yeti,
Output transformers would work, but it kinda takes the fun out of it. One of the advantages that solid state has over tubes (we'll ignore OTL designs for the moment) is the lack of an output transformer. It seems rather a pity to stick in a transformer if it's not absolutely necessary. McIntosh used to--and I think still do--incorporate autoformers in their amps. It was quite common in the older solid state pieces, but fell by the wayside as direct-coupled topologies came into vogue.
Through an odd chain of circumstance, I now have some short lengths of Al that would be ideal for small ribbon drivers. Since I already have on hand some nice, hefty magnets, I'm feeling this irrational desire to whip up a couple of 6" ribbon tweeters. This would fit right in with the circuit discussion above.
Grey
Output transformers would work, but it kinda takes the fun out of it. One of the advantages that solid state has over tubes (we'll ignore OTL designs for the moment) is the lack of an output transformer. It seems rather a pity to stick in a transformer if it's not absolutely necessary. McIntosh used to--and I think still do--incorporate autoformers in their amps. It was quite common in the older solid state pieces, but fell by the wayside as direct-coupled topologies came into vogue.
Through an odd chain of circumstance, I now have some short lengths of Al that would be ideal for small ribbon drivers. Since I already have on hand some nice, hefty magnets, I'm feeling this irrational desire to whip up a couple of 6" ribbon tweeters. This would fit right in with the circuit discussion above.
Grey
Two things:
First, I was just looking at some 55W 6V halogen bulbs for pretty cheap. About .65 ohms, I figure. Bias resistor/heat sink/room lighting rolled into one little glass nub--food for thought.
Second, and so painfully obvious that I wonder why it never hit me before: Why not use battery power?!! I can't think of an easier design to do it on. Cuts cost by allowing a cheap (noisy) power supply into the battery, right?
Bill
First, I was just looking at some 55W 6V halogen bulbs for pretty cheap. About .65 ohms, I figure. Bias resistor/heat sink/room lighting rolled into one little glass nub--food for thought.
Second, and so painfully obvious that I wonder why it never hit me before: Why not use battery power?!! I can't think of an easier design to do it on. Cuts cost by allowing a cheap (noisy) power supply into the battery, right?
Bill
Bill,
Don't forget that you'll be buring current like there's no tomorrow--the batteries will need to be kept charged. Other than that, I'd say that it's an interesting possibility.
I've been thinking along these lines, and as soon as I get the Aleph 2.60 thread a little further along, I'll turn my attention more this way, as I'm interested in spite of myself.
Grey
Don't forget that you'll be buring current like there's no tomorrow--the batteries will need to be kept charged. Other than that, I'd say that it's an interesting possibility.
I've been thinking along these lines, and as soon as I get the Aleph 2.60 thread a little further along, I'll turn my attention more this way, as I'm interested in spite of myself.
Grey
Let me qualify that battery thing:
What I was thinking of was using a cheap, noisy 6V (or 12V) power supply constantly charging a tractor or car battery in series. Wouldn't the battery serve as a fine noise filter? I could be way off base here, but I was thinking that the sluggish electrolytic energy transport of the battery would even out all but the lowest frequency ripple. If you want to add even more filtering, a 0.5F 20V Cap won't break the bank.
Can anyone out there shed some light on my battery-as-a-filter theory? Maybe it's already being done?
Bill
P.S. I see that I just earned my DIYaudio acolyte accolade! Celebrate with me!!
What I was thinking of was using a cheap, noisy 6V (or 12V) power supply constantly charging a tractor or car battery in series. Wouldn't the battery serve as a fine noise filter? I could be way off base here, but I was thinking that the sluggish electrolytic energy transport of the battery would even out all but the lowest frequency ripple. If you want to add even more filtering, a 0.5F 20V Cap won't break the bank.
Can anyone out there shed some light on my battery-as-a-filter theory? Maybe it's already being done?
Bill
P.S. I see that I just earned my DIYaudio acolyte accolade! Celebrate with me!!
Any takers on my battery question?
