more seriously, I have a french document on file that shows the power needed to match the SPL of instruments, it's a scary one to reads..don't think it need translation just go straight to the table.
Capacit dynamique d'un systme audio
I am not planning on a 120db in my room but still, the power requirements in the bass for the average/low sensitivity speaker is huge.. that's the very reason I went active!
Capacit dynamique d'un systme audio
I am not planning on a 120db in my room but still, the power requirements in the bass for the average/low sensitivity speaker is huge.. that's the very reason I went active!
you haven't listened to Clayderman's latest jewel I guess?
nope, I can understand music with a difference of that much, that would not be unusual, but a track that is at that level but then continuously at nearly 20db higher for that long would seem to me a very difficult track to record well too
I bought for the LPUHP GB as we need solid sections not for To2XX parts etc but rather for heatsink to PCB and threw in the rest to get bulk direct from the distributor. shoot me a PM if you want a bit, after i've cut to suit the group i'll probably sell a touch more of mine and then keep the rest
Hi Guys,
Well, the very last package shipped out the other day, and the extra 100 mosfets arrived last night, so I'm in the home stretch now!
I will have shipped a grand total of 158 packages by the time this whole thing is done, which is pretty mind boggling, especially looking back on it.
I'm about to send out the last payment requests for the remaining mosfets, so I've attached the mosfet order list with total and a small matrix for calculating shipping costs.
Please follow the instruction in the message carefully, and check the attached list for all the details.
Cheers,
Owen
Well, the very last package shipped out the other day, and the extra 100 mosfets arrived last night, so I'm in the home stretch now!
I will have shipped a grand total of 158 packages by the time this whole thing is done, which is pretty mind boggling, especially looking back on it.
I'm about to send out the last payment requests for the remaining mosfets, so I've attached the mosfet order list with total and a small matrix for calculating shipping costs.
Please follow the instruction in the message carefully, and check the attached list for all the details.
Cheers,
Owen
A very interesting document lolo.
Kind of gives us pause to think that a greater efficiency in the loudspeaker gives us quite a bit more potential realism in terms of dynamic range. Potentially it also could give a person a cleaner reproduction of the sound.
I demonstrated a system of that nature in the beginning of the year. Very beguiling in it's ability to capture your ears. First system I have made in many years that I did not want to turn off. It's baseline efficiency was 96 db/watt.
Kind of gives us pause to think that a greater efficiency in the loudspeaker gives us quite a bit more potential realism in terms of dynamic range. Potentially it also could give a person a cleaner reproduction of the sound.
I demonstrated a system of that nature in the beginning of the year. Very beguiling in it's ability to capture your ears. First system I have made in many years that I did not want to turn off. It's baseline efficiency was 96 db/watt.
Opc, great thanks for your work!
Yeah, took time, but at last amps are ready. I'm not used to SMD parts, so a couple of times i spent several minutes crawling on the floor searching dropped resistor. Luckily here's no carpet
Can anyone give some hints about heatsink size? I am planning to run amps at 50V (or less) on FETs, bias 350mA, passive cooling. Would be good to have also some overhead. If i understood correctly then according to Basic Theory i need around 6000cm3 heatsink. Is that number realistic?
kiwik
Yeah, took time, but at last amps are ready. I'm not used to SMD parts, so a couple of times i spent several minutes crawling on the floor searching dropped resistor. Luckily here's no carpet
Can anyone give some hints about heatsink size? I am planning to run amps at 50V (or less) on FETs, bias 350mA, passive cooling. Would be good to have also some overhead. If i understood correctly then according to Basic Theory i need around 6000cm3 heatsink. Is that number realistic?
kiwik
350mA of output bias with +-50Vdc = 35W of Pq.
An effective Tc that allows good reliability is ~60°C requiring a Ts of ~50°C
If your ambient is 25°C then delta Ts-a is ~25Cdegrees. At this delta expect the sink de-rating to be somewhere around 1.3 to 1.4
Rth s-a needs to be ~25C / 35W / 1.35DF ~=0.5C/W for one channel.
Note how many assumptions had to be made to get to 0.5C/W.
You need to insert your own operating conditions. Simply stating +-50V & 350mA does not get you close to the required solution.
An effective Tc that allows good reliability is ~60°C requiring a Ts of ~50°C
If your ambient is 25°C then delta Ts-a is ~25Cdegrees. At this delta expect the sink de-rating to be somewhere around 1.3 to 1.4
Rth s-a needs to be ~25C / 35W / 1.35DF ~=0.5C/W for one channel.
Note how many assumptions had to be made to get to 0.5C/W.
You need to insert your own operating conditions. Simply stating +-50V & 350mA does not get you close to the required solution.
Last edited:
thanks Andrew.
0.5C/W is not small value, I was planning on these sinks which are fairly large, but maybe I have to reconsider..
Radiateurs - Radiateur dissipateur thermique anodisé Noir 300x120x40
0.5C/W is not small value, I was planning on these sinks which are fairly large, but maybe I have to reconsider..
Radiateurs - Radiateur dissipateur thermique anodisé Noir 300x120x40
0.41C/W is bigger than 0.5C/W. It will allow the output stage to run cooler.thanks Andrew.
0.5C/W is not small value, I was planning on these sinks which are fairly large, but maybe I have to reconsider..
Radiateurs - Radiateur dissipateur thermique anodisé Noir 300x120x40
However, from the pic it looks like the backplate is quite thin. A few devices located well away from many of the fins will not cool as efficiently as a thicker backplate.
This depends... the tab on the mosfets is electrically tied to the output of the amp. This means you have two options:
1. Electrically isolate the mosfets from the heatsink
2. Tie the mosfets electrically to the heatsink and isolate the entire heatsink from everything else.
Option 2 allows better thermal performance as you can simply use a good thermal paste and get very low thermal resistance from case to sink. Electrically isolating a heatsink from a chassis, however, is not a trivial task. It would have to be an internal heatsink, fully protected from prodding fingers or falling metal object, and also fully supported and mounted with non conductive materials, likely plastic, which may break easily
It's probably easier to just isolate the mosfets unless you really have a good game plan for an complete isolated heatsink.
The measurements I ran on the bench were all with the mosfets tied directly to the heatsink.
Cheers,
Owen