-build-thread.html
Dave
http://www.diyaudio.com/forums/pass-labs/208880-f5x-euvl-approach-build-thread.html
Dave
This project has been long and arduous, and required more resources of time and money that the forums could spare. Which is why we're so impressed that the awesome results. diyAudio would like to produce the basic boards, which after all, have been designed to fit the "Hole Standard" for boards in the store, so will fit right in.
Whoever has the Gerbers should contact me via e-mail and we'll figure out the next step. If you've lost my e-mail address send me a PM
Variac
Whoever has the Gerbers should contact me via e-mail and we'll figure out the next step. If you've lost my e-mail address send me a PM
Variac
Variac,
As far as I know we have not changed anything, so your last set of Gerbers are still valid.
But if you want, fitzfish can send you again.
We leave it up to the forum to decide whether they wait for the Beta testers to finish,
or start making them available now.
As far as we know the boards are bug free.
Best,
Patrick
As far as I know we have not changed anything, so your last set of Gerbers are still valid.
But if you want, fitzfish can send you again.
We leave it up to the forum to decide whether they wait for the Beta testers to finish,
or start making them available now.
As far as we know the boards are bug free.
Best,
Patrick
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.