AudioSan, if you update your BOM, can you make the part numbers match the F5Tv3 schematic from Nelson's original article? The new board's silk numbering matches the v3 schematic.
How about options for R5 and R6 if using lower Idss for JFETs. At least for my build with 7.3mA 1K isn't working...I had to change to 2.2K to make it work. Lower it to 1.8K to make the bias adjustment more finer.
If I parallel 2 sets of JFETs (7.3mA) should R5 and R6 should be lower than 1K?
If I parallel 2 sets of JFETs (7.3mA) should R5 and R6 should be lower than 1K?
Hi Zen,
Got it.
However, since they are preparing the BOM, would it be nice if this kind of note or remarks will be included in the list.
Regards
Got it.
However, since they are preparing the BOM, would it be nice if this kind of note or remarks will be included in the list.
Regards
abadsantos - That kind or remarks would be nice have on the BOM or on it's own document. Would you like to help with it?
Hi Jim,
Its OK. You may consult others also who use the TOSHIBA FETs, I think they modified R5 and R6 also to a lower value (lower than 1K).
Regards.
Its OK. You may consult others also who use the TOSHIBA FETs, I think they modified R5 and R6 also to a lower value (lower than 1K).
Regards.
R5/R6 have been discussed in multiple F5 treads. this has been a chalange for many original F5 builders as wel. there will be some that need to go a little higher on R5/R6, but there is no standar.
Audiosan
The terminal blocks have gone to allow multiple ground connections on the Front End board and multiple output connections on the N & P Channel boards.
The terminal blocks have gone to allow multiple ground connections on the Front End board and multiple output connections on the N & P Channel boards.
oh. i see there is been some layout changes.
terminal blocks changed.
is C1 and C2 correct on the output boards? i can not tell from the pcb pic in the store.
also. i can not read the markings on the sec pair of J-fets on the front end board.
terminal blocks changed.
is C1 and C2 correct on the output boards? i can not tell from the pcb pic in the store.
also. i can not read the markings on the sec pair of J-fets on the front end board.
Last edited:
(another) Updated V3 BOM -
http://www.diyaudio.com/forums/attachment.php?attachmentid=374706&stc=1&d=1380891007
http://www.diyaudio.com/forums/attachment.php?attachmentid=374706&stc=1&d=1380891007
Source resistors issue with 2SJ/2SK
Hi all,
getting closer and closer to starting the build of the F5T as I have almost all parts on the desk.
Unfortunately I purchased a parts kit organized buy a forum member, buzzforb not to name him (excellent kit btw) without checking the value of source resistors needed for my specific build.
Thing is I am building it with SJ/SK and remember having read the need of 0.25 Ohm instead of 0.5 Ohm (Walter W. as well as EUVL, iirc).
As I am getting tired of sourcing parts I was wondering what drawbacks I should expect when using the SJ/SK outputs with 0.5 Ohm source resistance.
Any help is highly appreciated.
Cheers,
Max
Hi all,
getting closer and closer to starting the build of the F5T as I have almost all parts on the desk.
Unfortunately I purchased a parts kit organized buy a forum member, buzzforb not to name him (excellent kit btw) without checking the value of source resistors needed for my specific build.
Thing is I am building it with SJ/SK and remember having read the need of 0.25 Ohm instead of 0.5 Ohm (Walter W. as well as EUVL, iirc).
As I am getting tired of sourcing parts I was wondering what drawbacks I should expect when using the SJ/SK outputs with 0.5 Ohm source resistance.
Any help is highly appreciated.
Cheers,
Max
None at all.
The value of the source resistors simply determines the bias at which the fets will operate. In an ideal world, you would have no source resistors and the diodes would be unnecesasary. Unfortunately, fets without Rs tend to be unstable with nothing to temper their response. This is the main purpose of the resistors.As a side note, they do help linearize the response with a bit of local feedback. What you give up with source resistors is gain. They eat part of it. The use if diodes is a clever trick to get some of this back. As they slowly turn on, they act as a bypass for current. This increase in current results in an increase in gain. The increase in gain means more feedback and as a result, better performance in terms of distortion. Also with the increase in current, you get a turbocharged output, capable of delievering more current to the load. Kinda of a win/win situtation. Really wonderful amp for low Z loads, which is exactly what it was developed for.
Your .5r worth of resistance means you bias will be between .8 and .9a fet. Lowering this simply to .25 would simply mean a doubling in bias per fet, nothing else. I personally think that is too much heat per fet, but i am a rookie in these things. Euvl runs his at a higher bias in the f5x, but he give some of the heat back in the form of lower rail voltages. He gains this voltage back in the form of a bridged amp.
Audio is about tradeoffs.
Thanks buzzforb for the help!
Actually I do have exactly what I need for my intended 48V rails... don't I ?
Actually I do have exactly what I need for my intended 48V rails... don't I ?
Yes. I also hope you have liquid cooling, because 48V x .8A is 38W per fet. That is smokin, especially considering the close proximity of the fets on the boards that are available. One channel is hitting close to 160W with perfect heat transfer. Dont expect a long lifespan.
If I go lower on bias if needed that simply means I'll never experience the "turbo-effect" as the diodes will never turn on, right ?
To a large extent, yes. In situations where you need greater voltage swing, I would recommend bridged amps or more fets with less bias per fet, like BIg Papa amps.
Hi Buzz,
Which do you think will sound better:
Consider F5T V3 with 4 pairs each mono and Heat sink is not a problem.
32V rail running at 0.8A each or 42V running at 0.6A each fets.
Any advise?
Regards,
Which do you think will sound better:
Consider F5T V3 with 4 pairs each mono and Heat sink is not a problem.
32V rail running at 0.8A each or 42V running at 0.6A each fets.
Any advise?
Regards,
Higher rail voltages tend to improve distortion as it increases Yfs/gain. I do not know the sweet spot of these fets in that regard. From what little i know(second hand knowledge), Nelson uses up to 32v rails in his big amps, but then again, they are bridged, giving higher swing capability into the load. It is easy enough to try. Give it a shot.