Kilowatt
Two MJ21193/4s are required in series, despite their 250V rating, because the maximum current at a Vce of 150V is 0.8A. With a Vce of 75V the maximum current is just over 3A. For the same voltage and current capability, you would need 30 devices in parallel compared to 16 devices in series/parallel.
You are correct in saying that 16 MJ15003s in series/parallel will keep each device just within the SOA shown on the datasheet. However, bear in mind that datasheet graphs are based on typical devices. In reality, some transistors will be better than the graphs indicate but some will be worse. In engineering, all designers have to allow a factor of safety to cover variations in materials and components. If you design your output stage on the basis of keeping the transistors just inside the SOA you are asking for trouble. A safety margin is required for all amps, but the higher the power output the higher this margin needs to be.
A few further comments on things you have said.
Noise and RFI will be picked by the feed from the pole-mounted transformer to your amp and, without the low-pass filter at the input, will not be attenuated. You mention some inductive filtering. If this is to be capable of handling the 24A load and to have sufficient inductance to be effective, it will be large and heavy. You might just as well use a transformer.
With regard to your transformerless power supply you say 'I have no good evidence to suggest that that idea would leave me with a dead amp'. We are not particulary concerned about dead amps, what we are concerned about is the potential for dead people.
You say that your case will be non-conductive, but what about metalwork that may be exposed such as heatsinks.
You seem to be under the impression that, in order to achieve a high power output, all that needs to be done is to scale-up a lower output design. This is not the case. For example, a 200Wrms amp reqires a closed-loop gain of
40 for an input sensitivity of 1Vrms. Your 1800Wrms amp will require a closed-loop gain of 105 for the same input sensitivity. In order to provide sufficient feedback to reduce cross-over distortion etc to an acceptable level, the open-loop gain of your amp will need to be considerably higher than that for the 200Wrms amp. Higher open-loop gain leads to far greater difficulty in ensuring stability and, as others have pointed out, an oscillating amp can quickly destroy itself.
You say that you have read quite a bit about electronics and that you know all the basic principles, but have you really understood what you have read and think you know?
Geoff
Two MJ21193/4s are required in series, despite their 250V rating, because the maximum current at a Vce of 150V is 0.8A. With a Vce of 75V the maximum current is just over 3A. For the same voltage and current capability, you would need 30 devices in parallel compared to 16 devices in series/parallel.
You are correct in saying that 16 MJ15003s in series/parallel will keep each device just within the SOA shown on the datasheet. However, bear in mind that datasheet graphs are based on typical devices. In reality, some transistors will be better than the graphs indicate but some will be worse. In engineering, all designers have to allow a factor of safety to cover variations in materials and components. If you design your output stage on the basis of keeping the transistors just inside the SOA you are asking for trouble. A safety margin is required for all amps, but the higher the power output the higher this margin needs to be.
A few further comments on things you have said.
Noise and RFI will be picked by the feed from the pole-mounted transformer to your amp and, without the low-pass filter at the input, will not be attenuated. You mention some inductive filtering. If this is to be capable of handling the 24A load and to have sufficient inductance to be effective, it will be large and heavy. You might just as well use a transformer.
With regard to your transformerless power supply you say 'I have no good evidence to suggest that that idea would leave me with a dead amp'. We are not particulary concerned about dead amps, what we are concerned about is the potential for dead people.
You say that your case will be non-conductive, but what about metalwork that may be exposed such as heatsinks.
You seem to be under the impression that, in order to achieve a high power output, all that needs to be done is to scale-up a lower output design. This is not the case. For example, a 200Wrms amp reqires a closed-loop gain of
40 for an input sensitivity of 1Vrms. Your 1800Wrms amp will require a closed-loop gain of 105 for the same input sensitivity. In order to provide sufficient feedback to reduce cross-over distortion etc to an acceptable level, the open-loop gain of your amp will need to be considerably higher than that for the 200Wrms amp. Higher open-loop gain leads to far greater difficulty in ensuring stability and, as others have pointed out, an oscillating amp can quickly destroy itself.
