I have 2 headphones: Stax Lambda Pro circa 185 (electrostatic), and Grado SR80. The Lambda Pros are incredible, but I haven't compared them to recent high end headphones. Obviously I will not attempt to drive the electrostatics from an F5.
The Grados are very good for their cost (< $100). Most of the grado headphones have 32 ohm impedance and sensitivity of 98 dB/mW. I measured that my SR80s with a 440 Hz sine wave at a high, but tolerable sound level required .3V RMS = .424V peak. This means that your could probably power the F5 with 6 volts per rail is the output FETs have a low enough threshold voltage as many laterals do.
O
~8vac is more than enough for HD600, I would normally shoot for ~6-7 (about 2x balanced voltage gain on the 3vrms output of ES901X)). perhaps I should have been more clear, i'm talking about output into the Hi Z load, not power for the whole amp, but I didnt crunch any numbers, just baased on what i've used in the past to good effect.
I get great sound into the lowZ headphones in question with an amp (the wire balanced) powered with a dual bipolar supply that is current limited at +/-120ma per channel at +/-6.6vdc (so 1.532Wpc total power available to the amp). i'm aware they need more current than the hd600 and i'm aware the wire is AB, but i'm also well aware of what they actually need in reality to sound great, not theoretically.
With this amp and this power supply I have experience some of, if not THE cleanest, most brutal transient speed and low bass i've ever heard from them or any headphone for that matter (these are $1200 WW-MM-TT 3 way in ear custom fit musicians monitors) and cannot get close to half way with the amp set at unity gain with ~3vrms balanced output from a sabre. the only sound ive ever heard that rivals this for speed with this bass authority, are ATC active studio monitors. stax headphones match/possibly beat them for speed, but the lowest registers are missing.
so I think i'm pretty safe with my estimates. its actually worse than you think, because of the crosover they can present less than 10 ohms at some frequencies. but they are VERY efficient, what they need most is low output impedance and in my experience as long as the power into my hd600 is fine and the output impedance of the amp and noise is nice and low, then the requirements of the jh13 have a good chance to fall into place and if they dont adding higher current isnt going to fix it. they are a hard task and for sure not every amp is up to the job, but DC coupled amps with low impedance are the best candidates and I think the F5 is probably the likeliest of the pass amps to suit?
they would be more of a novelty as the wire was built for them and does an excellent job, building the amp to cope with 30Wpc to cater for them is out of the question. but I would build the amp for the hd600 and for testing other headphones at meets
I think thats enough words devoted to headphones though really
I get great sound into the lowZ headphones in question with an amp (the wire balanced) powered with a dual bipolar supply that is current limited at +/-120ma per channel at +/-6.6vdc (so 1.532Wpc total power available to the amp). i'm aware they need more current than the hd600 and i'm aware the wire is AB, but i'm also well aware of what they actually need in reality to sound great, not theoretically.
With this amp and this power supply I have experience some of, if not THE cleanest, most brutal transient speed and low bass i've ever heard from them or any headphone for that matter (these are $1200 WW-MM-TT 3 way in ear custom fit musicians monitors) and cannot get close to half way with the amp set at unity gain with ~3vrms balanced output from a sabre. the only sound ive ever heard that rivals this for speed with this bass authority, are ATC active studio monitors. stax headphones match/possibly beat them for speed, but the lowest registers are missing.
so I think i'm pretty safe with my estimates. its actually worse than you think, because of the crosover they can present less than 10 ohms at some frequencies. but they are VERY efficient, what they need most is low output impedance and in my experience as long as the power into my hd600 is fine and the output impedance of the amp and noise is nice and low, then the requirements of the jh13 have a good chance to fall into place and if they dont adding higher current isnt going to fix it. they are a hard task and for sure not every amp is up to the job, but DC coupled amps with low impedance are the best candidates and I think the F5 is probably the likeliest of the pass amps to suit?
they would be more of a novelty as the wire was built for them and does an excellent job, building the amp to cope with 30Wpc to cater for them is out of the question. but I would build the amp for the hd600 and for testing other headphones at meets
I think thats enough words devoted to headphones though really
I have a few questions about using fans...especially after your design inspired me to look into quiet fans. Don't mean to steal your thread though, your excellent work just gets me thinking.
