Baldwin Multi Waveform Organ

[rjsilva]

Really interesting details here! Thanks to everyone who has contributed.

[jbird604]

The more I peruse the manual, the more I believe this organ was an awesome idea, but actually not nearly as revolutionary as you might think. Yes, the major difference between the Multi-Wave and one of the large analog oscillator organs they built at the peak of their prestige is simply that the tones are being read off the disks by photocells instead of generated live by oscillators. There is no real-time playback, no more than in Allen's MOS system, but apparently the tonal result was more impressive than MOS. I'd lay that to the "long samples" stored on the tracks of the disk, in place of the single half-wave "recipe" used in MOS. And there may have been more frequency variation going on, due in part to the slight unsteadiness of the mechanical system.

Baldwin was always good at making a simple technology sound better. They employed their patented "tone expander" system as a basic building block of this organ, which probably kept it from sounding as sterile as the MOS organs were said to sound. Without the tone expander, I assume that all flute tones would have been locked in frequency and phase, since they were all read off the same spinning disk. Likewise with other tonal families. But you don't hear that in the recording, you hear a broad ensemble. That was Baldwin's forte. Too bad Allen didn't have something as effective to hang on to the outputs of a MOS organ.

But largely, this was just an analog organ with tremendous attention to doing everything that could be done back in the day. No skimping on keying circuitry or attack sounds. Keeping individual stops as independent as possible in the audio system. And using a really good tone source that they didn't even have to invent, since it's basically movie film technology.

The failure of this organ is a testimony to the rapid takeover of the organ industry by digital tone generation. No amount of novelty in the analog realm was going to hold ground for long, as digital would eventually be able to do it "faster, cheaper, better, and smaller."

[jbird604]

Ok, I'm going to revise some of what I said just above. Spent some more time reading the manual last night, and I see some interesting things that give me even more respect for that organ, and make me wonder how they were actually able to pull it off in the early 70's. Got sleepy before I could totally understand this, so I may have to revise some of my thoughts again in a few days...

(1) Contrary to my speculation, there is little or no apparent unification or borrowing. The stop list is small (for the basic setup), not much more than an Allen ADC Classic, for example. But they may in fact have a different tone source on their optical disks for each stop, though there may be a couple of stops used in more than one division, and one or two instances of using a rank as both a standard stop and as a percussive component. And the mixture is a composition created from high-pitched diapason ranks. But far less borrowing than a typical analog organ.

(2) There are 20 discrete audio channels, each carry just one or two stops. Each channel gets its own amp and speaker, though the speaker might be just one cone in the acoustic pipe array. No doubt they would've used multiple sets of the pipes, or else actual speaker cabinets, in a larger installation. But that's about as many discrete audio channels as you'll ever find in any organ that doesn't cost a real fortune.

(3) This is the part that puzzles and amazes me most: they would have had a tremendous density of audio stored on those optical disks. The base model had only two of the optical storage units, which held all the analog recordings used to create the stops. There are a total of 30 (yes, thirty!) "ranks" of optical pickups, each one presumably having 61, or at least 32, photocells, each cell, presumably, aligned with a unique track.

The layout of these 30 ranks are indicated on the diagram of the wave storage unit, and a chart relates where each stop comes from. If there were only two or three real ranks recorded, the diagram should show, for example, all the flute stops coming from the same optical pickup, or at least from the same tracks on the same disk. But that is not the case. In fact, on one disk there are three or more 8' stops that appear to be picked up on adjacent photo traces, but have three entirely different tones -- diapason, flute, and reed.

So I'm guessing they may have packed the audio information VERY tightly on those disks. If a single disk really contains 15 discrete ranks of 61 notes each, there would have to be 15x61=915 tracks encoded on each optical disk. That sounds like like a tiny number, compared to the magnetic tracks on a modern day hard drive, but in 1971 it meant making each optical track perhaps 1/100th of an inch wide, or less, considering the need for some guard band space. The focusing of the light beams and the alignment of the corresponding photo cells had to be done with great precision.

Maybe I'm wrong and there is a lot of borrowing here. But if one disk contains, let's say, just three discrete unit ranks that each serve as 16, 8, 4, and 2 foot sources, that would require 97 tracks per rank, a total of 291 for the disk, and that would reduce the spacing to about 1/30th inch. Still pretty tight, but less harrowing. I tend to think that they did it the hard way, given how good these organs sound in the recording.

The console looks very much like the console of the C-601 I used to have at my old church for a while. A super heavy and well-crafted console with excellent keys and pedals. A somewhat shabby combination action though, one of Baldwin's perpetual failings. But the Multi-Wave could've been a successful product had it not been overtaken so quickly by the rapidly improving digitals.

