F4 power amplifier

Jump to Latest
Variac

Variac

diyAudio Editor
Joined 2001
BurningAmp.org
Paid Member
How far is this from the airport?
Click to expand...

Well, if you aren't carrying too many F4's then you can take our rapid transit system, BART, directly from the airport now. Then switch to a bus to go over the hill to the site. Otherwise, a van from the airport shared with other passengers isn't too bad.. The driver can take you to the closest hotel to the event. Then take a taxi.

I'm sure we can get youto the event if you decide you really are coming. Places to stay at member's houses seem to be rare unless someone posts here that they can take someone..
 
Malotron

Malotron

Member
Joined 2005
19th century mad scientist look
Click to expand...

Captain Nemo, your F4 is ready!

A shot of the "bizarre heatsink tower" please.

-Mal

pic of maple/concrete rack


Albums I'm listening to through my F4:

Brokeback Field Recordings
Talking Heads "77"
Sparklehorse It's A wonderful Life
Neil Young Live - Massey Hall 1971
 

Attachments

  • rack.jpg
    rack.jpg
    93.5 KB · Views: 1,464
Variac

Variac

diyAudio Editor
Joined 2001
BurningAmp.org
Paid Member
Thanks guys for your very kind comments. I hope my wife feels similarly :whazzat:

I think she suspects though- it is being built on the kitchen table!

I have to leave the Bizarre Heatsink for a surprise at the Burning Amp show. It is more big than bizarre anyway..

Choky, I like to keep my picture off of the web so I won't be hunted down by angry ex-members, but you found it!! Now I'm ruined ;)

Variac
 
Variac

Variac

diyAudio Editor
Joined 2001
BurningAmp.org
Paid Member
This is an up to date one I believe.


http://www.diyaudio.com/forums/attachment.php?s=&postid=1263192&stamp=1185337158


Remember that Nelson changed the values for the resistors afterthe initial post

So any prudent builder would first go to Nelson's site as mentioned in the first post and find the file about the F4 , then download the entire NEW owner's manual

Then compare the BOM in question with the
schematic (Version 0 dated 6/4/07 I believe, but check for yourself.. )

the older BOM's are good resources for alternate parts- if they are changed to the correct value if required.

I was very brave and ordered different resistors , 3w resistors (got 5w) trimmer pots (bourns- mentioned in a different BOM, bridge, power caps, It was fun to mix and match and feel like I actually had an opinion of what is best!!

Digikey worked for me, except for the resistors, 'cause of my weird resistor desires!! and they sent me a new catalog twice as thick as my 5 year old one!!
 
Last edited:
S

Studley

Member
Joined 2005
Heatsink calc

A friend is helping me build my F4s (archetypal British understatement). He has done some calculations to determine how serious the heatsinks need to be but suggested I post the calcs on this thread to see if others think they stack up:

Each device is passing 0.43A so at 23v
(minus 0.2v
dropped across the resistor) the total power dissapated at
idle is 6 *
0.43 * 23 - 0.2 ) = 59W.

So, for a running temp of 50Deg, and room at 25deg, the
heatsink will be
at 25deg above ambiant. So the total thermal resistance
needs to be
25/59 or 0.424 C/W

The chip to case thermal resistance of the mosfet is 0.24
C/W, so for 6
devices thats 0.04 C/W, assuming efficient coupling from
the device to
the heatsink,

Assuming 1 C/W insulation for the device to heatsink,
again, thats in
parallel, so thats .16 C/W, so a total Chip to sink of 0.2
C/W, so that
leavs us needing

0.424 - 0.2 = 0.224 C/W


Any comments?
 
