As an experiment this entire amplifier could be run off of a couple of car batteries. It would be cheap and easy to implement. Then the remainder of the design could focus on the discrete components needed for this much current.
With the exceptional db/watt ratio of a long ribbon speaker the actual voltage needed for MUSIC levels should not be to terribly stressing on the output section. It is not uncommon for long pure ribbon designs to achieve 100-105 db/watt. Once these numbers are plugged in a 10 watt amp should be more than sufficient.
I have long wondered what the life expectancy is of a pure aluminum membrane without any backing support. Aluminum is not to reliable with repeated 180 degree stressing day in and day out. Using a thin Spectre plastic membrane which has silver, copper or aluminum deposited on it in a micron thin layer would be superior for longevity and extended low frequency response.
Tad
With the exceptional db/watt ratio of a long ribbon speaker the actual voltage needed for MUSIC levels should not be to terribly stressing on the output section. It is not uncommon for long pure ribbon designs to achieve 100-105 db/watt. Once these numbers are plugged in a 10 watt amp should be more than sufficient.
I have long wondered what the life expectancy is of a pure aluminum membrane without any backing support. Aluminum is not to reliable with repeated 180 degree stressing day in and day out. Using a thin Spectre plastic membrane which has silver, copper or aluminum deposited on it in a micron thin layer would be superior for longevity and extended low frequency response.
Tad
I believe many may suggest pure foil superior to a foil/plastic composite.
If an accurate fatigue model is established, perhaps the membranes could be replaced at intervals over the transducers lifetime to sustain optimal operation. I believe this is commonly done in the Aerospace industry. Aluminum isn't too expensive and it wouldn't be too difficult to make the system modular.
Fair,
You need to also take into account the increased level over a standard speaker vs linesource at listening distances of 4 M the line source will be much better than the numbers suggest vs pointsource .
I have them now being driven by amplifiers rated to be (Adcom )850 @ 2 ohms and the sound will grow and maintain good volume . for safety I'm using 10 amp speaker out fuses , which do pop on peaks with say classical or live large scale recordings , never on regular compressed studio stuff like R&B , pop ,etc ..
currently being driven by a ( best this way so far ) Threshold S500 and Adcoms 565 on the bass , Would love to do a class a and my decision is to build one with 25 watts @ 8 for 200 watts/ch@1 ohm stable ..
don't think i need 5000 watts , i would try it if you have ...
I have found that mylar sounds better than straight foil in the middle ...
Last edited:
Indeed, this I haven't taken into account.
Then I would calculate a circlotron power stage to get the output power of the biggest Aleph from Nelson Pass, the Aleph 1.2.
Stereophile: Pass Aleph 1.2 monoblock power amplifier
CLASS-A - DIY PASS LABS ALEPH 1.2 amplifier schematic & PCB
But I guess, the Aleph 1.2 itself could be also a good choice (distortion character more like a tube amplifier). Compare to the previous power stage proposal of me a very small power amplifier with very small requirements regarded the heatsinks - additional easy to clone and modify.
So I hope, you don't prefer high sound pressure levels like live events in your music room.
Last edited:
LOL....
No i like my hearing and 200 -250@ 1 ohm will be more than enuff , well 200 hundred solid watts not whatts ..
As to live music, yes ! I do want it so sound like live music and it does that pretty well now , it is just at that certain point on dynamic peaks , that the whole thing fall apart and i end up blowing fuses ..
a little more will get me over my hump and yes i plan to re-vamp the speakers a few more db's are available and is there , the speaker lost it's excellent balance when it was more efficient , so it will take some workand time on my part to get both ...
regards,
The power required for this ribbon is calculated assuming an extremely low resistive element. Aluminum is very efficient as an electrical conductor and heat transfer medium, however, as the driven element in a ribbon design would not something more robust like titanium are even steel suffice as the sound transfer source. The benefits would be that these materials are not as efficient for electrical conductors and could raise the low ohm threshold to a more acceptable level. You could also use a much thinner material because of the added strength.
Let me know if I am wrong here but I do not know of many, or any, musical instruments, voice membranes that are made of aluminum. In contrast the vast majority are plain old high tensile steel. We should be able to at least double if not triple the ohm figure which would make this project more realistic.
One other area to discuss is the best voltage for the discrete devices in the amplifier. I think mosfets would be almost mandatory for the output and they like 20+ volts, I read this somewhere in the Pass literature, in order to operate properly. That much voltage is going to produce prodigous amounts of heat.
Tad
Let me know if I am wrong here but I do not know of many, or any, musical instruments, voice membranes that are made of aluminum. In contrast the vast majority are plain old high tensile steel. We should be able to at least double if not triple the ohm figure which would make this project more realistic.
One other area to discuss is the best voltage for the discrete devices in the amplifier. I think mosfets would be almost mandatory for the output and they like 20+ volts, I read this somewhere in the Pass literature, in order to operate properly. That much voltage is going to produce prodigous amounts of heat.
