F3 Builders Thread

flg said:

Cost :confused: ...I'll total somethin up but it isn't cheap...

Well, what I've mentioned so far for Cs and expensive Rs is about $225 for 2 ch. ($100 of that is the PPT Thetas for the output!)

I've had a suitable PSU for a while but it would not be cheap to duplicate. I got the Xfrmr from an my old 5 ch amp that blew-up. Must be about 625-750VA, 2 monster shotcky diodes $??, 4 39,000uF big blue comp Grade caps and a Hammond 10mH, 10A choke :D Probably $250-300

I've not decided what H.S.s to use yet but most of my set-ups run about $120 for something like this... Plus the PCB $15

So I'm in it for about $600 and I'm not done yet... :xeye:
 
Don't stress too much about cost. I just wanted to get a rough idea so I can start to budget for mine.
I need to finish a JLH amp and Naim clone before I can commit spending on another amp.
I have finished the JLH but need to build a case, then test it. For the Naim clone I need to finish the second channel plus buy transformers (I will probably also use these transformers on the F3 for initial testing).

I will probably do it in stages over a few months for example. I already have the 220uF caps (which are actually 470uF caps) so I just need the output caps and film caps.
I am still not sure how much money I am prepared to spend on resistors yet. Ultimately I would like to buy Caddocks, Vishay Z foil, Vishay Dale cmf, and Mills.
 
Adjusting the "ghost load"

Fellow F3 builders,

In the original setup of my F3 I calibrated P2 to nearly a minimum value, setting the total distortion on the output at the lowest possible level. (load= 8 ohm + 1mH in series + 1uF parallel; 1kHz@ 1W)
Now I've been listening to different speakers I'm not sure if this is the way to go. Reducing the ac feedback of the output on the setting of the current source seems to sound more natural and less thin (but maybe less controlled...)
Can anyone share his/her ;) findings??

BTW the work on a "F6" a two cascoded LU1014D per channel version of the F3 is suffering from heatproblems (even in wintertime....) :( and stability problems. Real life differs from Spice..
 
thanh1973 said:
I am guessing that maybe you like the sound of 2nd Harmonic distortion (which is quite natural and musical), so in that case rather than tune it for lowest distortion try tuning the pot by ear while having a favourite cd playing (be careful though electricity can kill).

Well, that's my point, tuning P2 doesn not change the ratio between 2nd 3th etc.., only the amount of. And I can't believe I like more distortion :rolleyes:
Now I'm thinking of testing on a real life dynamic load (a high compliant, long stroke woofer in a panel) to see what happens...
 
SET triode amps run up to 2% 2nd harmonic distortion and are loved by many people.

A 12 string acoustic guitar is the equivalent of a 6 string acoustic guitar with 50% 2nd harmonic and sounds absolutely beautiful.
So I am not surprised at all that you like the sound with more 2nd Harmonic Distortion.
 
Member
Joined 2006
Paid Member
Telstar,

...but what heatsinks are you using ?

...mine are 300x150x80 mm rated at .25ºC/W ...which is the "minimum" recomended by NP....( read it in one of his posts here ).
On top of that the rest of the case is 5mm alu which will help a little.

But of course i'm interested to know before i finish the build.


:cool:
 
CeeVee said:
Telstar,

...but what heatsinks are you using ?

...mine are 300x150x80 mm rated at .25ºC/W ...which is the "minimum" recomended by NP....( read it in one of his posts here ).
On top of that the rest of the case is 5mm alu which will help a little.

Yeah, that's the size. My case is all alu, so it gets all hot ;)
I'm going to buy an infrared thermometer to check better, but I think it goes +30-40 over ambient.

Also, a good idea is to have tall feet.
 
Member
Joined 2006
Paid Member
The idea i have is that if it is in the region of 50ºC at the heatsink fins it's fine.

Hope some of the more informed guys here can give us some input on this.

My take , from the pictures of the real deal ( Firstwatt F3 ) is that
the size is right, haven't measured the real one.

The diference is that the original is black, which helps heat dissipation efficiency quite a bit.

:cool:
 
Hi all,
There seems to be some confusion about how much heat this F3 amp actually generates.

For each channel, the rail is close to 46 volts, add few more for diodes, resistors, transformer, etc to a neat round figure of 50V (dc).
Okay so far?

The current thru each channel is, or should be, approx 1.6 amps, so consumption is approx 80 Watts/ channel and with no power going out to the speakers, this is all dissapated by the heatsink (plus case).

Now, heatsinks - If you look at the Conrad sinks, for example, the 350mm long X 150mm high 15" X 6") flat back have a thermal rate of 0.28 *C/W (0.28 degree centigrade temp rise for each watt injected into it)
However, it's never this simple -  this 0.28 is measured at 80*C and as it's going to work about 50*C, the heatsink is LESS efficient and so the 0.28 is derated by a factor of about 1.15 (15%) so this becomes 0.32*C/W at 50*C. 
If your having a nice day and the ambient temp is 25*C, then you're looking for a temp rise of about 25*C. 

Pumping 80 watts into a sink of 0.32C/W will actually give you a calculated temp rise of about 27*C - close enough!

An example - Nelson's F3 heatsinks are about 350 X 125mm (15" x 5") with the 5 horizontal fins and the thermometer rises from room temp of 23*C to 46*C in the middle groove to give a temp rise of 23*C with no output into speakers - the amp does have a big 10mm slab of Ali across the front and also has it's top off that does increase dissapation a bit, but you get the idea.

The F3 clone runs the bigger Conrad sinks and it's temp rise is about 17*C = an R of 0.22*C/W. The developing shunt supply will increase consumtion up to about 2.5A/ ch (at about the same 50V total), then dissapation goes up to about 120W (per channel) which will increase the temp rise to about 25*C - this is a bit of a "guess-timate" and it'll be interesting to see what actually happens in practice..

Incidently, for some reason, this amp does show a quite clear difference with the internal wiring! - so I suggest you fit good quality wire right from those ultrafast diodes that I know everyone is now using(!) to the output terminals.
 
Member
Joined 2006
Paid Member
I'm using the Conrad heatsinks on my F3 clone and I'm seeing about a 30 to 32 celsius rise over ambient. The front has a couple good chunks off aluminum bar as part of the design and these also get very warm. I don't run the amp if the ambient temp is over 25 celsius.