I have a bunch of commercial amps that use fans. These are not common or even uncommon computer fans, instead they use what appears to be a large synchronous motor with a large laminated frame and large motor rotor but smallish blade. No shaped cowling at all, round blade spinning in a square box looks like a horrible design. In fact, it looks very much like the synchronous AC motor used in kids' cheap rubber-wheel rim-drive record player turntables. No brushes. Some kind of odd (to me) copper looking material encircles the laminated-core frame at one spot. The amplifiers have a SMPS with a large laminated core transformer and PWM regulator with discrete components (I think). Anyway, there's a simple (few components) fan control circuit which I don't really understand. The fans usually run at slow speed, unless you short the FET output.
http://mysite.verizon.net/gregorypinkowski/PR1800-schematic.pdf
These stock fans run pretty slow, but pretty noisy compared to some nearly silent PC fans I've lived with. Anyway, sorry to pollute an F5 thread with such an idea, but I want to use these amps for my stereo and A/V HT. So I have some questions and some observations:
1) Looking at my schematic, does the fan speed control vary drive frequency (it sure looks like a synchronous AC motor) or voltage or am I all wrong about the motor, and is it driven with pulsed DC or something? Fans have 2 wires, so there's no tach feedback like some PC fans. But there are more and more PC fans runnning on pulsed DC these days... Any chance of my substituting quieter fans? Would I need a different fan driver circuit?
2) Two fans on each amp seems like a poor choice. I guess it's for reliability, as I'm sure one on 'high' could handle the load and would certainly be noticed; though the airflow pattern would be all wrong. But I'm thinknig that two fans must make a lot more noise than one if their individual sounds make additional beats together.
3) I'm inspired by the IBM z/OS mainframes and some other tricks I've seen in big computers. All kinds of somewhat noisy stuff inside, but the shapes of the cabinet front and back labyrinths create balanced opposed flows that cancel (and absorptive lining material helps no doubt). Fan noise is mostly high-frequency so it should be easy to muffle, cancel, and absorb. On the other hand, a bunch of little fans seems like a poor choice if the amps are racked in cases with muffled intakes and exhausts that might reduce the air pressure at the intake and increase the air pressure at the exhaust, possibly also exaggerating the poor performance of any weakest fan (and possibly creating really bad flow patterns if a fan fails and the hot air fro the others starts turning it backward).
4) Your design makes much more sense, with the fans on the INSDIE of the box, already muffled somewhat by the tunnel. My fans are literally sticking out the back of the amp cabinets, each in its own little metal box, like an afterthought.
5) Fewer bigger fans are much quieter than lots of small ones. My big squirrel-cage furnace air handler is silent when running slow. A cabinet of HP servers with dozens of 2" fans screaming has permanently damaged my hearing.
I guess what I should do is come up with a new fan-control circuit, take the fan boxes and intake grilles off the amps, put the amps in racks with a large squirrel-cage fan INSIDE the rack, and carefully design some kind of intake and exhaust mufflers? Or take off the fan boxes and mount PC fans? I'd love to hear any comments.
I have a bunch of commercial amps that use fans. These are not common or even uncommon computer fans, instead they use what appears to be a large synchronous motor with a large laminated frame and large motor rotor but smallish blade. No shaped cowling at all, round blade spinning in a square box looks like a horrible design. In fact, it looks very much like the synchronous AC motor used in kids' cheap rubber-wheel rim-drive record player turntables. No brushes. Some kind of odd (to me) copper looking material encircles the laminated-core frame at one spot. The amplifiers have a SMPS with a large laminated core transformer and PWM regulator with discrete components (I think). Anyway, there's a simple (few components) fan control circuit which I don't really understand. The fans usually run at slow speed, unless you short the FET output.
http://mysite.verizon.net/gregorypinkowski/PR1800-schematic.pdf
These stock fans run pretty slow, but pretty noisy compared to some nearly silent PC fans I've lived with. Anyway, sorry to pollute an F5 thread with such an idea, but I want to use these amps for my stereo and A/V HT. So I have some questions and some observations:
1) Looking at my schematic, does the fan speed control vary drive frequency (it sure looks like a synchronous AC motor) or voltage or am I all wrong about the motor, and is it driven with pulsed DC or something? Fans have 2 wires, so there's no tach feedback like some PC fans. But there are more and more PC fans runnning on pulsed DC these days... Any chance of my substituting quieter fans? Would I need a different fan driver circuit?