Tarasque

Tarasque

Member
Joined 2002
All aspects complete:

Rth____________________°K/W_____delta T
Heatsink_________________0,3_______20,7
Tmount-Heatsink___________0,1_______6,9
Isolator-Transistormounting__0,3_______3,5
Isolator-Transistor_________0,2________2,3
Transistor-Junction-case_____0,83______9,5__IRFP240 = 0,83


Dissipation W/transistor_____11,5____117watt total
amount of transistors________6
dissipation per channel______69W
dissipation stereo__________138W
rail voltage________________23V
total bias_________________1,5A

ambient T ________________25°C_____77°F
Junction T_______________67,9°C____154,21°F
Mounting temperature_____52,6°C____126,68°F
Heatsink temperature_____45,7°C_____114,26°F
 
D

diy2000

Disabled Account
Joined 2007
supermodel.awardspace.com
Variac said:
My power supplies and chassis are about done,,,

but I got my order from Digikey today. The mica insulators for the output devices are a tiny bit SMALLER than the metal on the back of the MOSFETS. It seems to me that the spacing should be enough, with the low voltage.. but if not the consequences would be very unpleasant.. What do you think?
Click to expand...

with np's amps you should always leave wide margins
-most of his amps run hot- use insulators bigger than
the devices thats what i reckon

cheers

;)
 
L

lurcher300b

Member
Joined 2005
lurcher.org/nick/audio
Hi, I am the person Studly has asked to do the build, can I just gheck something?

every day calculus :


U=23 x 2
I=,43 x 3 (3 verticals)

Ptot = 46 x 0,46 x 3 ~64W

Tmax ~ 50C
Tamb~ 25

Dt=25

Rth = (25 / 64) x 0,85 = 0,33


so-you'll have ~ 50C at heatsink
Click to expand...

Should that not be

Ptot = 46 * 0.43 * 3 = 59.34

(ignoring the small amount of power that will be dissapated in the resistors)

But I am not sure where

Rth = (25 / 64) x 0,85 = 0,33{/quote]

Comes from. Ok the Chip to case on the device is 0.85, but as you have 6 in parallel, then if you are using the total power diss, it should be 0.85/6 and subtracted not multipled.

Maybe I am reading this wrong, is the target a heatsink temp of 25 above ambiant or is it the chip temperature?
Click to expand...
 
Zen Mod

Zen Mod

Official Court Jester
Joined 2003
zenmod.in.rs/
Paid Member
lurcher300b said:
Hi, I am the person Studly has asked to do the build, can I just gheck something?



Should that not be

Ptot = 46 * 0.43 * 3 = 59.34

(ignoring the small amount of power that will be dissapated in the resistors)

But I am not sure where

Rth = (25 / 64) x 0,85 = 0,33{/quote]

Comes from. Ok the Chip to case on the device is 0.85, but as you have 6 in parallel, then if you are using the total power diss, it should be 0.85/6 and subtracted not multipled.

Maybe I am reading this wrong, is the target a heatsink temp of 25 above ambiant or is it the chip temperature?
Click to expand...


well- every-day calculus is example what is needed,what not ;
let's not over-engineer what is just for engineering .

forget about resistor dissipation in overall picture- there are just 6 of them,each with less than 0,1W dissipation

........ (25/64) x 0,85........
25 is Dt between heatsink and air [C]
64 is sum of dissipation [W]
0,85 is just experience factor , nothing else - where are summed usual losses etc.

presuming that mounting of output devices is "good enough" or "proper" (in other words) , shooting at heatsink temperature instead on chip temperature is completely sane ;

you must shoot chip temperature in case when output devices are much closer to their limits than in this case
Click to expand...
 
marginwalker

marginwalker

Member
Joined 2006
here's how i see it:

Rth(j-a) = (Tj max - Tamb) / Ptot

Ptot = 46 * 0.43 * 3 = 59.34

where:

T(j max) is the max of the junction temp we want to allow
Ptot is Vrails * Ibias
Rth(junction-ambient) = Rth(junction-case) + Rth(heatsink) = 0,83 (for IRFP240) + Rth

the matter is to know exactly which Tj you want... according to IRFP240 datasheet it seems that Tj max is 125°C but i guess that safe operating temp will be lower.

Let's say Tj max 100°C, so your heatsink should have Rth = 0,43°C/W


Is it correct? :confused:
 
You must log in or register to reply here.
Top Bottom