You say that you have read quite a bit about electronics and that you know all the basic principles, but have you really understood what you have read and think you know?
Geoff
Geoff said:
Hi Geoff,
You touched on a major issue in that first part there. Another aspect of this is slew rate. To get the same distortion performance at high power as a low power amp requires the high power amp have a faster slew rate, all other things being equal. This makes the amp that much harder to constuct and more likely to oscillate (ie self destruct or roast tweeters).
Kilowatt,
I may be wrong but I get the impression that a big part of the motivation for your design idea is that you want to get high power on a shoestring budget. This is not impossible, but beating the current state of the art in amp design is not an easy task and you are not very likely to stumble across any world shaking innovations on your first project. I find that most of my "revolutionary" ideas end up not being practical because I overlooked something stupid (like my idea only working if I get to ignore the 2nd law of thermodynamics). As a totally off topic side comment I have never liked that law and hearby request that I be exempted from it ;^)
I'm not trying to be nasty but some of your comments sound like you really don't understand what we are talking about here. I suggest you start slower and try your hand at building some smaller amps using proven designs before striking out into totally uncharted territory. There is nothing like watching $100 worth of output transistors go poof or melting the end off a screwdriver to make you slow down and be more cautious (been there, done that, got the box of toasted parts to show for it).
There have been several messages cautioning you about the safety risks with what you are proposing. There are aspects to your design ideas that would result in liability lawsuits and recalls (along with probable criminal prosecution) if a manufacturer used them. Don't count on a plastic case protecting you from electrocution either. The problem is that you could very easily end up with 240 VAC on your signal ground which means that everthing plugged into your amp (tuner, preamp, CD player, turntable etc) will be electrified also. This is a very real problem with old guitar amps and they kill a few musicians every year. What you are proposing creating is a mankiller and you (or someone you love) is likely to be it's first victim.
Phil Ouellette
Such an interesting topic. If you'll pardon an
audio geezer reminiscing:
Quite a few years ago I built a pair of AC line
driven power amps and took them to the CES show.
They had no power transformers, the AC voltage was
rectified directly into 200 volt capacitors, for
about +-160 volts. As I recall, they did about
5 KW into 8 ohms using bridged output stages.
The output stages consisted of MJ15022 and 15024
in series and parallel for a total of 80 devices
arranged in 8 banks of 10. The topology was the
Stasis Amp, and was similar to the S1000, except
for higher voltage front end devices and the series
outputs.
It worked quite well, but at the show everyone was
afraid to touch the volume control.
If I had it to do over, I would look at Mosfets,
such as IRFP360's, rated at 400 volts, and arranged
in a quasi-complementary topology. This would get
you past the second breakdown phenomenon of Bipolars
and would be fairly easy to drive. Rated at 250 watts
each, I would still consider 80 of them in 4 banks of
20.
Certainly using the AC line directly is a scary
proposition, as you have 160 volt rails referenced
to ground, and this could be a lethal voltage, but
let me point out that just about every amplifier is
ground referenced, and creating these voltages with
an AC line transformer would be just as lethal.
All told, I don't recommend this as a project for
anyone who hasn't already built a lot of amplifiers.
audio geezer reminiscing:
Quite a few years ago I built a pair of AC line
driven power amps and took them to the CES show.
They had no power transformers, the AC voltage was
rectified directly into 200 volt capacitors, for
about +-160 volts. As I recall, they did about
5 KW into 8 ohms using bridged output stages.
The output stages consisted of MJ15022 and 15024
in series and parallel for a total of 80 devices
arranged in 8 banks of 10. The topology was the
Stasis Amp, and was similar to the S1000, except
for higher voltage front end devices and the series
outputs.
It worked quite well, but at the show everyone was
afraid to touch the volume control.
