I have a pair of Oris 150 horns and am looking at building SE amplifiers for them. I have a pair of large dual 15 inch woofer cabinets (and drivers of course) that will be used with a two channel Crown XLS 1002 amplifier that has a built in 24db slope crossover with level controls. The horns will sit on the bass cabinets.
The amp I'm looking to build for the horns is a 46 single ended amplifier but since I don't need to produce any bass below 150Hz or so I was wondering if it would be a good idea to use the coupling capacitor between the driver tube and the 46 to start rolling off the bass around that range. I'm also wondering if it'd be good to even have a cap in front of the driver tube for the same purpose. The idea is to prevent the amp and the drivers in the horn from having to reproduce the low frequencies below the cutoff point.
The crown amp is going to be used balanced from the preamp but it also has a balanced pass through output. I'm looking to use that pass through connection to connect the SE amps via Jensen input transformers (JT-11P4-1) on the inputs of the 46 amps.
So I guess what I'm asking is what capacitor value range should I use? And if I use one before each tube, I'm assuming that would give me a 12db roll off but would I use the same values for both stages? Is doing both a good idea or should I just use the one coupling cap between stages for a 6db roll off?
Any suggestions? Thoughts?
The amp I'm looking to build for the horns is a 46 single ended amplifier but since I don't need to produce any bass below 150Hz or so I was wondering if it would be a good idea to use the coupling capacitor between the driver tube and the 46 to start rolling off the bass around that range. I'm also wondering if it'd be good to even have a cap in front of the driver tube for the same purpose. The idea is to prevent the amp and the drivers in the horn from having to reproduce the low frequencies below the cutoff point.
The crown amp is going to be used balanced from the preamp but it also has a balanced pass through output. I'm looking to use that pass through connection to connect the SE amps via Jensen input transformers (JT-11P4-1) on the inputs of the 46 amps.
So I guess what I'm asking is what capacitor value range should I use? And if I use one before each tube, I'm assuming that would give me a 12db roll off but would I use the same values for both stages? Is doing both a good idea or should I just use the one coupling cap between stages for a 6db roll off?
Any suggestions? Thoughts?
These look to have a 77cm mouth diameter, and probably a 150Hz flare rate, so the driver will be pretty much unloaded down at 150 Hz, where the largest acoustic power demands of music lie. Any way you can reduce the excursions that the driver must make will minimize distortion.
You don't mention your intended crossover point, so let's call it 350Hz. Two single pole rolloffs at the same frequency are equal to a two pole rolloff of Q=0.5. Putting these about an octave below crossover is fairly conservative; you could come up nearer to crossover frequency. Blending (summing the acoustic outputs from the disparate drivers) is tricky with horns because their own response near mouth cutoff and flare rate cutoff are both drastic. If you stay far enough away from cutoffs, you could even use a 24dB/oct active filter and ignore driver/horn responses, but it's hard to get that far away.
To choose capacitors, you can calculate the value of each by setting their capacitive reactance equal to the sum of source resistance plus load resistance at your chosen frequency. You can find calculators on the web, but the capacitor's value is reciprocal of (2 times pi times frequency times desired capacitive reactance) where capacitive reactance is equal to the total of resistance against which it's working. Forgive me if this is obvious to you, but it's hard to guess folks' tech level on a board.
All good fortune,
Chris
You don't mention your intended crossover point, so let's call it 350Hz. Two single pole rolloffs at the same frequency are equal to a two pole rolloff of Q=0.5. Putting these about an octave below crossover is fairly conservative; you could come up nearer to crossover frequency. Blending (summing the acoustic outputs from the disparate drivers) is tricky with horns because their own response near mouth cutoff and flare rate cutoff are both drastic. If you stay far enough away from cutoffs, you could even use a 24dB/oct active filter and ignore driver/horn responses, but it's hard to get that far away.
To choose capacitors, you can calculate the value of each by setting their capacitive reactance equal to the sum of source resistance plus load resistance at your chosen frequency. You can find calculators on the web, but the capacitor's value is reciprocal of (2 times pi times frequency times desired capacitive reactance) where capacitive reactance is equal to the total of resistance against which it's working. Forgive me if this is obvious to you, but it's hard to guess folks' tech level on a board.
All good fortune,
Chris
Thanks for the info!
