Hey folks, I've been stumbling around the forums looking and reading, but mainly drowning in verbiage that is still slightly out of my realm of expertise.
The short of things, is that I am wondering if something like this is possible, and if there are available plans out there that anyone could recommend, or perhaps be willing to contribute your own thoughts and designs.
I am building a digital organ for my church. There will be roughly 32 channels of audio: 28 for 8' and smaller "pipes" (64Hz and higher) and a couple of subs for 16' pipes (down to 32Hz) and a sub-sub or two down to 16Hz.
I am not worried about the subs.
I actually want to discuss using a full-range design for the 64Hz and up "pipes."
One design that immediately grabbed my attention was the Windermere^2 design with the upper driver at a 45 degree angle. I'm not sure why, but that design just looks like it would be well suited for this application. (It even looks similar to many wooden organ pipes!)
http://http://www.creativesound.ca/pdf/Lake-District-maps-080909.pdf
I have built a few different designs that I've found online, including the TubaHT by Bill Fitzmaurice, as well as his TLAH Pros. I've also build the Fostex BK-12's, and made a semi-unsuccessful CSS Trio12 BIB.
Since I am financing this project almost entirely on my own dime, including the console build, electronics, wiring, computer, etc. I am looking to find a speaker design that is economical and somewhat easily produced.
With so many channels, I'm looking for nice fidelity, but I don't really need tons of volume. The idea is for a single speaker to emulate only a pipe or two at any given time - ie: I'm not building a 2-channel stereo for listening to pipe organ music, but instead a 32 channel audio system to reproduce these sounds.
My questions are:
The Windermere^2 driver (CSS EL70) from what I gather is discontinued. I'm assuming it to be futile to try and force a different driver to work, but has anyone had any experience with this?
I've never designed a speaker enclosure of any type (successfully), so can anyone shed some light on what other designs exist that are easily "reverse engineered" so I can figure out how to move forward?
Thanks for reading.
The short of things, is that I am wondering if something like this is possible, and if there are available plans out there that anyone could recommend, or perhaps be willing to contribute your own thoughts and designs.
I am building a digital organ for my church. There will be roughly 32 channels of audio: 28 for 8' and smaller "pipes" (64Hz and higher) and a couple of subs for 16' pipes (down to 32Hz) and a sub-sub or two down to 16Hz.
I am not worried about the subs.
I actually want to discuss using a full-range design for the 64Hz and up "pipes."
One design that immediately grabbed my attention was the Windermere^2 design with the upper driver at a 45 degree angle. I'm not sure why, but that design just looks like it would be well suited for this application. (It even looks similar to many wooden organ pipes!)
http://http://www.creativesound.ca/pdf/Lake-District-maps-080909.pdf
I have built a few different designs that I've found online, including the TubaHT by Bill Fitzmaurice, as well as his TLAH Pros. I've also build the Fostex BK-12's, and made a semi-unsuccessful CSS Trio12 BIB.
Since I am financing this project almost entirely on my own dime, including the console build, electronics, wiring, computer, etc. I am looking to find a speaker design that is economical and somewhat easily produced.
With so many channels, I'm looking for nice fidelity, but I don't really need tons of volume. The idea is for a single speaker to emulate only a pipe or two at any given time - ie: I'm not building a 2-channel stereo for listening to pipe organ music, but instead a 32 channel audio system to reproduce these sounds.
My questions are:
The Windermere^2 driver (CSS EL70) from what I gather is discontinued. I'm assuming it to be futile to try and force a different driver to work, but has anyone had any experience with this?
I've never designed a speaker enclosure of any type (successfully), so can anyone shed some light on what other designs exist that are easily "reverse engineered" so I can figure out how to move forward?
Thanks for reading.
Also - do you want the speakers (and cabinets) to provide any harmonic distortion / tone shaping, or are they basically each reproducing a full-range "pre-processed" signal? What is the desired aesthetic effect? How much harmonic content is in each channel vs. fundamental?
As you can tell I don't know much about digital pipe organs.
As you can tell I don't know much about digital pipe organs.
It turns out that the new Pluvia Seven (CHS70) is very close in terms of loading requirements to the EL70. Scott has already checked the sims and says they will drop in althou there could be a small amount of tweaking for optimum results. I have been thinking of (can't recall whether i have actually asked) working with Scott to do a rework of the entire Lake District set for the new driver.
dave
diyaudio's member xrk971 has modeled: the original Karlson coupler from 1951, his own "XKi" K-coupler and GregB's "Karlsonator" which has been built in 2 inch to 12 inch sizes. (the original plans are for 6.5", 8" and 12" driver sizes) - mini Karlsonators have been built with double drivers as has the XKi approach using inexpensive Dayton PA130 fullrange - there are threads in the fullrange forum covering all these designs.
Fulmer slotted designs would aesthetically fit
XKi with two inexpensive Dayton PA130 5" fullrange
here's a 10" coax in my Karlsonator12
Norman Fulmer speaker patent
http://www.google.com/patents/US2787332
Fulmer slotted designs would aesthetically fit
XKi with two inexpensive Dayton PA130 5" fullrange
An externally hosted image should be here but it was not working when we last tested it.
here's a 10" coax in my Karlsonator12
Norman Fulmer speaker patent
http://www.google.com/patents/US2787332
Last edited:
Let me see if I can address all of the above in one global response and perhaps clarify a few things.