If I had it to do over, I would look at Mosfets,
such as IRFP360's, rated at 400 volts, and arranged
in a quasi-complementary topology. This would get
you past the second breakdown phenomenon of Bipolars
and would be fairly easy to drive. Rated at 250 watts
each, I would still consider 80 of them in 4 banks of
20.
Certainly using the AC line directly is a scary
proposition, as you have 160 volt rails referenced
to ground, and this could be a lethal voltage, but
let me point out that just about every amplifier is
ground referenced, and creating these voltages with
an AC line transformer would be just as lethal.
All told, I don't recommend this as a project for
anyone who hasn't already built a lot of amplifiers.
Fresh-T, I shouldn't need a larger heat sink if it's cooled, should I?
I might have to have an inductuve filter outside my case, plugged into a jack in the amp that is in series with the power supply. It would have to handle about 50A for the whole amp, it would be huge. Could I cut back on noise if I ran it from a portable generator? One hell of a generator though.
As far as gain goes, I know I will need higher gain an the lower power design. I already knew that. Increasing gain is just part of the scaling up. I could use devices with higher Hfe, or add some transistors in darlington with the outputs. About slew rate, that's a good point, but if an amp is very simple, and originally very stable, it seems to me like I would ok upscaling it. Maybe not though, I'm sure you guys know what you are talking about. If the amp was going to start oscillating, would it do so as soon as it has power, or when it tried to produce a certain frequency?
Nelson Pass had a good point about the safety issue you are all so concerned about. 300V is 300V. Period. It can kill wether it's off the line or from a big transformer. If you think I have a dangerous supply here, what about tube amps? They run higher voltage yet, at plenty of current. About making the amp safer though, should I use an isolation transformer at my inputs? That way, my supply voltage couldn't come out the front of the amp.
I might have to have an inductuve filter outside my case, plugged into a jack in the amp that is in series with the power supply. It would have to handle about 50A for the whole amp, it would be huge. Could I cut back on noise if I ran it from a portable generator? One hell of a generator though.
As far as gain goes, I know I will need higher gain an the lower power design. I already knew that. Increasing gain is just part of the scaling up. I could use devices with higher Hfe, or add some transistors in darlington with the outputs. About slew rate, that's a good point, but if an amp is very simple, and originally very stable, it seems to me like I would ok upscaling it. Maybe not though, I'm sure you guys know what you are talking about. If the amp was going to start oscillating, would it do so as soon as it has power, or when it tried to produce a certain frequency?
Nelson Pass had a good point about the safety issue you are all so concerned about. 300V is 300V. Period. It can kill wether it's off the line or from a big transformer. If you think I have a dangerous supply here, what about tube amps? They run higher voltage yet, at plenty of current. About making the amp safer though, should I use an isolation transformer at my inputs? That way, my supply voltage couldn't come out the front of the amp.
Hi Kilowatt,
An isolation transformer would be a very good idea if you are running from the single phase, single ended AC power you find in your residential wall outlets. The TV industry uses no-transformer power supplies (at least they use to, I haven't done much with TV's since the '80s). These could be a bear to work on since you couldn't just hook your scope ground up to the signal ground (there is a huge potential difference from the TV's ground to earth ground) unless you power the TV from an isolation transformer.
When you directly rectify the 117 VAC (USA) mains input you end up with your negative supply being very close to earth ground potential, your signal ground being around 100 V and your positive supply being around 200 V. If you connect the signal ground to earth ground sparks are gonna fly.
The main reason why it's normal to use a transformer inside an amp (beside noise filtering, getting the rail voltages right and boosting the current capability) is to make it possible to create a plus and minus supply voltage with a center tapped ground from a single ended AC input. The signal ground is then connected to earth ground (through the line cord) and this ensures that the signal ground (which gets connected to all sorts of human accessible knobs and jacks) never has a dangerous voltage on it and nobody gets toasted.