2) Two fans on each amp seems like a poor choice. I guess it's for reliability, as I'm sure one on 'high' could handle the load and would certainly be noticed; though the airflow pattern would be all wrong. But I'm thinknig that two fans must make a lot more noise than one if their individual sounds make additional beats together.
3) I'm inspired by the IBM z/OS mainframes and some other tricks I've seen in big computers. All kinds of somewhat noisy stuff inside, but the shapes of the cabinet front and back labyrinths create balanced opposed flows that cancel (and absorptive lining material helps no doubt). Fan noise is mostly high-frequency so it should be easy to muffle, cancel, and absorb. On the other hand, a bunch of little fans seems like a poor choice if the amps are racked in cases with muffled intakes and exhausts that might reduce the air pressure at the intake and increase the air pressure at the exhaust, possibly also exaggerating the poor performance of any weakest fan (and possibly creating really bad flow patterns if a fan fails and the hot air fro the others starts turning it backward).
4) Your design makes much more sense, with the fans on the INSDIE of the box, already muffled somewhat by the tunnel. My fans are literally sticking out the back of the amp cabinets, each in its own little metal box, like an afterthought.
5) Fewer bigger fans are much quieter than lots of small ones. My big squirrel-cage furnace air handler is silent when running slow. A cabinet of HP servers with dozens of 2" fans screaming has permanently damaged my hearing.
I guess what I should do is come up with a new fan-control circuit, take the fan boxes and intake grilles off the amps, put the amps in racks with a large squirrel-cage fan INSIDE the rack, and carefully design some kind of intake and exhaust mufflers? Or take off the fan boxes and mount PC fans? I'd love to hear any comments.
Can someone look at the schematic and tell me what the stock fans are driven with?
http://mysite.verizon.net/gregorypinkowski/PR1800-schematic.pdf
Despite the big motors, my current fan blades in unshrouded square boxes appear to be the noisiest and least efficient design. They are, however, incredibly reliable; whereas I've seen PC fans just eventually gum up in Los Angeles air depite any claims of "seled bearings"... If the fans don't like so much resistance to flow, I might do well to mount two "input" fans at the front and two "exhaust" fans at the rear; both panels will acutally accomodate that without too much difficulty. But the prospect of paying for 36 fans might be a consideraiton.
http://mysite.verizon.net/gregorypinkowski/PR1800-schematic.pdf
Despite the big motors, my current fan blades in unshrouded square boxes appear to be the noisiest and least efficient design. They are, however, incredibly reliable; whereas I've seen PC fans just eventually gum up in Los Angeles air depite any claims of "seled bearings"... If the fans don't like so much resistance to flow, I might do well to mount two "input" fans at the front and two "exhaust" fans at the rear; both panels will acutally accomodate that without too much difficulty. But the prospect of paying for 36 fans might be a consideraiton.
It looks like the fans run on AC line voltage controlled by a silicon controlled rectifier.
If you want a small Class A amp that is quiet, there is only one solution IMHO -- water cooling.
You should still have some heat sink, to give the amp a chance to safely shut down in case the water cooling fails.
Assuming of course cost is no issue and compactness is everything.
There are enough PC based cooling towers and pumps on the market.
Just put them in the balcony, and run the tubes to the amp.
http://zalman.co.kr/ENG/product/Product_Read.asp?idx=183
And yes, done it before. Works like a charm.
Patrick
You should still have some heat sink, to give the amp a chance to safely shut down in case the water cooling fails.
