I get the heat outside by running two coolant lines to a raiator outside. Amps inside, heat outside. The lightweight version is for outdoor use and needs to be moved quickly in the event of monsoon rains and for storage when not in use. The TEC is merely to make the dies cooler. If I can supply better cooling why wouldn't I? It would allow me to ask more questions like, "with a Honda Civic radiator and a 600gal/hr pump, how much bias can we run?
Some time ago, Cray Computors had some model like YMP or something that, acttually looked like a giant phone booth. The one on display in their lobby actually did have glass panels all over it. it was totally filled with Flourenert. A liquid flourocarbon substance that is inert (does not conduct electrons). The boards were in the liquid. The liquid has a boiling point that is reasonably low. When the hot devices cause the fluid to boil, the boiling bubbles carry away all the heat that created the bubble as the bubble floats to the top. That bubble was immediately replaced by cooler fluid not hot enough to form another bubble yet. The tank was chilled by an external cooler.
This is called nucleate boiling. The device temp is very well regulated by the boiling point of the fluid. If a hot device were surrounded with liquid, it would be below boiling point. If it's above boiling point it would have to be surrounded with gas. but that is unlikely due to the disire for all the gas to rise to the surface and more fluid to take it's place.
There are a few caveates to this. Flourenert was about $350-$550 a US Gal
Flourenert produces a poison gas when burned (not boiled). I had a liquid burn in project that used $20,000 worth of this stuff 
This is called nucleate boiling. The device temp is very well regulated by the boiling point of the fluid. If a hot device were surrounded with liquid, it would be below boiling point. If it's above boiling point it would have to be surrounded with gas. but that is unlikely due to the disire for all the gas to rise to the surface and more fluid to take it's place.
There are a few caveates to this. Flourenert was about $350-$550 a US Gal
Flourenert produces a poison gas when burned (not boiled). I had a liquid burn in project that used $20,000 worth of this stuff 3M makes a lot of these liquids and you can pretty much specify your boiling point. Many of them will not conduct electricity. Obviously this is a necessity We were going to use this in an inverter so that we could get the entire size and weight way down, then my boss started acting up more than usual and I took on a job as stay at home dad. Anyway, you can also use these liquids in heatpipes to guarantee a lower temp on the device.
The one we were looking at was used in fire protection systems so if your server room caught fire this liquid would put out the fire, but not damage the electronics.
Uriah
We were going to use this in an inverter so that we could get the entire size and weight way down, then my boss started acting up more than usual and I took on a job as stay at home dad. Anyway, you can also use these liquids in heatpipes to guarantee a lower temp on the device.The one we were looking at was used in fire protection systems so if your server room caught fire this liquid would put out the fire, but not damage the electronics.
Uriah
Awhile back I asked one of the computer whiz kids about getting the noise out of the computer (fans)and maybe trying to get the heat from the amps outside in summer - I think he's still laughing at me.
Apparently, this conundrum is "pretty small beer" (summer/winter trick and a few hundred watts), and in comparison to the examples of Owen, flg, Uriah, etc it must seem really basic.
It seems that longer runs of cooling tubes are quite normal and even don't seem to need pumps (?) and the bigger surprise is that if you can use a standard consumer system, they're not particularly expensive - much of it's rather neat engineering.
I feel so "out of place" in these computer worlds and ignore the "looks" from the kids but it does get results and they do need customers, ol' farts and all!
Curious thing - many of them seem to have obvious hearing problems and are still in school - not future audiophiles, I doubt! Rather disturbing, really.
Apparently, this conundrum is "pretty small beer" (summer/winter trick and a few hundred watts), and in comparison to the examples of Owen, flg, Uriah, etc it must seem really basic.
It seems that longer runs of cooling tubes are quite normal and even don't seem to need pumps (?) and the bigger surprise is that if you can use a standard consumer system, they're not particularly expensive - much of it's rather neat engineering.
I feel so "out of place" in these computer worlds and ignore the "looks" from the kids but it does get results and they do need customers, ol' farts and all!
Curious thing - many of them seem to have obvious hearing problems and are still in school - not future audiophiles, I doubt! Rather disturbing, really.
f5- ps question
hey ya guys
perhaps a dumb quesiton but what the heck;
I'm looking at the First Watt PS 0 R1 from 2005-looks like the template that peter daniels is recommending for the F5- ie CRC with dual bridges etc. On the primary side of the transformer(I'm using the GB 300va 18+18 -nice piece ,well priced) the schematic shows a capacitor connecting(bridging) the two red leads and the two blacks leads intersecting this "bridge" with thermistors.
I understand the thermistors- slows the inrush of current-
why the cap? does that establish a virtual ground? Is the value of the cap important relative to the specifics of the transformer- does the cap need to be there or could you just hook up the black leads through the thermistor?the cap is listed as .0033 line- is this mfd or farads=3300mfd?
inquiring minds wanna know
Throw this old dog a bone
thanks for your input
rob
hey ya guys
perhaps a dumb quesiton but what the heck;
I'm looking at the First Watt PS 0 R1 from 2005-looks like the template that peter daniels is recommending for the F5- ie CRC with dual bridges etc. On the primary side of the transformer(I'm using the GB 300va 18+18 -nice piece ,well priced) the schematic shows a capacitor connecting(bridging) the two red leads and the two blacks leads intersecting this "bridge" with thermistors.