I was looking at your earlier posts and noticed that you plan on running this from 230 VAC which if you are in the USA actually works to your advantage. US residential AC power service is 230 center tapped and the 117 VAC we normally connect to is developed between one of the hot leads and a neutral wire (that ends up being tied to earth ground in your circuit breaker panel). If you directly rectify the 230 center tapped line voltage then theoretically this isn't any different than using a center tapped transformer inside your amp. The big limitation is that your plus and minus rail voltages will not be symetrical (at the same voltage level about ground) unless whoever wired your house was very carefull about balancing loads and nothing changes while using your amp. If you allow for a posible mismatch in the supply rail voltages in your amp design then this can be made to work.
I still would not do this because the performance of the amp will be dependent on conditions that are not under your control (a refrigerator kicks in or someone uses a microwave oven and the power feeding your amp gets even more unbalanced). Any imbalance on the two legs of the AC power will result in an increase in current flowing in the ground wire which will cause GFI interrupters to trip. Even if you don't have a GFI on your amplifiers power circuit, the extra ground current means that your signal ground is going to be noisy (expect to hear lots of hum from this amp).
So my final word is: Not impossible, just not something I would want in my house. Decent toroidal transformers are not that expensive. 1500VA toroidal transformers from Plintron are US $113 each and that's pretty affordable for what it gets you.
http://www.plitron.com/
Phil Ouellette
An isolation transformer would be a very good idea if you are running from the single phase, single ended AC power you find in your residential wall outlets. The TV industry uses no-transformer power supplies (at least they use to, I haven't done much with TV's since the '80s). These could be a bear to work on since you couldn't just hook your scope ground up to the signal ground (there is a huge potential difference from the TV's ground to earth ground) unless you power the TV from an isolation transformer.
When you directly rectify the 117 VAC (USA) mains input you end up with your negative supply being very close to earth ground potential, your signal ground being around 100 V and your positive supply being around 200 V. If you connect the signal ground to earth ground sparks are gonna fly.
The main reason why it's normal to use a transformer inside an amp (beside noise filtering, getting the rail voltages right and boosting the current capability) is to make it possible to create a plus and minus supply voltage with a center tapped ground from a single ended AC input. The signal ground is then connected to earth ground (through the line cord) and this ensures that the signal ground (which gets connected to all sorts of human accessible knobs and jacks) never has a dangerous voltage on it and nobody gets toasted.
I was looking at your earlier posts and noticed that you plan on running this from 230 VAC which if you are in the USA actually works to your advantage. US residential AC power service is 230 center tapped and the 117 VAC we normally connect to is developed between one of the hot leads and a neutral wire (that ends up being tied to earth ground in your circuit breaker panel). If you directly rectify the 230 center tapped line voltage then theoretically this isn't any different than using a center tapped transformer inside your amp. The big limitation is that your plus and minus rail voltages will not be symetrical (at the same voltage level about ground) unless whoever wired your house was very carefull about balancing loads and nothing changes while using your amp. If you allow for a posible mismatch in the supply rail voltages in your amp design then this can be made to work.
I still would not do this because the performance of the amp will be dependent on conditions that are not under your control (a refrigerator kicks in or someone uses a microwave oven and the power feeding your amp gets even more unbalanced). Any imbalance on the two legs of the AC power will result in an increase in current flowing in the ground wire which will cause GFI interrupters to trip. Even if you don't have a GFI on your amplifiers power circuit, the extra ground current means that your signal ground is going to be noisy (expect to hear lots of hum from this amp).
So my final word is: Not impossible, just not something I would want in my house. Decent toroidal transformers are not that expensive. 1500VA toroidal transformers from Plintron are US $113 each and that's pretty affordable for what it gets you.
http://www.plitron.com/
Phil Ouellette
Kilowatt
You are in danger of showing your complete lack of knowledge and understanding of power amplifier design and perhaps even basic electronics and electrical theory and practice. To be let loose on a project such as this can only result in disaster.
