Nope....fantastic boards, but since this is a beast those wishing to stretch V3 to full class A will need other option like split input and power boards...like BA2/BA3....
Which is easy to do with this layout ...just clip the two ends of the board and voila....split input board.
yes they are
i'm using 250mm x 120mm x 73mm heatsinks for a JLH 10W (pr. ch.) and 500x150x73mm for F5 on steroids
Nope....fantastic boards, but since this is a beast those wishing to stretch V3 to full class A will need other option like split input and power boards...like BA2/BA3....
Which is easy to do with this layout ...just clip the two ends of the board and voila....split input board.
even easyer to clip them at the drawing bord
yes they are
My F5 is pretty compact, like Firstwatt, 300cm deep /12cm high /8cm fins
runs at around 50ºC in Lisbon Summer.
Standard power rails !
Me have Aleph-J monaurals with 400mm heatsinks on the left and right side.
Eeh, but just a hair over 2.5" high !
My F5 is pretty compact, like Firstwatt, 300cm deep /12cm high /8cm fins
runs at around 50ºC in Lisbon Summer.
Standard power rails !
my F5 will run å liiiitle higher bias
almost dobble it's dual outputsMe have Aleph-J monaurals with 400mm heatsinks on the left and right side.
Eeh, but just a hair over 2.5" high !
hehe
sounds like modu's dissepante 2U 400mm
Would a "Driver board" and a pair of "Output boards" make things more flexible? And easier to accommodate monoblocks and stereo chassis? No doubt. But it also adds a bunch of cost...
What about having a board with seams and the ability to break away the outputs? Like the BrianGT Aleph board. This is more complicated than I thought at first glance because the thermistor circuit need to travel with the outputs, but I think it could be accommodated into your board with a minimum of fuss.
Measured Heatsink Performance
I made some temperature measurements of my F5 amplifier which uses 7" long HeatsinkUSA E007 heatsink. As shown in F5 Power Amplifier post 10205 http://www.diyaudio.com/forums/pass-labs/121228-f5-power-amplifier-1021.html#post2453420, the heasinks consists of a heat tunnel produced by connecting halves of the E007 heatsink. The heatsink is cooled with a Noctua NF-S12-B fan. The F5 amplifier has a standard bias current of 1.3A, resulting in about 62W per channel power dissipation.
Here are the temperature measurements:
22C Inlet air temperature
32C Heatsink near inlet and near mosfet
36C Heatsink near exhaust and near mosfet
14C/62W => R = .226 C/W for 1/2 E007
A full E007 would provide R = .113 which is enough for the F5-Turbo at 180 watts dissipation.
I made some temperature measurements of my F5 amplifier which uses 7" long HeatsinkUSA E007 heatsink. As shown in F5 Power Amplifier post 10205 http://www.diyaudio.com/forums/pass-labs/121228-f5-power-amplifier-1021.html#post2453420, the heasinks consists of a heat tunnel produced by connecting halves of the E007 heatsink. The heatsink is cooled with a Noctua NF-S12-B fan. The F5 amplifier has a standard bias current of 1.3A, resulting in about 62W per channel power dissipation.
Here are the temperature measurements:
22C Inlet air temperature
32C Heatsink near inlet and near mosfet
36C Heatsink near exhaust and near mosfet
14C/62W => R = .226 C/W for 1/2 E007
A full E007 would provide R = .113 which is enough for the F5-Turbo at 180 watts dissipation.
Only 100 watt?
It's a PP class-A amp, so according to THIS it's around 40% efficient. 100 watts output will need about 250 watts of dissipation... Per channel...
Although we don't have a direct quote to work from, let's see what Papa says about this, from the original F5 manual; bold is mine.
Nelson Pass said:
Inferring from this it's a safe bet to say that his sample heatsink, which is 48 square inches (120 square centimeters) is required for every transistor!!
So, our worst-case scenario F5 v3, biased fully class-A, with 8 transistors per channel, and then 16 in a stereo chassis, you come up with 768in^2, or 5.3 square feet (1.6 square meters) of heatsink required.
So, a normal-looking, flat back heatsink, convection cooled is going to be staggeringly large. The heatsink profile will need to be much more complex. And probably fan-cooled.
Somebody please correct me if my back of the napkin calculations are incorrect.
Last edited:
Only 100 watt?
It's a class-A amp, so it's basically 25% efficient. 100 watts output will need 400 watts of dissipation. Although we don't have a dorect quote to work from, let's see what Papa says about this, from the original F5 manual; bold is mine.
Inferring from this it's a safe bet to say that his sample heatsink, which is 48 square inches (120 square centimeters) is required for every transistor!!
So, our worst-case scenario F5 v3 with 8 transistors per channel, and then 16 in a stereo chassis, you come up with 768in^2, or 5.3 square feet (1.6 square meters) of heatsink required.
So, a normal-looking, flat back heatsink, convection cooled is going to be staggeringly large. The heatsink profile will need to be much more complex.
Somebody please correct me if my back of the napkin calculations are incorrect.
and you wont get the heat spread over the sinks with the outputs mounted on the PCB
How must building two chassis be cheaper than one ......
Precisely JV ...
Split boards mean wires , mono blocs mean additional cost .........
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.