I understand the thermistors- slows the inrush of current-
why the cap? does that establish a virtual ground? Is the value of the cap important relative to the specifics of the transformer- does the cap need to be there or could you just hook up the black leads through the thermistor?the cap is listed as .0033 line- is this mfd or farads=3300mfd?
inquiring minds wanna know
Throw this old dog a bone
thanks for your input
rob
Re: f5- ps question
Rob- That capacitor is there to reduce noise coming in from the AC line. The value is in microfarads, so that's equivalent to 3.3nF if you like. "line" just indicates that it's across the two legs of the AC line. Its value isn't critical, and the amplifier will work well without it. It's one of those things that an experienced designer will put in to help out... particularly when it costs $.20.
edit: I see NP answered while I was distracted... and, Ah!.. "line rated" is the more correct answer. I learn something new every day!
Rob- That capacitor is there to reduce noise coming in from the AC line. The value is in microfarads, so that's equivalent to 3.3nF if you like. "line" just indicates that it's across the two legs of the AC line. Its value isn't critical, and the amplifier will work well without it. It's one of those things that an experienced designer will put in to help out... particularly when it costs $.20.
edit: I see NP answered while I was distracted... and, Ah!.. "line rated" is the more correct answer. I learn something new every day!
transformer question and supply caps?
I am in the process of building a f5 anybody have any ideas where i can get good supply power caps[anybody use any mundorf caps?]also i was going to buy transformers from antek[any input?] I am also building a aleph 3 it calls for a 21v trans,but antek only has 20v or 22v which one would be better?thanks in advance peter
I am in the process of building a f5 anybody have any ideas where i can get good supply power caps[anybody use any mundorf caps?]also i was going to buy transformers from antek[any input?] I am also building a aleph 3 it calls for a 21v trans,but antek only has 20v or 22v which one would be better?thanks in advance peter
A have made it as well.
Just now I am listening with my KEF IQ3 which I normally use as computer speakers in my “home office when my daughter is not at home”. I admire the sound. Very detailed, very firm bass, etc.. Thanks to Nelson and to all gurus in this thread.
I am going to compare it with my Accuphase+B&W next weekend. (I hope the Accu will be the winner).
Just now I am listening with my KEF IQ3 which I normally use as computer speakers in my “home office when my daughter is not at home”. I admire the sound. Very detailed, very firm bass, etc.. Thanks to Nelson and to all gurus in this thread.
I am going to compare it with my Accuphase+B&W next weekend. (I hope the Accu will be the winner).
Same details, following the resent discussions.
Room temperature Tr= 22.5 (quite comfortable here in France)
The heatsink temperature (measured on the outer side) = 52.5 (0.3 C/W, two transistors mounted on the distance of about 10 cm )
The transistor temperature (measured on the mounting screw) = 65.9
The transistor temp (measured on the plastic body) = 69.2
Seems OK at the bias current of 1.2 A.
What I have noticed after some very preliminary tests:
Some drift of the voltage bias. I have noticed that there is a drift of the voltage on the R11,12 of about 10 mV resulting in approximately the same change of the output voltage (the input is shorted). Have not noticed any influence of this drift on the sound.
Room temperature Tr= 22.5 (quite comfortable here in France)
The heatsink temperature (measured on the outer side) = 52.5 (0.3 C/W, two transistors mounted on the distance of about 10 cm )
The transistor temperature (measured on the mounting screw) = 65.9
The transistor temp (measured on the plastic body) = 69.2
Seems OK at the bias current of 1.2 A.
What I have noticed after some very preliminary tests:
Some drift of the voltage bias. I have noticed that there is a drift of the voltage on the R11,12 of about 10 mV resulting in approximately the same change of the output voltage (the input is shorted). Have not noticed any influence of this drift on the sound.
Re: transformer question and supply caps?
for caps - check Apex Jr. site
elecres said:I am in the process of building a f5 anybody have any ideas where i can get good supply power caps[anybody use any mundorf caps?]also i was going to buy transformers from antek[any input?] I am also building a aleph 3 it calls for a 21v trans,but antek only has 20v or 22v which one would be better?thanks in advance peter
for caps - check Apex Jr. site
Alexyz:
Do you have the thermal comp circuit in your amp? If so, is the thermistor close enough to the FET? Is it the correct thermistor?
My F5 doesn't exhibit any drift from what I can see, other than the rails moving slightly throughout the day as the line voltage varies. I just checked it this morning after a good three months of running, and it's still right where I had set it originally.
Congrats on finishing it up! It's a superb amp.
Cheers,
Owen
Do you have the thermal comp circuit in your amp? If so, is the thermistor close enough to the FET? Is it the correct thermistor?