If you think you can increase the open-loop gain of an amp by just increasing the Hfe of certain devices or by dropping in some additional transistors and still maintain stability into practical loads, then think again. It's not that simple!
Have you considered how you are going to maintain thermal stability and how you are going to control the quiescent current so that it doesn't vary with temperature? What about overload and short-circuit protection? Output DC offset control? AC and DC fusing? These all need redesigning if the output power is increased to the levels you are talking about. Then there's the compensation and slew rate issues to be resolved.
If you need to ask when an amplifier will start to oscillate or if a portable generator output is quieter than the mains then you don't even have the basic knowledge necessary to understand a project of this nature, yet alone to design and construct it safely.
Geoff
You are in danger of showing your complete lack of knowledge and understanding of power amplifier design and perhaps even basic electronics and electrical theory and practice. To be let loose on a project such as this can only result in disaster.
If you think you can increase the open-loop gain of an amp by just increasing the Hfe of certain devices or by dropping in some additional transistors and still maintain stability into practical loads, then think again. It's not that simple!
Have you considered how you are going to maintain thermal stability and how you are going to control the quiescent current so that it doesn't vary with temperature? What about overload and short-circuit protection? Output DC offset control? AC and DC fusing? These all need redesigning if the output power is increased to the levels you are talking about. Then there's the compensation and slew rate issues to be resolved.
If you need to ask when an amplifier will start to oscillate or if a portable generator output is quieter than the mains then you don't even have the basic knowledge necessary to understand a project of this nature, yet alone to design and construct it safely.
Geoff
Haldor, that's just why I want to use 240VAC to supply this amp, because there is a center tap. If it were possible to effectively center tap a standard AC supply with no transformer, I would do that. I will have to worry about imbalance though, because most of the things in my house run from one side of the 240. 🙂
Geoff, I am still learning. There are many things that I know, and many things that I don't. I have built a few amps before, just nothing of this magnetude. Things have gone fine before. I don't happen to know much about oscillation of bad amp ciruits, I haven't needed to before. I can safely assume that a generator would be pretty noisy because of it's close proximity to an engine. If I need more information to do this project, I guess I do, you've said that enough. Some of your posts, and others too, have seemed somewhat condescending, but I assure you I have taken no offence because I know you're probably right. I'm sure you're just trying to be helpful and save me a lot of headache.
Since I have no schematic or PCB layout for this amp, it seems I need to learn some things about audio amp design.
Sorry I come across as an idiot, I'm just trying to get help with my amp.
Geoff, I am still learning. There are many things that I know, and many things that I don't. I have built a few amps before, just nothing of this magnetude. Things have gone fine before. I don't happen to know much about oscillation of bad amp ciruits, I haven't needed to before. I can safely assume that a generator would be pretty noisy because of it's close proximity to an engine. If I need more information to do this project, I guess I do, you've said that enough. Some of your posts, and others too, have seemed somewhat condescending, but I assure you I have taken no offence because I know you're probably right. I'm sure you're just trying to be helpful and save me a lot of headache.
Since I have no schematic or PCB layout for this amp, it seems I need to learn some things about audio amp design.
Sorry I come across as an idiot, I'm just trying to get help with my amp.
One thing no one really said for sure: can I or can I not make the 1800W units with a design much like the one here ( http://sound.westhost.com/project03.htm ), for example, but using different devices and supply? This is assuming optimal layout and use of minor passive components. It would have pretty much the same schematic.
Also, since no one thinks there's much hope that I can build the amp, why wasn't that mentioned earlier in the thread?
[Edited by Kilowatt on 11-20-2001 at 09:46 PM]
Also, since no one thinks there's much hope that I can build the amp, why wasn't that mentioned earlier in the thread?
[Edited by Kilowatt on 11-20-2001 at 09:46 PM]
Since I am, by default, the resident tube critter here, I find that I need to point out yet another incorrect statement...this time about tube circuits.
