F5 Turbo Builders Thread - Page 50 - diyAudio
Go Back   Home > Forums > Amplifiers > Pass Labs

Pass Labs This forum is dedicated to Pass Labs discussion.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 23rd October 2012, 07:56 AM   #491
cdfac is offline cdfac  United States
diyAudio Member
 
Join Date: Aug 2009
Location: Iowa City
Quote:
Originally Posted by AudioSan View Post
its 0.28C/W pr SINK. so 0.14C/W pr side. they can handle a 2ch V2 fine.
thanks, i always appreciate your insight AudioSan. looking again at the numbers, i think you must be right about what is what.

but if you don't mind, i'd like some help understanding why a single 165mm(H)X 300mm(D) sink is roughly 10% worse at dissipating heat than a single 200mm*200mm. the former has roughly 20% more surface area, and i read elsewhere on this thread or another that a lower, wider sink is more efficient than a taller, narrower one. is a "square" actually the optimal profile here?
  Reply With Quote
Old 23rd October 2012, 09:34 AM   #492
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
Taller sinks have warm air wafting over the fins. That reduces the effectiveness of the upper parts of the fins.

A very rough calculation (model/simulation) based on using sqrt(height) gives some idea of what to expect.

Take a 50mm high 200mm wide sink. Let's say it has Rth s-a = 1C/W
Double it to 100mm high and the Rth s-a is likely to be around 0.7C/W (sqrt[50/100]=0.7)
Double again to 200mm high and Rth s-a is likely to be around 0.5C/W (sqrt[50/200]=0.5)
Double again to 400mm high and Rth s-a is likely to be around 0.35C/W
Very short and very tall heatsinks move further away from the sqrt rule.

Now instead double the width. All the fins in the 200mm wide version are running at the same temperatures as the 400mm wide version.

Remember, when Rth s-a is measured and specified, it is for the back plate at isothermal temperature. This means the interface surface of the whole backplate is at the same temperature (Ts). This means there are no cooler spots/areas to reduce effectiveness of the fins.

We need to modify the model to take account of delta T (Ts-Ta). The difference in temperature between A air (Ta) and S backplate (Ts). The sink manufacturers usually give direction in how to derive the delta T de-rating Factor. For delta T ~30C degrees expect DF in the range 1.3 to 1.4

The next correction is for non isothermal operation. I have not seen any manufacturer giving data on this correction.
Some independent software does allow investigation of this. I adopt a much simpler method. Make the extremities out at the corners of the heatsink be at a radius] of less than 10 times the thickness of the backplate. This gets the backplate surface to near the same temperature as the hot spot under the heating device. Copper heatsink can probably manage with a bigger ratio, maybe 12 to 15 times thickness.
This rule would require the 400mm wide heatsink to have a thicker backplate than the 200mm wide heatsink if one expects the modeling to predict operational temperatures.

BTW,
this radius to thickness also applies to the fins. Short fins are effective even though they can be relatively thin. Long fins must be thicker to be as effective. Just apply the 10:1 ratio for fin depth:thickness. This also explains why fins are often tapered and yet not lose much in the way of dissipation effectiveness.
__________________
regards Andrew T.

Last edited by AndrewT; 23rd October 2012 at 09:44 AM.
  Reply With Quote
Old 23rd October 2012, 12:01 PM   #493
diyAudio Member
 
Join Date: Feb 2009
and there might be rilled fins. then the surface area will be much greater.
  Reply With Quote
Old 23rd October 2012, 12:10 PM   #494
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
If I understand "rilled" then it depends on how the fins have been shaped.
If the fins "look" line they have been bent into their new shape then basically we have a longer fin of the same thickness. Yes that increases surface area but it also increases it's length:thickness ratio. It still needs to be able to conduct heat right out to the tip.

But there is another way to ripple the surface of the fin. This is to make one side wavy (or zig-zagged) and to make the other side wavy but out of phase this inserts narrowing of the fin without increasing the length. So it keeps the length the same and the surface area increases but the length:thickness ratio goes up if the necked thickness is used as reference.
I believe the first (in phase wavy) method is better then the necked version.

I think you exaggerate using the descriptor "much".

How much does the area actually increase for examples you have seen?
__________________
regards Andrew T.
  Reply With Quote
Old 23rd October 2012, 06:53 PM   #495
diyAudio Member
 
Join Date: Feb 2009
i was thinking of this.
Attached Images
File Type: jpg 9740hs_.jpg (85.2 KB, 435 views)
  Reply With Quote
Old 24th October 2012, 01:19 PM   #496
diyAudio Member
 
Join Date: May 2005
Location: Lisbon Portugal
Guys I need an opinion, I have been planing to build an f5t v 2 as I always tought I could use the extra steroids!

Now I've been thinking a bit better and I believe an higher bias f5 should do the trick.
I have a pair of rega rs3 which are 6ohm speakers with aprox 89db sensitivity. and would like to have at least 35-40W in class A and then some headroom in AB.