My F5 doesn't exhibit any drift from what I can see, other than the rails moving slightly throughout the day as the line voltage varies. I just checked it this morning after a good three months of running, and it's still right where I had set it originally.
Congrats on finishing it up! It's a superb amp.
Cheers,
Owen
P/S Caps
elecres,
for caps, try Matt Campbell ay " the Audio Capacitor Company" in England - has Mundorf, Jensen and those F&T's at very good prices - excellent value.
Zenmod,
How's your back doing this week?
If you can manage it, you might ask someone from the US to send you a tube of "Hot Hemp Muscle Rub" from the "Merry Hempsters" store in Oregon USA - costs about $3 -4.
It's marvellous at reducing back pain (and any other muscle aches) - good for plumbers, squashplayers, electronic techs! - everybody, actually - extraordinary product - no hallucigens unfortunately!
elecres,
for caps, try Matt Campbell ay " the Audio Capacitor Company" in England - has Mundorf, Jensen and those F&T's at very good prices - excellent value.
Zenmod,
How's your back doing this week?
If you can manage it, you might ask someone from the US to send you a tube of "Hot Hemp Muscle Rub" from the "Merry Hempsters" store in Oregon USA - costs about $3 -4.
It's marvellous at reducing back pain (and any other muscle aches) - good for plumbers, squashplayers, electronic techs! - everybody, actually - extraordinary product - no hallucigens unfortunately!
TeaBag stopped by today with Lbayjr.
We had a good listen to a bunch of Steve Hoffman mastered stuff. Its easy to become less a fan of a certain group or style and more a fan of anything Steve has remastered. Its really incredible.
Anyway, we heard TeaBag's F5 through the Wavelength pre and through the Lightspeed. Really great amp. Grabs the Audio Nirvana speakers by the throat and shows them who is boss. Big difference I noticed from the solid state I have been using is that the attack on the drums with the F5 is real tight, not flubby. Cymbals were awesome! Not a one note bass, not a one not cymbal anymore. Also found out how important good speaker wire is. Always thought it was a bunch of snake oil and I am sure a bunch of it is pure oil but Larry had some Kimbers and they replaced my braided silver plated copper. Seems the Kimbers are probably made of the same stuff, but way more strands than my 3 and nice connectors on the ends. Mine is just bare wire. Instant improvement in sound with them in place. I am sure with $30-$50 I could make something that will be close.
Both preamps had strong suits and I really had fun listening to both. Junk CDP but hey we deal with what we have and mine is a cheapo Technics. It has sounded better when a DAC was on it so thats in the future, then maybe a better CDP after the amp transforms itself to an F5
Lots of fun. Wish you guys had been there.
Gotta say that maybe the heatsinks need a preheater or something. It took a good hour to get the best sound out of the F5 and the sinks never got HOT. They got good and warm, but the 10 second test was a no flincher Oh, and his sinks are not enormous. Maybe 12x6x1.5 or so. I really do think that it might be good to use a heatsink the size Nelson is suggesting and not go 'all out' unless you want to wait 'all night' to hear it perform at its best. And I am serious about that preheater. Why not? Glad I had the opportunity to hear it and gives me some build ideas. Thanks TeaBag for dropping by and showing off the F5.
Will be organizing audio get togethers from my new home in Houston this summer. See you there
Uriah
We had a good listen to a bunch of Steve Hoffman mastered stuff. Its easy to become less a fan of a certain group or style and more a fan of anything Steve has remastered. Its really incredible.
Anyway, we heard TeaBag's F5 through the Wavelength pre and through the Lightspeed. Really great amp. Grabs the Audio Nirvana speakers by the throat and shows them who is boss. Big difference I noticed from the solid state I have been using is that the attack on the drums with the F5 is real tight, not flubby. Cymbals were awesome! Not a one note bass, not a one not cymbal anymore. Also found out how important good speaker wire is. Always thought it was a bunch of snake oil and I am sure a bunch of it is pure oil but Larry had some Kimbers and they replaced my braided silver plated copper. Seems the Kimbers are probably made of the same stuff, but way more strands than my 3 and nice connectors on the ends. Mine is just bare wire. Instant improvement in sound with them in place. I am sure with $30-$50 I could make something that will be close.
Both preamps had strong suits and I really had fun listening to both. Junk CDP but hey we deal with what we have and mine is a cheapo Technics. It has sounded better when a DAC was on it so thats in the future, then maybe a better CDP after the amp transforms itself to an F5
Lots of fun. Wish you guys had been there.
Gotta say that maybe the heatsinks need a preheater or something. It took a good hour to get the best sound out of the F5 and the sinks never got HOT. They got good and warm, but the 10 second test was a no flincher
Oh, and his sinks are not enormous. Maybe 12x6x1.5 or so. I really do think that it might be good to use a heatsink the size Nelson is suggesting and not go 'all out' unless you want to wait 'all night' to hear it perform at its best. And I am serious about that preheater. Why not? Glad I had the opportunity to hear it and gives me some build ideas. Thanks TeaBag for dropping by and showing off the F5.Will be organizing audio get togethers from my new home in Houston this summer. See you there
Uriah