I see 250W computer power supplies with 5V 30A taps for sale on the net at less than $10 each. Ah, the classic love story: supply meets demand. These'll do the trick, I think.
Bill
I see 250W computer power supplies with 5V 30A taps for sale on the net at less than $10 each. Ah, the classic love story: supply meets demand. These'll do the trick, I think.
Bill
Yeah, in a car's electrical system, the battery is the filter. But there's one problem with using battery power: where's the center tap/ground?
That's a good point. Actually, I was thinking of using two matched batteries, one in series on the plus side, one on the neg. I guess that would be a virtual ground? Actual grounding would have to be a third wire.
Bill,
I was flipping mental coins as to whether to post this here or in the Line Array thread, and decided to go for this one. This is an idea I'd intended to take a poke at in the SOZ w/current sources, etc. thread, but I'll toss it at you to stimulate your thinking:
Look at the schematic for a SOZ. There's a 1 ohm resistor connecting the two sources. See it?
Replace that with a driver.
Note that the driver will have to be purely resistive (ribbons and planar drivers are good candidates, here) because any variation in impedance (i.e. with frequency) between the two sides of the amp will change the gain, which is a no-no. I've yet to test this, but I see no reason why it wouldn't work pretty well.
Since the SOZ will now be functioning as a pair of followers, you'll need more gain going in, but, as such a low resistance requires very little in the way of voltage swing to get things moving, it shouldn't be all that difficult a problem.
Yes, you can use less than one ohm there.
Grey
I was flipping mental coins as to whether to post this here or in the Line Array thread, and decided to go for this one. This is an idea I'd intended to take a poke at in the SOZ w/current sources, etc. thread, but I'll toss it at you to stimulate your thinking:
Look at the schematic for a SOZ. There's a 1 ohm resistor connecting the two sources. See it?
Replace that with a driver.
Note that the driver will have to be purely resistive (ribbons and planar drivers are good candidates, here) because any variation in impedance (i.e. with frequency) between the two sides of the amp will change the gain, which is a no-no. I've yet to test this, but I see no reason why it wouldn't work pretty well.
Since the SOZ will now be functioning as a pair of followers, you'll need more gain going in, but, as such a low resistance requires very little in the way of voltage swing to get things moving, it shouldn't be all that difficult a problem.
Yes, you can use less than one ohm there.
Grey
Very intererstink...
I'm looking up the SOZ schematics now.
But don't some/most planars have a lumpy impedence curve? How bout a zobeled dynamic? It's won't be exactly purely resistive...but...
I'm looking up the SOZ schematics now.
But don't some/most planars have a lumpy impedence curve? How bout a zobeled dynamic? It's won't be exactly purely resistive...but...
Bill,
Drivers such as the Magneplanars are purely resistive. There are still a lot of the old Infinity EMIMs and EMITs floating around used. Speaking of used, there's always the Stratherns. Bohlender-Graebeners (new). And various ribbons, of course. The assumption here is that you use them <i>sans</i> passive crossovers, transformers, etc. In other words, we're talking about a bi/tri/whatever amped system where the amp sees the resistance of the driver, only, without other reactive components getting in the way.
Nearly all of these are bewteen .1 and 5 ohms...in other words, they're ideal for what I'm proposing.
Grey
P.S.: Drat your hide! I need to get working on the Aleph-X, and you've got me thinking about the SOZ thing again...grumble, grumble, grumble...
Drivers such as the Magneplanars are purely resistive. There are still a lot of the old Infinity EMIMs and EMITs floating around used. Speaking of used, there's always the Stratherns. Bohlender-Graebeners (new). And various ribbons, of course. The assumption here is that you use them <i>sans</i> passive crossovers, transformers, etc. In other words, we're talking about a bi/tri/whatever amped system where the amp sees the resistance of the driver, only, without other reactive components getting in the way.
Nearly all of these are bewteen .1 and 5 ohms...in other words, they're ideal for what I'm proposing.
Grey
P.S.: Drat your hide! I need to get working on the Aleph-X, and you've got me thinking about the SOZ thing again...grumble, grumble, grumble...
- Status
- Not open for further replies.