While the woofer amp has a built in digital 24db crossover I'm just looking to roll off the lower bass in the horn tube amps. Many people run the driver in the Oris horn full range and let the roll off at horn cutoff do the job but I thought I could put a pair of first order line level high pass filters, one at the input and one between stages doing exactly what you said, preventing the driver from reproducing (as much) low frequency content I won't hear anyway.
I was looking at crossing over at around 150 to 200Hz but if that's too low I can probably go a bit higher than that. The Oris does have a small bass reflex enclosure behind the driver if that makes any difference so the driver isn't flopping in free air like an open baffle.
Don't worry about being patronizing as I've just started looking into this and am not that knowledgeable. I was thinking of using an active crossover but figured I could get a significant improvement with two cascaded single order filters built into the horn amps, I just don't know how to properly implement it.
Thanks again.
While the woofer amp has a built in digital 24db crossover I'm just looking to roll off the lower bass in the horn tube amps. Many people run the driver in the Oris horn full range and let the roll off at horn cutoff do the job but I thought I could put a pair of first order line level high pass filters, one at the input and one between stages doing exactly what you said, preventing the driver from reproducing (as much) low frequency content I won't hear anyway.
I was looking at crossing over at around 150 to 200Hz but if that's too low I can probably go a bit higher than that. The Oris does have a small bass reflex enclosure behind the driver if that makes any difference so the driver isn't flopping in free air like an open baffle.
Don't worry about being patronizing as I've just started looking into this and am not that knowledgeable. I was thinking of using an active crossover but figured I could get a significant improvement with two cascaded single order filters built into the horn amps, I just don't know how to properly implement it.
Thanks again.
Classic horn designs place the driver's fundamental resonance way up into its working range in order to have it damped by the horn's acoustic impedance. For example, a Klipschorn woofer has its fundamental resonance around 80 or 90 Hz, and compression drivers have almost no air chamber at all, very tiny.
Modern designs are tending towards a loosening of these demands, and assuming much lower acoustic outputs (typical modern home use!), so rules of thumb can often be ignored. I do still worry about operation too near horn mouth or flare rate cutoff frequencies, but I'm a worrier.
With horns, there are no Thiele/Small/Linkwitz/D'Appolito models with simple moving pistonic masses, free air loading, etc. Instead, even the horn shapes themselves are based on known incorrect assumptions (flat wave fronts in "exponential" horns as a particularly bad example) and usually a requirement that dimensions be calculable with an equation. End rant mode. (Tractrix is surprising very very close to a true exponential expansion, only deviating near the mouth.)
So integrating these tempermental beasts into a loudspeaker doesn't have an "ideal" solution. Instead, what we want to do is to balance competing desires for lowest excursions (and thus lowest distortions), meaning highest crossover frequencies, against best summing of the varied drivers' outputs for flattest response, and the most "single-source" listener perception, meaning lowest crossover frequencies.
I think your plan to optimize the horn driving amp for its task is perfectly spot-on, and should be done even if you later try an added active crossover. It's certainly the right place to start your exploration, and might well prove to be the complete solution.
All good fortune,
Chris
Modern designs are tending towards a loosening of these demands, and assuming much lower acoustic outputs (typical modern home use!), so rules of thumb can often be ignored. I do still worry about operation too near horn mouth or flare rate cutoff frequencies, but I'm a worrier.
With horns, there are no Thiele/Small/Linkwitz/D'Appolito models with simple moving pistonic masses, free air loading, etc. Instead, even the horn shapes themselves are based on known incorrect assumptions (flat wave fronts in "exponential" horns as a particularly bad example) and usually a requirement that dimensions be calculable with an equation. End rant mode. (Tractrix is surprising very very close to a true exponential expansion, only deviating near the mouth.)
So integrating these tempermental beasts into a loudspeaker doesn't have an "ideal" solution. Instead, what we want to do is to balance competing desires for lowest excursions (and thus lowest distortions), meaning highest crossover frequencies, against best summing of the varied drivers' outputs for flattest response, and the most "single-source" listener perception, meaning lowest crossover frequencies.
I think your plan to optimize the horn driving amp for its task is perfectly spot-on, and should be done even if you later try an added active crossover. It's certainly the right place to start your exploration, and might well prove to be the complete solution.
All good fortune,
Chris
I could check in the BD Design forum but this is more of an amplifier design question than anything else. I currently have everything except the 46 amp in question so that's the next project. I have a couple of SE amps on hand (a Korneff 45, a Decware ZKIT1 and a pair of Amp Camp Amps) but don't want to modify them as they're already built and working fine, etc. I'll probably try them full range as I'm building the 46 amp.
