Hello folks. I finally decided to start saving money for a pair of Son of Zen monoblocks. In the meantime, I'd like to clear my ideas about the building details, and most important the thermal issues.
First of all, let me specifiy that the amps will be driven single-ended (I've heard that efficiency is further reduced that way... But I don't have balanced stuff at all).
Well, let's start. I'd like to build *at least* two 15W monoblocks... But I noticed that, if we exclude heatsinks (which will be the subject of this thread), for only 50 EUR more I can buy the stuff for a nice 25W version.
The problem is that I have no experience in what profiled heatsinks can handle.
I'd like to build each monoblock following the Aleph model, because I find it amazingly beautiful and it seems like the easiest way:
I can find heatsinks similar to those, in measure of about 30x30cm (12x12in.); I could use three of them for each amp, as front - left - right sides of the cabinet, such as in the Aleph, with the following components on them:
- 15W version:
- front heatsink: 35W (MOS1) + 35W (MOS2) + 9W (R7) = ~80W
- left heatsink: 75W (R1) + 75W (R3) + 9W (R5) = ~160W
- right heatsink: 75W (R2) + 75W (R4) + 9W (R6) = ~160W
- 25W version:
- front heatsink: 45W (MOS1) + 45W (MOS2) + 15W (R7) = ~105W
- left heatsink: 120W (R1) + 120W (R3) + 15W (R5) = ~255W
- right heatsink: 120W (R2) + 120W (R4) + 15W (R6) = ~225W
If I'm right in interpreting mr Pass' graphs, these would be the power ratings to dissipate on the heatsinks.
What do you guys think about my idea? Would 12x12inches panels be good to handle it, especially for the more ambitious 25W version?
Thank you very much
Giacomo
First of all, let me specifiy that the amps will be driven single-ended (I've heard that efficiency is further reduced that way... But I don't have balanced stuff at all).
Well, let's start. I'd like to build *at least* two 15W monoblocks... But I noticed that, if we exclude heatsinks (which will be the subject of this thread), for only 50 EUR more I can buy the stuff for a nice 25W version.
The problem is that I have no experience in what profiled heatsinks can handle.
I'd like to build each monoblock following the Aleph model, because I find it amazingly beautiful and it seems like the easiest way:
An externally hosted image should be here but it was not working when we last tested it.
I can find heatsinks similar to those, in measure of about 30x30cm (12x12in.); I could use three of them for each amp, as front - left - right sides of the cabinet, such as in the Aleph, with the following components on them:
- 15W version:
- front heatsink: 35W (MOS1) + 35W (MOS2) + 9W (R7) = ~80W
- left heatsink: 75W (R1) + 75W (R3) + 9W (R5) = ~160W
- right heatsink: 75W (R2) + 75W (R4) + 9W (R6) = ~160W
- 25W version:
- front heatsink: 45W (MOS1) + 45W (MOS2) + 15W (R7) = ~105W
- left heatsink: 120W (R1) + 120W (R3) + 15W (R5) = ~255W
- right heatsink: 120W (R2) + 120W (R4) + 15W (R6) = ~225W
If I'm right in interpreting mr Pass' graphs, these would be the power ratings to dissipate on the heatsinks.
What do you guys think about my idea? Would 12x12inches panels be good to handle it, especially for the more ambitious 25W version?
Thank you very much
Giacomo
Assuming that I'm following your proposal correctly, I think you'll have trouble moving the heat from the heatsink the output device is mounted on to the one 90 degrees from it. In other words, you'll need to spread the heat around. Offhand, the only solution I can think of is to use something like 1/4" plates of metal (aka heat spreaders) bolted to the inside of the heatsinks, but even then you'll be facing a butt joint at the corner and the heat won't move across that joint very efficiently.
You can reduce the heat generated by the SOZ considerably by using a current source instead of biasing resistors. It will also improve the performance and make your single ended input nearly as good as a balanced input. I did a thread about this once upon a time called something off the wall like "SOZ With Current Sources." (I like calling things by really mysterious and obscure names so no one knows what the thread's about...)
Grey
You can reduce the heat generated by the SOZ considerably by using a current source instead of biasing resistors. It will also improve the performance and make your single ended input nearly as good as a balanced input. I did a thread about this once upon a time called something off the wall like "SOZ With Current Sources." (I like calling things by really mysterious and obscure names so no one knows what the thread's about...)
Grey
Thank you Grey. Even if this is my first post, I'm an "old reader" of this forum, and I know many of you by nick I know the thread you're talking about too, I'll read it again and figure out the "new" circuit. (If I recall correctly, I could be in trouble finding the final scheme, so expect me to be back here and ask for help )
I understood what you mean about heat transferring, but I was hoping that each one of the three faces could handle its own heat. I try to explain better: I put some RRR on side 1, some MOS on side 2, and some RRR on side 3... I would expect that each side dissipates its own heat. So side 1 takes care of its three resistors, side two keeps its MOSFETs cool and healthy, and side 3 dissipates from the other three resistors. What's wrong with this "each for its own" way to arrange dissipation?