Tad
OK,
We did the group thing over the last couple days and would have to agree , 5000 watts would seem to be the way to go
Currently running A pr of mono bloc adcoms on the bass ( 585@4 ohm ) and an Threshold S500 -II on the 2 way direct drive Ribbon panels ..
Best combination to date with this much swing , the music is very dynamic clear , detailed, with very good tonality and a lot of swing and energy ., important if trying to replicate live sound/music .
OK, so now i can live with the s500 sound but with the necessary mods to live at 1 ohm . I don't need 250 @ 8 , 50- 100 @ 8 doubling down to 1 would be perfect and of course with the necessary PS and heat sinking ( the S500 is very hot to the touch after 1 hr )
I do not want to mod the S500 , but would prefer to do a DIY that would get me to my goals , any recommendation as to which of the Pass designs currently being used here would do so with similar or better sonic character to the S500 .
It seems that all the current pass desigs being used here is class A only and that would not be practical for a 1 ohm load . Could there be a compromise and have a sliding class a bias ? Load is low , but very resistive with good phase angle .
Any recommendations ?.....
We did the group thing over the last couple days and would have to agree , 5000 watts would seem to be the way to go
Currently running A pr of mono bloc adcoms on the bass ( 585@4 ohm ) and an Threshold S500 -II on the 2 way direct drive Ribbon panels ..
Best combination to date with this much swing , the music is very dynamic clear , detailed, with very good tonality and a lot of swing and energy ., important if trying to replicate live sound/music .
OK, so now i can live with the s500 sound but with the necessary mods to live at 1 ohm . I don't need 250 @ 8 , 50- 100 @ 8 doubling down to 1 would be perfect and of course with the necessary PS and heat sinking ( the S500 is very hot to the touch after 1 hr )
I do not want to mod the S500 , but would prefer to do a DIY that would get me to my goals , any recommendation as to which of the Pass designs currently being used here would do so with similar or better sonic character to the S500 .
It seems that all the current pass desigs being used here is class A only and that would not be practical for a 1 ohm load . Could there be a compromise and have a sliding class a bias ? Load is low , but very resistive with good phase angle .
Any recommendations ?.....
Last edited:
S.Wayne,
As someone stated earlier the Pass/Thagard A-75 with the 12 outputs per rail could be a good candidate. It moves to class AB when initial bias setting has played out.
With the original design Nelson states in the article that it should handle 75 plus amps.
You can regulate the frontend from a separate supply and eliminate heavy rail sag interferring with the frontend.
Tad
As someone stated earlier the Pass/Thagard A-75 with the 12 outputs per rail could be a good candidate. It moves to class AB when initial bias setting has played out.
With the original design Nelson states in the article that it should handle 75 plus amps.
You can regulate the frontend from a separate supply and eliminate heavy rail sag interferring with the frontend.
Tad
Commercial Amp in Current Drive Mode wanted for Full Range Ribbons - Overview wanted
I am looking arround for an overview of various models in that mode (= Transconductance-Amplifier resp. Power-OTA).
It is intended to drive without transformer the low-impedance loudspeaker Apogee model "Scintilla"(version with low impedance).
Which amps - except that one by
http://www.ultrasound-hifi.com/Us_wh_1/FR/Amplificatori/amplificatori_en.html -
are here available in new and used condition on world's market ?
http://www.audiocircle.com/index.php?topic=86222
I am looking arround for an overview of various models in that mode (= Transconductance-Amplifier resp. Power-OTA).
It is intended to drive without transformer the low-impedance loudspeaker Apogee model "Scintilla"(version with low impedance).
Which amps - except that one by
http://www.ultrasound-hifi.com/Us_wh_1/FR/Amplificatori/amplificatori_en.html -
are here available in new and used condition on world's market ?
This is the URL:
http://www.audiocircle.com/index.php?topic=86222
Attachments
Last edited:
I forgot to mention, that for the user, who asked me for a list of suitable amp devices, it is not important, to hear very loud while his listening sessions. Thus no high power is necessary. Most important for him is a lovely and musical sonic quality (similar like the character of good single ended amp devices like Aleph or ZEN).
BTW - the only models I know with good sonic quality connected to this ribbon speaker from so called voltage drive mode output stages are the Krell "MRA" (vintage in the meantime) and the "SA Reference" from Plinius Audio (from currently availabe devices the best in this days - so I think).
StereoTimes - Event - A Visit to Krell by Steve Ekblad
Plinius SA-Reference power amplifier | Stereophile.com
Plinius SA-REF review from the experts at whathifi.com
May I suggest using a class-d amp.
A good design running on a lower than normal voltage, would be able to put out a lot of current and thereby watt for this use.
If the circuit employs an over current circuit, it will probably need to be adjusted for the higher current, but that should be doable.