I have thought of using a pair of the Marchand XM46 passive line level crossover boards but I figure it'd be cheaper and easier to just use a pair of C/R high pass networks on the input and in between the stages of the amplifier circuit. It just seems a simple and elegant way to cut down on the IM distortion in the amp and the doppler distortion in the driver.
I also have to get some friends over to help me move the bass cabinets from the garage into place as they're huge. I had knee surgery not too long ago and I need to be careful.
I have thought of using a pair of the Marchand XM46 passive line level crossover boards but I figure it'd be cheaper and easier to just use a pair of C/R high pass networks on the input and in between the stages of the amplifier circuit. It just seems a simple and elegant way to cut down on the IM distortion in the amp and the doppler distortion in the driver.
I also have to get some friends over to help me move the bass cabinets from the garage into place as they're huge. I had knee surgery not too long ago and I need to be careful.
The OP is suggesting a pole at 150Hz or more, so this would be strongly dominant even with feedback around it. Your point about overload recovery times is important, and another advantage to minimizing the size of coupling capacitor from driver to output stage grid (the location of clipping in all optimized conventional valve amplifiers) in *any* design.
All good fortune,
Chris
All good fortune,
Chris
I was thinking primarily about achieving better integration between mid-bass and the horn, to achieve the best overall result.
Many seem to favour driving the LF/lower mid amplifier from the output of the mid/HF amplifier. In other words the mid/HF amplifier runs full range. This may sound counter-intuitive but provides useful example of where someone can benefit from others who have actual experience with these systems.
With the usual input stage feedback return won't the driver correct for the inter stage low frequency loss by boosting bass to the limit of its excess gain?
Since this is a 45 SE then GNB seems unlikely anyway. Even so I'ld be inclined to eliminate the low frequency load on the driver stage power supply by setting the 150 Hz limit at the input coupling cap. The builder can decide if adding another cap in the signal path is a compromise. An added benefit is the inter stage cap can still be used to double the attenuation slope at lower frequencies.
Yes indeed. I think we would not disagree. And although loop feedback is probably not an issue here, if it were, the "internal" high pass filter would be even higher frequency, open loop, to give the desired closed loop high pass frequency.
The late, great DF96, to whom we can give thanks, used to argue for a dominant pole as early as possible to minimize signal excursion throughout the amplifier. I would argue that the dominant pole should be at the clipping point, to minimize overload time. But both together is more better.
All good fortune,
Chris
The late, great DF96, to whom we can give thanks, used to argue for a dominant pole as early as possible to minimize signal excursion throughout the amplifier. I would argue that the dominant pole should be at the clipping point, to minimize overload time. But both together is more better.
All good fortune,
Chris
Jean-Michel LeCleach would recommend using group delay to determine where his exponential horns should be crossed. In addition, polar measurements would show you where this would fit into the system.
Let us pause for a moment of thanks for JMLC, who contributed so much, and who we lost just a few years ago.
May the Goddess bless us all,
Chris
May the Goddess bless us all,
Chris
Thanks to all that have replied, I'm not planning on any negative feedback on the amplifier.
With the Oris horns being around 106db efficient according to Bert Doppenberg with my FE206ES-R drivers I only really need one good watt and the 46 looks like a good candidate, especially if I roll off the low end a bit. I'm debating about what driver to use, I have some cool globe mesh 27 tubes but I'm also considering using 6J5's (or the loctal equivalent). I have a Korneff integrated but that's a lot of amplification stages and I'm going to be using a preamp to drive both the Crown bass amp and the tube amp. I bought myself an Analog Discovery 2 for my birthday and I have a bunch of caps on order as well to do some prototyping. With the audio suite software I'll be able to tinker with cap values and hopefully get a good result.
With the Oris horns being around 106db efficient according to Bert Doppenberg with my FE206ES-R drivers I only really need one good watt and the 46 looks like a good candidate, especially if I roll off the low end a bit. I'm debating about what driver to use, I have some cool globe mesh 27 tubes but I'm also considering using 6J5's (or the loctal equivalent). I have a Korneff integrated but that's a lot of amplification stages and I'm going to be using a preamp to drive both the Crown bass amp and the tube amp. I bought myself an Analog Discovery 2 for my birthday and I have a bunch of caps on order as well to do some prototyping. With the audio suite software I'll be able to tinker with cap values and hopefully get a good result.