Tube circuits do not have a lot of current on the rails. Yes, the rail voltage is high, but typical bias for a 6550 (the most common output tube) is on the order of 50mA. That's .05A. My tube amps (130W or so, output) use four 6550s per channel for a total of about 200mA. That's .2A total. Even if you combine both channels together (they're monoblocks, but let's not quibble), the total current is only about .5A. Try that on a solid state amp and see how far it gets you.
A buddy of mine is fond of saying that once you've got it in your head, you're not happy 'till you get it in the rump. I think we've got a prime example, here.
Grey
Tube circuits do not have a lot of current on the rails. Yes, the rail voltage is high, but typical bias for a 6550 (the most common output tube) is on the order of 50mA. That's .05A. My tube amps (130W or so, output) use four 6550s per channel for a total of about 200mA. That's .2A total. Even if you combine both channels together (they're monoblocks, but let's not quibble), the total current is only about .5A. Try that on a solid state amp and see how far it gets you.
A buddy of mine is fond of saying that once you've got it in your head, you're not happy 'till you get it in the rump. I think we've got a prime example, here.
Grey
Turn your back and this thread's exploded! Congratulations Kilowatt, you've got us all hooked.
Kilowatt, you're right, it's not impossible to use the ESP amp and beef it up. It would be impractical for all the reasons others have listed above and a few they we haven't thought of yet. Better to start afresh. If you want 1200hp you could start with a 4 cylinder engine and force fuel and air down it's throat, much better to start with a V8, it'll be cheaper in the long run.
You obviously do have some electronics knowledge, but recognise that others on this forum with more knowledge and experience would not take the path you are considering.
I've said it before multiple cheap 100-200W amps (bridge them if you like). The way to go mate! You've got multiple drivers. Start with as may kits as you can afford now and build up the numbers over time. Try beefing one up, if you smoke it, you won't have all your eggs in one basket.
I suspect that if your successful I'll hear it from here!
Regards WALKER
Kilowatt, you're right, it's not impossible to use the ESP amp and beef it up. It would be impractical for all the reasons others have listed above and a few they we haven't thought of yet. Better to start afresh. If you want 1200hp you could start with a 4 cylinder engine and force fuel and air down it's throat, much better to start with a V8, it'll be cheaper in the long run.
You obviously do have some electronics knowledge, but recognise that others on this forum with more knowledge and experience would not take the path you are considering.
I've said it before multiple cheap 100-200W amps (bridge them if you like). The way to go mate! You've got multiple drivers. Start with as may kits as you can afford now and build up the numbers over time. Try beefing one up, if you smoke it, you won't have all your eggs in one basket.
I suspect that if your successful I'll hear it from here!
Regards WALKER
I would work on your enclosure design more. It's very inportant in low freq's, and can gain you alot of output.
Kilowatt
Nobody is calling you an idiot, least of all me. We have all had to learn at sometime, usually from our mistakes. It is one thing to learn from a mistake that fries a couple transistors or blows a fuse or two. Our concern is that a simple mistake on the project you are proposing could have far more serious (and expensive) consequences.
I'm afraid I have to disagree with Walker. In my opinion the power output ESP circuit cannot be pushed to the level you require without so much redesign that the finished circuit would bear very little, if any, resemblence to the original. This circuit is only one of many traditional 3-stage amp designs that have been published on the web and elsewhere. If you want to gain an understanding of how this topology works try Doug Self's 'Audio Power Amplifier Design Handbook'.
I do agree with Walker in his analogy with car engines. If you can find a schematic for, say, a 500W into 4ohm design (there are one or two on the web) then you stand more chance of being able to stretch it to something near to your requirements.