There will be approximately 30 full range drivers, which will have the various sound samples distributed across them. The algorithm inside the software looks at the number of output channels it has to work with and distributes the "load" (ie. number of notes you are playing at an exact moment) across the available channels it has to work with, attempting to avoid any two notes on the same output channel at any given time. (There are additional settings and algorithms to choose from inside the software, but for the sake of this conversation, this is all that seems relevant at this point.)
It is very unlikely to play 32 simultaneous pitches (as you only have 10 fingers and 2 feet (usually)), but it is very likely to play many differing tone colors simultaneously. So, it is possible to further divide the audio channels by tone color. Being able to separate out the samples/output by both pitch (frequency) and timbre (tone color) means having the ability to keep distortion down, and clarity preserved.
Pipe organ frequencies break down like this: (using the A=440 tuning, so there is some degree of flexibility here and frequencies can vary slightly)
32' Pipes = 16Hz fundamental
16' Pipes = 32Hz fundamental
8' Pipes = 64Hz fundamental
ETC.
So, what I'm looking to do is find a design that can handle the 8', 4', 2' and higher stops. Ideally the lowest it should ever need to play is 64Hz. And since the software algorithm is so good it can utilize whatever outputs it has available by cycling pitches away from others, then I just need a cabinet capable of producing a nice, clean, balanced output.
The idea is not to have a small number of speakers trying to handle the entire organ, but rather, as many full range speakers as possible, so that each pipe sample is reproduced accurately.
The sub channels are typically mix-downs of other routing, but I will likely devote a large setup, utilizing a couple of crossovers, in a biamping situation to create another larger full-range channel or two just for those. (That is a separate conversation, though as I'm trying to focus on the smaller pipes for the moment.)
There are other considerations as well, such as the room, which is very large, and the location of the speakers once they are installed. (See attached pictures)
The speakers will live up on top of the large wooden rooms that flank both sides of the altar. There is roughly 12x12' of space on each side for all the speakers to live.
There will be approximately 30 full range drivers, which will have the various sound samples distributed across them. The algorithm inside the software looks at the number of output channels it has to work with and distributes the "load" (ie. number of notes you are playing at an exact moment) across the available channels it has to work with, attempting to avoid any two notes on the same output channel at any given time. (There are additional settings and algorithms to choose from inside the software, but for the sake of this conversation, this is all that seems relevant at this point.)
It is very unlikely to play 32 simultaneous pitches (as you only have 10 fingers and 2 feet (usually)), but it is very likely to play many differing tone colors simultaneously. So, it is possible to further divide the audio channels by tone color. Being able to separate out the samples/output by both pitch (frequency) and timbre (tone color) means having the ability to keep distortion down, and clarity preserved.
Pipe organ frequencies break down like this: (using the A=440 tuning, so there is some degree of flexibility here and frequencies can vary slightly)
32' Pipes = 16Hz fundamental
16' Pipes = 32Hz fundamental
8' Pipes = 64Hz fundamental
ETC.
So, what I'm looking to do is find a design that can handle the 8', 4', 2' and higher stops. Ideally the lowest it should ever need to play is 64Hz. And since the software algorithm is so good it can utilize whatever outputs it has available by cycling pitches away from others, then I just need a cabinet capable of producing a nice, clean, balanced output.
The idea is not to have a small number of speakers trying to handle the entire organ, but rather, as many full range speakers as possible, so that each pipe sample is reproduced accurately.
The sub channels are typically mix-downs of other routing, but I will likely devote a large setup, utilizing a couple of crossovers, in a biamping situation to create another larger full-range channel or two just for those. (That is a separate conversation, though as I'm trying to focus on the smaller pipes for the moment.)
There are other considerations as well, such as the room, which is very large, and the location of the speakers once they are installed. (See attached pictures)
The speakers will live up on top of the large wooden rooms that flank both sides of the altar. There is roughly 12x12' of space on each side for all the speakers to live.
Attachments
Can you also divide the sound by voice? In old (1960's) analog Allen organs, the reed stops, especially trumpet, were fed into front loaded horn speakers, strings and flutes were in various combinations of conventional speakers. They used the speakers which had the best characteristics for the sound they wanted to reproduce.
Also, if you have speakers on both sides, you should divide the organ's output by manual, e.g. great on right, swell on left, etc. You do not want a solo stop jumping between the two sides of the room.
Also, if you have speakers on both sides, you should divide the organ's output by manual, e.g. great on right, swell on left, etc. You do not want a solo stop jumping between the two sides of the room.
Yes, you are correct and routing will be done that way. But for the sake of this thread, in trying to find suitable speakers, I omitted all of those little details that only we true organ-lovers understand.
But yes, they will be divided further as needed.
I'm just looking for the right speaker design at this point to handle the majority of the ranks.
I'm working on a similar project, and am considering the Pensil P7.3 for the 8' and 4' ranks. These will provide an 8 octave range (64Hz - 16KHz).
Still considering the upperworks.
Are you using a mono sample?
Still considering the upperworks.
Are you using a mono sample?
You did, and Scott has not forgotten.
I'm about half-way through (figured I'd work them out then send as a job-lot so we don't need to traverse the electronic equivalent of a PC wiring loom or my tame black-hole filing system trying to find them. )I've heard conflicting takes on using mono vs. stereo. For the time-being, the consensus is stereo.
Thank for the tip. I will definitely look into those. I have started the Windermere^2 design already, but this will be the 4th or 5th speaker design I've tried out for this project.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
My initial attraction to the Pensil P7.3 is that they look VERY easy to build, and are considerably less expensive. (Only one driver!) When building only a pair, that's not much of a comparison, but since I'm looking at building 12-16 PAIRS, that really adds up.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.