As a DIY project it would open for even more options.
Today you can get FETs which can easily switch 50 - 100 A in a TO220 package, at switching frequencies over 250 kHz .... should be enough to get the good old Apogees to shine
Kind regards Baldin
A good design running on a lower than normal voltage, would be able to put out a lot of current and thereby watt for this use.
If the circuit employs an over current circuit, it will probably need to be adjusted for the higher current, but that should be doable.
As a DIY project it would open for even more options.
Today you can get FETs which can easily switch 50 - 100 A in a TO220 package, at switching frequencies over 250 kHz .... should be enough to get the good old Apogees to shine
Kind regards Baldin
I haven't heard most of the various class-D topologies. But I have heard the Tripath TA3020 similar like aboutMay I suggest using a class-d amp.
A good design running on a lower than normal voltage, would be able to put out a lot of current and thereby watt for this use.
If the circuit employs an over current circuit, it will probably need to be adjusted for the higher current, but that should be doable.
As a DIY project it would open for even more options.
Today you can get FETs which can easily switch 50 - 100 A in a TO220 package, at switching frequencies over 250 kHz .... should be enough to get the good old Apogees to shine
Kind regards Baldin
600W Class-D Audioverstärker - t0mmy.org
so as the ICEPOWER (B&O/Sanyo)
http://www.diyaudio.com/forums/clas...ic-about-rev-engin-not-complete-finished.html
and various 2 years old UcD models from
UcD
at normal 4-8 ohm two/three-way loudspeakers
UcD is the best (very good for class-D).
Compare to analog amp topologies (even Class AB) I would say, Class-D is still not preferable.
Maybe on ribbon speakers my perception is another. And maybe in the meantime Class-D is better in all respects than any analog audio amp, no matter, whether PP or single ended.
As long as I myself have not experienced this, I'm not so sure.
several months ago I start this thread:
http://www.diyaudio.com/forums/clas...hout-custom-made-ics-like-tripath-ta2020.html
Perhaps there appear some interesting class-D devices, that are new to me.
Last edited:
I do think class-d is the future, but all designs are not equal.
I for one had a very big LC Audio The End running for many years, but I swaped it with a small class-d amp (home build) simply because it sounded better
Anyway, what I was stating was class-d amps inherited ability to put out a lot of amps at low voltages, makin them ideal for driving e.g. the Scintellas mid directly. So e.g. try a Hypex UcD400 at half the voltage specified ( say +-40VDC), and try it out
As said you will have to do something about the over current protection to make it go all the way. But I'm sure Hypex can help pointing out which few resistors needs to be changed to increase the max current.
Kind regards Baldin
I for one had a very big LC Audio The End running for many years, but I swaped it with a small class-d amp (home build) simply because it sounded better
Anyway, what I was stating was class-d amps inherited ability to put out a lot of amps at low voltages, makin them ideal for driving e.g. the Scintellas mid directly. So e.g. try a Hypex UcD400 at half the voltage specified ( say +-40VDC), and try it out
As said you will have to do something about the over current protection to make it go all the way. But I'm sure Hypex can help pointing out which few resistors needs to be changed to increase the max current.
Kind regards Baldin
really ?
I thought the heat from driving exstreme low impedance would kill it fast ?
or trigger the 'kill relay' ?
even at low voltage, it will still try to deliver high wattage
unless ofcourse you use a 'limiter', of some kind
Switching losses rises with voltage over the mosfets. Therefore lower the voltage, and you are able to drive more amps, and therefore higher power into lower resistance.
Same actually goes for class-ab amps, but there you will have to put more transistors in the output stage to handle the increase in current. With new mosfets, you can stay with only 2 and still drive high power, with low loss.
See slide 9: D vs. AB
http://www.irf.com/product-info/audio/classdtutorial2.pdf
or this design http://www.irf.com/technical-info/refdesigns/iraudamp9.pdf giving more than 1kW into 2 ohms using 4 TO220 fets. (not saying that this design is what you should use for the Apogees, but just to show what I mean with driving low impedance)
Remember that you don't burn off half the power in the amp with class-d, here you should be able to get some 80-90 percent efficiency => small heatsinks and half size psu
Kind regards Baldin
Same actually goes for class-ab amps, but there you will have to put more transistors in the output stage to handle the increase in current. With new mosfets, you can stay with only 2 and still drive high power, with low loss.
See slide 9: D vs. AB
http://www.irf.com/product-info/audio/classdtutorial2.pdf
or this design http://www.irf.com/technical-info/refdesigns/iraudamp9.pdf giving more than 1kW into 2 ohms using 4 TO220 fets. (not saying that this design is what you should use for the Apogees, but just to show what I mean with driving low impedance)
Remember that you don't burn off half the power in the amp with class-d, here you should be able to get some 80-90 percent efficiency => small heatsinks and half size psu
Kind regards Baldin