Acording to the f5 manual at 1.3A bias per device the math is:
2.6*2.6*8=54W and half of that is class A so around 27W
so for a 6ohm speaker to get to the 35-40W target I would have to bias each device between 1.7 and 2 amps using the same voltage. Correct? here come some questions:

Providing adequate heatsinking (I'm thinking 0.2W/C per channel, using a 300x210x40 heatsink) the trasistors should support this right?

Increasing the supply voltage means I would be able to bias the devices at a lower current and still get the same power in class A right?

Now, if I go the turbo route, I would have 2 devices per side and would only bias each at 0.9 or 1A to get to the desired power, and wouldn't need to increase the supply voltage right? Or should I increase the voltage and use lower bias?

So is it better to have 2 devices per side biased at a higher A or 4 devices at a lower bias?
  Reply With Quote
Old 24th October 2012, 01:38 PM   #497
diyAudio Member
 
Join Date: Feb 2009
Quote:
Originally Posted by MASantos View Post
Guys I need an opinion, I have been planing to build an f5t v 2 as I always tought I could use the extra steroids!

Now I've been thinking a bit better and I believe an higher bias f5 should do the trick.
I have a pair of rega rs3 which are 6ohm speakers with aprox 89db sensitivity. and would like to have at least 35-40W in class A and then some headroom in AB.

Acording to the f5 manual at 1.3A bias per device the math is:
2.6*2.6*8=54W and half of that is class A so around 27W
so for a 6ohm speaker to get to the 35-40W target I would have to bias each device between 1.7 and 2 amps using the same voltage. Correct? here come some questions:

Providing adequate heatsinking (I'm thinking 0.2W/C per channel, using a 300x210x40 heatsink) the trasistors should support this right?

Increasing the supply voltage means I would be able to bias the devices at a lower current and still get the same power in class A right?

Now, if I go the turbo route, I would have 2 devices per side and would only bias each at 0.9 or 1A to get to the desired power, and wouldn't need to increase the supply voltage right? Or should I increase the voltage and use lower bias?

So is it better to have 2 devices per side biased at a higher A or 4 devices at a lower bias?
thats what i did too. 1.7A bias to drive 6ohm. however. if you like it to run at higher power then 35W at 6ohm in class A/B. then you also need higher rail voltage.
if you want som A/B headroom, then twice the class A power is a minimum. under that is little to no point. that meens +/-32V anyways. so you realy want a turbo V1 or V2.
  Reply With Quote
Old 24th October 2012, 02:48 PM   #498
Tea-Bag is offline Tea-Bag  United States
not politcally affiliated
diyAudio Member
 
Tea-Bag's Avatar
 
Join Date: Feb 2006
Location: Kennebunk
Blog Entries: 7
I like the overall sound of the Turbo in any format better than the original, regardless # of outputs. I've tried them all. More laid back and grunty.
My setup uses 4 of outputs each side biased lower @ ~30v.
This way I keep less wattage on each hotspot and just crank the bias until your heatsinks max out, or get reaches 65c, which ever occurs first. I pretend it's a XP-30A
It probably goes to class B quicker this way, but ah well. Compromises.
  Reply With Quote
Old 24th October 2012, 04:22 PM   #499
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
35W into 6r0 requires ~20.5Vpk and 3.4Apk.
Bias set to slightly more than 1.7A satisfies the ClassA requirement for the 3.4Apk.
Supply that holds up to better than ~25Vdc satisfies the 20.5Vpk.
If you want more ClassA output then increase the bias current.
If you want more power then increase the supply voltage.

All that was posted in #496 para1 to para3 is correct.

I have a feeling that para 4 onward indicate you don't really understand power.
Stick to voltages and currents and don't confuse the understanding by including power.
__________________
regards Andrew T.
  Reply With Quote
Old 24th October 2012, 06:38 PM   #500
diyAudio Member
 
Join Date: May 2005
Location: Lisbon Portugal
Regarding transformer choice, NP says in the f5t article that va rating should be at least twice the dissipation of the amp.

With 4 devices per channel and 1A bias at 32v I et 32x8=256w dissipation, so I should go with a trafo with 512va plus extra headroom make that 600VA. Is
my math correct?
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
F5 Turbo Circuit Boards chchyong89 Pass Labs 2363 27th September 2014 02:10 PM
B1 builders thread Bas Horneman Pass Labs 696 14th July 2014 12:57 PM
F5 Turbo is posted Nelson Pass Pass Labs 635 6th April 2014 10:51 AM
F5 turbo problems Rob Dingen Pass Labs 17 11th January 2013 08:38 PM
F5 Turbo ............. if only !!!!!! a.wayne Pass Labs 23 21st January 2012 02:24 AM


New To Site? Need Help?

All times are GMT. The time now is 09:56 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2