Assuming of course cost is no issue and compactness is everything.
There are enough PC based cooling towers and pumps on the market.
Just put them in the balcony, and run the tubes to the amp.
http://zalman.co.kr/ENG/product/Product_Read.asp?idx=183
And yes, done it before. Works like a charm.
Patrick
Last edited:
I do know that water-cooled motorcycle drivelines weigh less than air-cooled; plus they are more compact with better mass centralization and perform better for similar displacement. So you're probably still right, you probably can package a smaller and lighter water and fan cooled amp than an air-cooled one.
If you want a small Class A amp that is quiet, there is only one solution IMHO -- water cooling.
You should still have some heat sink, to give the amp a chance to safely shut down in case the water cooling fails.
Assuming of course cost is no issue and compactness is everything.
There are enough PC based cooling towers and pumps on the market.
Just put them in the balcony, and run the tubes to the amp.
::: Zalman, leading the world of Quiet Computing Solutions :::
And yes, done it before. Works like a charm.
Patrick
What do you consider quiet? The Noctua fan I use has a 10 dB(A) noise level at the speed I run it. I assume that you are using the amplifier to generate music at some level. Do you think that you will hear this fan? It is far quieter that other ambient noises like refrigerators, traffic, ...
Thanks! That makes some sense, as the motors look like AC motors. It also means I can probably measure the AC, and transform/rectify/filter it to run much quieter DC fans.
Ihquam, Your idea for cooling the amp is totally awesome ! Could something like this be used to switch the fan on or off as needed ? See link > Digi-Key - 317-1502-ND (Manufacturer - CS703515Z)
I haven't tried any fan control circuits yet. My first approach would be a 2 speed control. Run the fan at low (or minimum) speed below 50C and use the thermostat to turn it on at full speed at 50C. A 2nd thermostat could be used for an intermediate speed at 40C.
I have had 4 of the very same 7" heat-sinks your using for almost 2yrs now. I could not find a machinist willing to cut them down for less than $100.00 so they sat collecting dust.... I would very much like to construct a amplifier using your lay-out the only difference is I will be using the Peter Daniels F5 pcb's. If made available I would be very interested in your pcb's. Thanks for posting this really great design ! 
I had my heatsinks cut for $5 by a local metal materials supplier. I that is needed is a suitably large metal bandsaw.
I have had 4 of the very same 7" heat-sinks your using for almost 2yrs now. I could not find a machinist willing to cut them down for less than $100.00 so they sat collecting dust.... I would very much like to construct a amplifier using your lay-out the only difference is I will be using the Peter Daniels F5 pcb's. If made available I would be very interested in your pcb's. Thanks for posting this really great design !
Heatsink-usa has a 6.080 low fin extrusion that I had thought to get in a 13.5" length and mount it to the top of the 7.0". For this I'm gonna need a more precise cut. Using the low fin HS allows me to construct a chassis using 2) removable sides attached to it for easy access. Plus I think it would look cool in black ! Take a look let me know what you think. As far as cutting the HS I plan to start with a Jig-saw, then use a cross-slide type vise mounted in a drill press with a milling bit to square things up.
Last edited:
Ihquam, If I had offended you from my last post I'm truly sorry ! I really believe your layout is amazing, I plan to construct my F5 using your Heat-Tunnel design. If any of the F5 amps were to be called a F5T it should be yours because of the tunnel design... Also it is air assisted much like a turbo only not driven by exhaust gases. Best Wishes, Thor61 
I was in no way offended. I do not understand your inability to find someone with the proper saw to cut the heatsink at a reasonable price. Do you have access to a power cutoff saw or a table saw? Blades for those kinds of saws to cut aluminium are not to expensive.
Ihquam, If I had offended you from my last post I'm truly sorry ! I really believe your layout is amazing, I plan to construct my F5 using your Heat-Tunnel design. If any of the F5 amps were to be called a F5T it should be yours because of the tunnel design... Also it is air assisted much like a turbo only not driven by exhaust gases. Best Wishes, Thor61
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.