[MarkS]

I have the fancy brochure packed away in a box somewhere here. (I also own the demo record but my Shure V15 Type III is shot.)

As I recall, they claimed that a computer was used to draw the waveforms. And there were two waveforms per "pipe" for accuracy. For some reason, I believe they claimed that the waveforms were read using servo-controlled lasers. Is is possible?

The console appeared to be a Klann. If so, it would explain the problems with the very early solid state combination action.

[jbird604]

The service manual only covers the organ itself of course, and doesn't offer any theory of operation. Not even a pic or diagram of the photocell arrangement. That's why I wish I could see one of these in person.

The more I think about it, the harder it is to believe it even worked at all. No matter how fast that optical disk was spinning, you can't have a "loopless" pattern corresponding to each of the 12 pitches of the tempered scale. In order for the old TOFG pitch synthesizers to work, they had to be driven by an oscillator running at around 4 MHZ, and even at that speed, the exact pitches of all the notes were a bit compromised. So the math seems very hard to work out.

What I mean by "loopless" is that the pattern in any given track on the disk HAD to be totally symmetrical and could not have any spot where they took a break or cut out a bit in order to stitch the end back to the beginning. That would cause an unacceptable weirdness in the pitch and nobody would've tolerated that. Since there is no computer involved, I'm certain the system didn't just read the sample once and then re-sample it electronically, which is what today's digitals do, and what the MOS organs did in a rudimentary way.

If they had used 12 different disks in each stack, one for each note of the chromatic scale in all the octaves, they could of course accomplish that, much as the Hammond organ does. But they would've needed a massive system of gears to run all those disks in perfect sync at the right speeds, which would be different for all the notes. Of course some kind of servo-controlled drive motors regulated by sophisticated speed sensors might accomplish that without gears or belts. But that would still be a lot of disks.

Someone mentioned that Kimball had an optical organ previously, and it was plagued with belt issues, driving numerous disks from a common shaft, I suppose. So the concept just seems impossible to implement.

But the fact remains that the demo recordings sound remarkable.

[Admin]

As I recall, they claimed that a computer was used to draw the waveforms. And there were two waveforms per "pipe" for accuracy. For some reason, I believe they claimed that the waveforms were read using servo-controlled lasers. Is is possible?

It doesn't appear that the disks were laser scanned for playback. They were spun and the photo-electric pickups were fixed. It's possible that lasers were used to either burn the waveforms on the disk, or to produce the negative used to photo-etch the disks, but the pictures of the drum don't support the idea lasers were employed for playback.

I wouldn't be surprised if a computer was employed to draw the waveforms. Remember that the angular velocity of the disk changes with diameter. For any given waveform at any given pitch, a re-entrant pattern would have to computed based on its circumference and angular velocity at that circumference. It's really no wonder given the tolerances involved that this system would prove to be impractical.

[quadibloc]

I had learned, from another thread in this forum, that a Canadian, Frank Morse Robb, had invented an electronically-amplified tonewheel organ before Hammond. (He was even before Richard H. Ranger, an American who also did that.) But from photos of the tonewheels in Robb's wave organ, I saw they were cylinders, and the teeth - parallel to the cylinder surface, not perpendicular - were cut in the shape of waveforms rich in harmonic content. So he was aiming at the same goal as was part of at least some of the designs of photoelectric disk organs.

[myorgan]

Did Frank Morse Robb pre-date Thaddeus Cahill? Some time ago on the Forum I learned of the Telharmonium (https://en.wikipedia.org/wiki/Telharmonium). Something also makes me think I saw a video, but I could be wrong.

Michael

[DrGeorge]

I repaired Baldwin and other organs in the mid 1960's and again in early 70's. I learned repair from a much older tech who told me Baldwin had pursued optical organ technology in the 1930's, pre WW2, and these used drums of photo electric tracks rather than disks like Kimball and others. But these were never were sold, moving to electronic master oscillator-divider organs after WW2 like Models 5 and 10. However, in their photo disk research Baldwin had created precision disk engraving machines which were put to another use in a Little Rock, Arkansas division, Baldwin Electronics Inc. (BEI), still in business as BEI Precision. BEI made extremely high precision optical shaft encoders, converting rotary shaft position of seconds of arc into digital format for a computer, using optical disks. I attended an engineering ‘lecture' about 1966 by BEI engineers and years later worked on tape equipment using BEI servo motors with encoders.
History at this website: http://www.beiprecision.com/heritage/

Although Baldwin's analog organs stayed with MO-divider versus oscillator per note (Allen, Conn et al), they added things such as the electronic “Choratone Projector” (Patent 3004460) from the 1950's to create a sort of pipe chorus sound, later, a solid state version called something else but similarly added beats to the band-split composite sound without detuning the oscillators. Baldwin also used multiple MO-divider sets, off-tuned a few cents from the ‘main' and wired to the keyboard so octaves weren't locked, starting in transistor Model 6 (2 main generator sets) and Model 11 (3 main sets plus 2 celeste), a compromise in number of oscillators to keep tuned while avoiding ‘locked' octaves. The three-manual Model 12 was all tube and may not have used this scheme.