If your aim is for a particular SPL at the listening position, as opposed to being the proud owner of the largest amp in town, then I also agree with Walker that the modular approach is the way to go, for all the reasons he has outlined. Maximise the efficiency of the speakers to limit the power required to achiveve your desired SPL and then split it into managable chunks. You will need a similar number output devices and the heatsinking and power supply requirements will be similar whether you go for 2x1800W or 18x200W.
I am surprised that you say "Also, since no one thinks there's much hope that I can build the amp, why wasn't that mentioned earlier in the thread?". My impression is that this is what we have been trying to say, in various different ways, all along.
Geoff
[Edited by Geoff on 11-21-2001 at 05:32 AM]
Nobody is calling you an idiot, least of all me. We have all had to learn at sometime, usually from our mistakes. It is one thing to learn from a mistake that fries a couple transistors or blows a fuse or two. Our concern is that a simple mistake on the project you are proposing could have far more serious (and expensive) consequences.
I'm afraid I have to disagree with Walker. In my opinion the power output ESP circuit cannot be pushed to the level you require without so much redesign that the finished circuit would bear very little, if any, resemblence to the original. This circuit is only one of many traditional 3-stage amp designs that have been published on the web and elsewhere. If you want to gain an understanding of how this topology works try Doug Self's 'Audio Power Amplifier Design Handbook'.
I do agree with Walker in his analogy with car engines. If you can find a schematic for, say, a 500W into 4ohm design (there are one or two on the web) then you stand more chance of being able to stretch it to something near to your requirements.
If your aim is for a particular SPL at the listening position, as opposed to being the proud owner of the largest amp in town, then I also agree with Walker that the modular approach is the way to go, for all the reasons he has outlined. Maximise the efficiency of the speakers to limit the power required to achiveve your desired SPL and then split it into managable chunks. You will need a similar number output devices and the heatsinking and power supply requirements will be similar whether you go for 2x1800W or 18x200W.
I am surprised that you say "Also, since no one thinks there's much hope that I can build the amp, why wasn't that mentioned earlier in the thread?". My impression is that this is what we have been trying to say, in various different ways, all along.
Geoff
[Edited by Geoff on 11-21-2001 at 05:32 AM]
Geoff, I do agree with you about the ESP amp, (no, this isn't a mutual admiration society), it would look nothing like the original if it were stretched. In fact it wouldn't be the same circuit any more, better to start with a muscle amp. Unfortunately even a 500W 4ohm amp may be outside Kilowatt's design constraints, (cost and device selection).
Regards WALKER
Regards WALKER
There are probably some schematics on the web that I could get what I want from. I've looked at many of them. Of course I'd want to know if it worked before putting power to something expensive but questionable. What I probably need is a simple design that has a PCB layout with it, or something that is already designed. Maybe there are some high impedance circuits that have high enough supply voltage that I can put in parallel.
Anyway, here's why I want to build just one high power amp instead of many smaller ones. I guess there are more advantages using lots of little amps, but it's the few things that the big beast has over the large system of smaller amps that are inportant for my purposes. A big amp has the ability to drive a single big driver per channel. That's not exactly what I plan to do now, but some day I may want to. You can't just parallel up 20 little amps (to get my 7200W, I'd need at least 36 cheap little 200W amps, and that's not cheap when you add that many together) to a 12 ohm load and expect to get much. It's kind of like that even with my current plan for speakers though. The most power/quality(maybe)/price that I could find was the 1kW Audiobahn ALUM12X speakers. They are dual 6 ohm. Even set up for three ohms each, I would need some big amps to drive them anyway. Another thing, a single amp, with two channels, is much quicker and easier to transport and set up than a combined system with countless components. With a system like the one I am proposing, the whole system-everything, can originate from one outlet, rare as that type of outlet is (hey, could be worse, I could make it plug into three phase 🙂 ), rather than having to plug into multiple standard outlets or be on a 100A breaker.
