F5 power amplifier

Jump to Latest
#861
jackinnj said:
The plot which I published depicted the P-channel replaced. I then replaced the N-channel as well and the result is the same -- still peaks at around 500kHz
Click to expand...

The closest I've got is to get a 1 dB peak at about 800 KHz.

Personally I wouldn't bother with anything at that value, but yours
appears to be a little more severe.

Looks to me like you'll have to break down and install 2 compensation
caps. one across each of the 50 ohm resistors on the feedback loop.
(By the 50 ohm I mean the parallel 100 ohm resistors.)

Suggested initial value would be .0022 uF.
 
#862
Re: Complementary MOSFETs

EUVL said:

While waiting for the Fairchild EU group buy, which is forcast now from Digikey to be 10 weeks (!!! ), we have got a few Toshiba 2SK1530's / 2SJ201's to play with.

The pdf attached plots drain current vs Vgs referenced to bias point, with both values inverted for the P-device to ease comparison. The bias current is 2A for both FETs, Vgs is 2.38V for the N-device, and -2.30V for the P-device. Transconductance at bias is about 4.5 S.

Red line is 2SK1530; light blue crosses are 2SJ201. It was a hell of a job to make the two sets of data visible from one another in excel.

A picture says a thousand words.


Patrick
Click to expand...


Hi Patrick

Did you ever try these devices in the design?
I would be interested to know how they faired

Cheers

Tim
 
#863
Prototype is 60% ready (PCBs, matched FETs, heatsinks, mechanical parts, ....), just waiting for PRP resistors to arrive from the US. Probably at German Custom right now.

Shall report again when I finish soldering & measuring.
Do not have an distortion analyser yet, neither Fairchild FETs, so no comparison to the original ....

I hope I do not end up putting caps on the feedback path.
Otherwise no point doing "current feedback" ......


Patrick
 
#866
Re: Re: Mega Power Hyper Balanced F5 Amp

Magura said:



That's not a compromise 😎

cute idea😀

Magura 🙂
Click to expand...


I really like the idea of a simple amp that is capable of driving low impedance loads.

I could probably even put a slow quiet fan on the heatsinks, but the (probably custom) >2KVA transformer + 100,000's of uF per channel would be a killer of a cost.

Cheers

Tim
 
#867
I wonder if the board can be 10-15 cm away from the outupt transistors . Pratically speaking may I have 15cm lenght wires ( max) between the output Fets and the board . I want to save a 2sJ109 and build a " central " board for both channels in the chassis .
In that case I would be able to use 2sk389 also ...

Thanks
 
#868
stefanobilliani said:
I wonder if the board can be 10-15 cm away from the outupt transistors . Pratically speaking may I have 15cm lenght wires ( max) between the output Fets and the board . I want to save a 2sJ109 and build a " central " board for both channels in the chassis .
In that case I would be able to use 2sk389 also ...

Thanks
Click to expand...


if you look at pics on 6moon , you'll see pretty long traces , as usual - with all FW amps (if not even all PL) ;

anyway - I think that hassle needed for finding 2+2 tiny Jfets is worth for playing safe ......... if not just for saving 109/389 for something more critical .....

YMMV ......... :rofl:
 
#869
Zen Mod said:



if you look at pics on 6moon , you'll see pretty long traces , as usual - with all FW amps (if not even all PL) ;

anyway - I think that hassle needed for finding 2+2 tiny Jfets is worth for playing safe ......... if not just for saving 109/389 for something more critical .....

YMMV ......... :rofl:
Click to expand...

Ok I have been able to find 2sJ74 V . I also have 2sk170 BL and 2sk170 V . So it should be ok . Lets hope to receive them .

:clown:
 
#870
I wonder if the board can be 10-15 cm away from the outupt transistors . Pratically speaking may I have 15cm lenght wires ( max) between the output Fets and the board
Click to expand...

stefanobilliani ... I use that approx. of wire between mainpcb, and output- board, without any problems at all.

Jesper.
 

Attachments

  • pic2-large.jpg
    pic2-large.jpg
    27.8 KB · Views: 2,512
#873
Re: Thermistor

lykkedk said:


stefanobilliani ... I use that approx. of wire between mainpcb, and output- board, without any problems at all.

Jesper.
Click to expand...

Thanks alot . I am making little boards around the output chips now . In the F5 the jfet are both very close to the input , I am indecided if following that approach or put them more in 2 different boards each for Pchannel side and Nchannel side , close to the big chips .

apassgear said:
According to the Pic’s of the F5 the thermistor that Nelson seems to be using is a Panasonic, available from DK

http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=PNT117-ND
Click to expand...

Would be lovely to organize a gorup buy for that .
 
#877
In post #781 ( http://www.diyaudio.com/forums/showthread.php?postid=1513301#post1513301 ) I presented a compact PCB layout for F5 with Q3 and Q4 placed very close to each other. Here is a new version (bottom view, 2 x 10 cm) with Q3 and Q4 placed far apart to allow easier use of two heatsinks per channel. Parts nomenclature is the same as in previous version. As Mr. Pass said, circuit is simple enough for p2p approach but PCB makes bias/DC offset adjustments easier.
 

Attachments

  • f5-long.gif
    f5-long.gif
    12.7 KB · Views: 2,813
#878
juma said:
In post #781 ( http://www.diyaudio.com/forums/showthread.php?postid=1513301#post1513301 ) I presented a compact PCB layout for F5 with Q3 and Q4 placed very close to each other. Here is a new version (bottom view, 2 x 10 cm) with Q3 and Q4 placed far apart to allow easier use of two heatsinks per channel. Parts nomenclature is the same as in previous version. As Mr. Pass said, circuit is simple enough for p2p approach but PCB makes bias/DC offset adjustments easier.
Click to expand...


Very nice , can I have a pair of them ?
 
You must log in or register to reply here.
Top Bottom