While attending a factory tech school (Bill Stevens et al) about 1972 in Fayetteville, during the factory tour I saw a wave organ being built and learned a bit of how it worked, but never serviced one. The Wave speaker cabinets were extremely unusual designs.

The Baldwin patents on the ‘multiwave' organ explain the theory. It has origins in the late 1940's. based on two disks, pitch disk and voice disk. I had also heard that the optical apertures were computer generated/machined to create the voice waveforms, but that must have been in later versions. The pitch disk seems have been ‘binary' (square wave on-off), something like the shaft encoders, moving across shaped apertures for the waveform disk. Note in the later patent there were separate apertures for Chiff. The pitch disk had a finite number of apertures, same limitation to note frequency resolution as the number of gear teeth in a Hammond.

No lasers were involved in the playback of the disks, instead an bright incandescent bulb, like from a projector. The waveform disks appear to be patterns to generate the desired harmonic structure rather than a ‘soundtrack' or any direct recording of a pipe sound.

You can read the patents, just enter the number, at www.pat2pdf.org

2576759 Means for producing tones, Edward M. Jones, filed 1949
3023657 PhotoElectric Musical Instruments, Jones, Love, filed 1955
3249678 PhotoElectric Organ, Jones filed 1962
3930430 Photo Electric Organ with Chiff, Jones &Wayne, filed 1974

[jbird604]

This topic continues to interest me, as the demo recordings were so authentic-sounding. Of course, a dirt cheap discount store keyboard of today has more sophisticated tone generation capability than these monsters ever did, so there would seem to be no real "musical" reason to resurrect one. Still, it's like old cars and old trains and such -- some people just want to know how they worked.

That's why I would jump at the chance to actually see one, even one that didn't work, to settle some questions that still don't make sense to me.

[gtc]

... there would seem to be no real "musical" reason to resurrect one. Still, it's like old cars and old trains and such -- some people just want to know how they worked.

I stumbled onto this topic via a link in a web article, that in turn was linked within a web article, etc, etc.

When I heard that LP in the YouTube video I was blown away by the sound.

Although not tube-based, given its magnificent sound and its optical disc technology, this is just the sort of thing that I would definitely consider taking on as a resto project, especially given that there is factory documentation available, although with the proviso that the organ was donated to me and that all I had to pay was the (reasonable) shipping cost to my workshop.

It's not as though I don't have a long list of resto projects staring me in the face, but the thought of getting an 'exotic' beast like this up and running again really excites me.

The photos provided by spharorgan are priceless. I wonder where he got those from?

[myorgan]

When I heard that LP in the YouTube video I was blown away by the sound.

The photos provided by spharorgan are priceless. I wonder where he got those from?

GTC,

The sound on the recording was excellent, I agree-even though the pitch was unstable on the record. If you read the first comment underneath the video, however, you get a different story. The poster states (s)he has played all the Baldwin organs in San Antonio, and none still exist. As the poster further states, I wonder why?

I believe Spahr Organ managed to get his hands on one of the organs, as his company is in Cincinnati, where Baldwin headquarters were for decades. That may have had something to do with his access to one of these multi-waveform organs. If he is who I think he is, I worked with his brother for a period of time. The brother is also a musician but has chosen a different career field.

Michael

[gtc]

Actually, when I go back and read the thread properly he did say "Cincinnati's Masonic Temple downtown still has and uses TWO of these organs ... I'm actually running out there next week [Mar 2018] to do some repairs on it."

I wonder what happened to the TWO of them?

Love the poster's reference to "a contumacious organist". I'm going to declare contumacious my word of the week.

[myorgan]

I also found two interesting articles about the organ here:

• https://friendsofmusichall.org/cinci...astings-organ/
• http://www.musicincincinnati.com/sit...nary_Tale.html

Other observations:

• It appears the Hook & Hastings organ lasted 23 years before needing to be rebuilt or repaired.
• Then 19 years later in 1920, Austin essentially rebuilt the organ as one of their instruments while retaining the organ case & the Hook & Hastings name.
• Then 40 years later in 1960, the organ was a candidate for replacement or repair due to neglect, but no one stepped up to the plate for repair.
• Then in 1974 the Baldwin replaced the pipe organ, which was thrown away by all accounts. As far as I can tell, 47 years later the Baldwin is still going strong.