If you consider that the resistors sides are way hotter (255W vs 105W) than the MOSFET side, isn't it better if we keep all them separated?
Thank you again....
Giacomo
I know the thread you're talking about too, I'll read it again and figure out the "new" circuit. (If I recall correctly, I could be in trouble finding the final scheme, so expect me to be back here and ask for help )I understood what you mean about heat transferring, but I was hoping that each one of the three faces could handle its own heat. I try to explain better: I put some RRR on side 1, some MOS on side 2, and some RRR on side 3... I would expect that each side dissipates its own heat. So side 1 takes care of its three resistors, side two keeps its MOSFETs cool and healthy, and side 3 dissipates from the other three resistors. What's wrong with this "each for its own" way to arrange dissipation?
If you consider that the resistors sides are way hotter (255W vs 105W) than the MOSFET side, isn't it better if we keep all them separated?
Thank you again....
Giacomo
I'm studying the situation... (thank you all)
Well, I still don't have the specifications for any heatsink, because I only have some "potential sources" from which I could buy them... Still need to investigate. It's just an idea, for now.
I've read with A LOT of interest the Zen Variations which add SuSy to the Son of Zen, and yes, thanks woody, in particular ZV 7. I don't have the right instrumentation to calibrate the gorgeous transformer-loaded version, but SuSy with active current sources seems a really, really nice and "doable" project.
I'm referring to this one:
http://img103.imageshack.us/img103/8030/zv7rhu2.gif
Let's talk a moment about coupling with a preamp: I only have (and it's beautifully firm in my setup) an Audible Illusions Modulus 3A preamp... A S.E. all tube preamp with an awful 1.2 kohm output impedance. Then I'm a bit scared by the low input impedance of ZV7-R amp.
I think that I would need something like 12kohm, better if 50 or so, in place of the 1.5k Mr. Pass wrote in the article, to make it a good match for my Audible Illusions preamp. One thing which could, maybe, come in handy, is the following: my preamp has a huge 24dB (or something around that value) gain... So, if there's a solution to lower the ZV7-R gain and raise the input impedance, it can be VERY interesting for my particular case.
As usual, thank you so much for your help.
Regards,
Giacomo
Well, I still don't have the specifications for any heatsink, because I only have some "potential sources" from which I could buy them... Still need to investigate. It's just an idea, for now.
I've read with A LOT of interest the Zen Variations which add SuSy to the Son of Zen, and yes, thanks woody, in particular ZV 7. I don't have the right instrumentation to calibrate the gorgeous transformer-loaded version, but SuSy with active current sources seems a really, really nice and "doable" project.
I'm referring to this one:
http://img103.imageshack.us/img103/8030/zv7rhu2.gif
Let's talk a moment about coupling with a preamp: I only have (and it's beautifully firm in my setup) an Audible Illusions Modulus 3A preamp... A S.E. all tube preamp with an awful 1.2 kohm output impedance. Then I'm a bit scared by the low input impedance of ZV7-R amp.
I think that I would need something like 12kohm, better if 50 or so, in place of the 1.5k Mr. Pass wrote in the article, to make it a good match for my Audible Illusions preamp. One thing which could, maybe, come in handy, is the following: my preamp has a huge 24dB (or something around that value) gain... So, if there's a solution to lower the ZV7-R gain and raise the input impedance, it can be VERY interesting for my particular case.
As usual, thank you so much for your help.
Regards,
Giacomo
Oh my God... Freeze!
I've just had something like a flash... I discovered the First Watt F4 amp (Nelson Pass strikes back... He's better than the Empire).
Maybe I found something which could demolish all the current problems... My preamp has such a gain (up to 30dB, with internal switches turned to "HI") which I always considered an issue, and now I discovered this amp, capable to make it a strong value... Simply using the huge gain of the preamp, the F4 could perfectly match it to my 98.5dB/W/m speakers (Royal Device Laura mkII), with the simplest DC/NoFeedback circuit around...
Isn't THIS what I was looking for, mates?
Giacomo
I've just had something like a flash... I discovered the First Watt F4 amp (Nelson Pass strikes back... He's better than the Empire).
Maybe I found something which could demolish all the current problems... My preamp has such a gain (up to 30dB, with internal switches turned to "HI") which I always considered an issue, and now I discovered this amp, capable to make it a strong value... Simply using the huge gain of the preamp, the F4 could perfectly match it to my 98.5dB/W/m speakers (Royal Device Laura mkII), with the simplest DC/NoFeedback circuit around...
Isn't THIS what I was looking for, mates?
Giacomo
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.