[Edited by Kilowatt on 11-21-2001 at 04:42 PM]
Anyway, here's why I want to build just one high power amp instead of many smaller ones. I guess there are more advantages using lots of little amps, but it's the few things that the big beast has over the large system of smaller amps that are inportant for my purposes. A big amp has the ability to drive a single big driver per channel. That's not exactly what I plan to do now, but some day I may want to. You can't just parallel up 20 little amps (to get my 7200W, I'd need at least 36 cheap little 200W amps, and that's not cheap when you add that many together) to a 12 ohm load and expect to get much. It's kind of like that even with my current plan for speakers though. The most power/quality(maybe)/price that I could find was the 1kW Audiobahn ALUM12X speakers. They are dual 6 ohm. Even set up for three ohms each, I would need some big amps to drive them anyway. Another thing, a single amp, with two channels, is much quicker and easier to transport and set up than a combined system with countless components. With a system like the one I am proposing, the whole system-everything, can originate from one outlet, rare as that type of outlet is (hey, could be worse, I could make it plug into three phase 🙂 ), rather than having to plug into multiple standard outlets or be on a 100A breaker.
[Edited by Kilowatt on 11-21-2001 at 04:42 PM]
Kilowatt,
Please, start your designs with a simple one. All of these guys here have much more experience in audio design and, as fas as I know, I agree with them about your project. Non isolated power supplies and very high power demmand a lot of care and experience to achieve good results and it could be dangerous too.
regards,
Please, start your designs with a simple one. All of these guys here have much more experience in audio design and, as fas as I know, I agree with them about your project. Non isolated power supplies and very high power demmand a lot of care and experience to achieve good results and it could be dangerous too.
regards,
Her is a circuit that will suit u well:
http://users.ece.gatech.edu/~mleach/superamp/
Schematics & PCB layout.
Why don't u construct 8 900Watt@12ohm amps from this schem with your +/- 150volt, for your 8 subs...
You could make them im one chassis if it needs to be portable. Problem then with one (chassis) powercord is that the ciruitbreaker needs to be over 30A (for 240volt main).
http://users.ece.gatech.edu/~mleach/superamp/
Schematics & PCB layout.
Why don't u construct 8 900Watt@12ohm amps from this schem with your +/- 150volt, for your 8 subs...
You could make them im one chassis if it needs to be portable. Problem then with one (chassis) powercord is that the ciruitbreaker needs to be over 30A (for 240volt main).
Oh, I wouldn't worry about that, I'm plugging it into a 50A mains.
And thanks a lot for the link! It's a little complicated, but I like it very much anyway. 🙂
[Edited by Kilowatt on 11-22-2001 at 12:27 PM]
And thanks a lot for the link! It's a little complicated, but I like it very much anyway. 🙂
[Edited by Kilowatt on 11-22-2001 at 12:27 PM]
High power project...something for you?
Hello, this link is maybe interesting for you:
http://homepages.picknowl.com.au/glenk/default.htm
This is all about high power amplification, and maybe he can give you some answers too.
Good luck.
(guess it's not the first one you saw..;-)
Hello, this link is maybe interesting for you:
http://homepages.picknowl.com.au/glenk/default.htm
This is all about high power amplification, and maybe he can give you some answers too.
Good luck.
(guess it's not the first one you saw..;-)
Perhaps an output transformer?
Hello KiloWatt,
If you are planning for a mains powered amp, just do it like the old days, use an output transformer (RCF did it as well some time ago)
The output transformer will prevent a direct contact with the mains..and you can use some form of feedback too.
And you can adjust the output impedance as well.
[Edited by HaaiFaai on 11-22-2001 at 12:43 PM]
Hello KiloWatt,
If you are planning for a mains powered amp, just do it like the old days, use an output transformer (RCF did it as well some time ago)
The output transformer will prevent a direct contact with the mains..and you can use some form of feedback too.
And you can adjust the output impedance as well.
[Edited by HaaiFaai on 11-22-2001 at 12:43 PM]
- Status
- Not open for further replies.