Interesting lengths of time when comparing the pipe organ to the electronic organ. So far, the electronic appears to have lasted the longest! However, according to the articles I linked above, the jury's out on whether the Baldwin was an improvement over the pipe organ or not.

Thank you for resurrecting this thread.

Michael

[Organkeys Jones]

That Cincinnati Baldwin did not last 47 years. Looks like it was replaced by an Allen R-450 many years ago.

https://www.allenorgan.com/www/insta...085/page1.html

[myorgan]

Organkeys,

It's always nice to have all the facts. I was wondering about the organ lasting that long, but I didn't have all the facts.

Michael

[gtc]

I also found two interesting articles about the organ here:

• https://friendsofmusichall.org/cinci...astings-organ/
• http://www.musicincincinnati.com/sit...nary_Tale.html

So that's Cincinnati Music Hall, but spahrorgan mentioned Masonic Temple.

• Then in 1974 the Baldwin replaced the pipe organ, which was thrown away by all accounts. As far as I can tell, 47 years later the Baldwin is still going strong.

If that's the case, then I can see jbird saddling up for a ride there asap.

Maybe we should call them and confirm.

However, according to the articles I linked above, the jury's out on whether the Baldwin was an improvement over the pipe organ or not.

Love this quote: “Those things are not pipe organs. They're phony. I can see playing them on gigs and for skating rinks. They don't belong with the CSO. Something should be done to get some real music sounds back into Music Hall.”

[jbird604]

Yes, I'm still quite curious about this device. I'd still like to know for sure how they managed to get all the necessary waveforms recorded and stored, and how they can faithfully play them back with optics. It seems to me that it would be necessary to have dozens and dozens of spinning disks in order to hold all the waveforms at the ready, available for instantaneous playback at the moment of keying. And that there would need to be elaborate schemes for spinning the disks at carefully controlled speeds.

Maybe I'm missing something, but it seems to me that the system would require one optical pickup for every note of every rank. In other words, as many optical pickups as an equivalent pipe organ would have pipes. And that would seem to be a costly way to build an organ. But maybe there's something I'm not quite understanding about how it works.

I do have the service manual, and it's got pictures and diagrams, but I still don't really understand how this thing makes music, and I'd love to see one up close. (Not sure that I'd drive to Cincinnati just to see one though.)

[Organkeys Jones]

Looks like that Cincinnati Baldwin was replaced by an Allen R-450 many years ago.

https://www.allenorgan.com/www/insta...085/page1.html

[Organkeys Jones]

I fixed the link. Had previously copied from an early post on this thread, but now it is direct from Allen. Sorry about that.

[Admin]

Yes, I'm still quite curious about this device. I'd still like to know for sure how they managed to get all the necessary waveforms recorded and stored, and how they can faithfully play them back with optics. It seems to me that it would be necessary to have dozens and dozens of spinning disks in order to hold all the waveforms at the ready, available for instantaneous playback at the moment of keying. And that there would need to be elaborate schemes for spinning the disks at carefully controlled speeds.

If you refer back to post #23 you can see that there are only three or four large rotating disk assemblies each with over a dozen pickup assemblies. Each pickup appears to contain the full compass of an individual rank as I'm counting roughly 64 connection points on each pickup. From this you can see that the optical tracks must have been incredibly close together. Each pickup is slightly offset from the others implying that the various waveforms are interlaced on the disk. Note as well that the higher pitched ranks are positioned toward the outer diameter, and the lower pitched voices toward the center. This makes sense since the angular velocity of the disk decreases with diameter. As I mentioned earlier in this topic, the trick is to somehow figure out the correct pattern and placement of the waveform on the disk in order arrive at the desired pitch at the same time ensuring that the pattern is reentrant.

As I also mentioned in post #25, Welte, had an optical organ like this back in the thirties, the mechanism of which is pictured in that post. It appears from the description on the photos that Welte had employed twelve assemblies, one for each note of the chromatic scale. As Baldwin appears to have been able cram much more information onto a single disk, it most have been incredibly difficult to align those pickups in order to avoid crosstalk. Due to the size of the tracks, the units must have been well sealed against dust as given the size of tracks even a small spec of which would result in noise.

I'm assuming that the tone generation was free running with all the tones available simultaneously for keying electronically. It did occur to me that perhaps the "pickups" were transmitters consisting of an array of LEDs which were switched individually to effect keying, but that doesn't make design sense as envelope shaping would still have to applied downstream of the disks.

In short, an incredibly complex system with substantial mechanical manufacturing and stability challenges. It's really hard for me to believe that after all these years any on these organs are functional and playable.