That is a fine machine! stike

Obviously not analog, but the Uhbik Frequency Shifter is pretty nice:

u-he.com - UHBIK

Encore Electronics' module uses a digitally driven quadrature oscillator:

Encore Electronics

One more thing to try...Set up a phaser..I like the Spx 90-900 symphonic patch. adjust depth and speed.Can do that also.You might find using a DDL plugin using the modulation section will work also.The mod speed can do an effect like that. Just play around a bit.Mario.

Hey Mario,

I just ordered an old SPX90 to try it. For $200....

I remember when these things were new, and $200 was a fortune!

Thanks..

Awesome...love them.Good price too.Let me know how it works out.M

Hi. I was fortunate to discover this forum after doing a google search on a mis-spelt version of my name. I am Keith Adkins (not Keith Atkins) and I worked as an engineer at what used to be called EMI studios - now known as Abbey Road studios - in London in the early seventies, in the period immediately following the break-up of the Beatles. I am the designer of the frequency translator used on Dark Side of the Moon by the Pink Floyd and other albums. Anyone further interested in this is welcome to contact me via the forum.

Keith, welcome to Gearslutz. I'm sure there will be many here wanting to pick your brain.

I'm not old enough to answer your question. I'll check back in 10-15 years.

I did this on a song I mixed that had this sort of feel on the chorus, and the swirlies we got with the Leslie, miced stereo and blended back in were really neat. It also hapened to be a Goff modified cabinet that they lent us, so it had really fantastic distortionthumbsup
Not the sound you are looking for, but it is certainly a worthy trick to keep up your sleeve.

edit, deleted what I wrote as I thought about it and found I was probalby wrong.

What it almost sounds like to me on that oohs ahhs part that I too have always loved is that either someone is riding the control for the device (what sounds to me like an autophaser type of circuit which is basically the same pitch altering delay-based circuit as what we're talking about here) so it's more active on the ahhs than the oohs.

But maybe it was something cool like an envelope follower that increased the mix of the pitch altering circuit with louder input levels! That would be great and not tough to do. Basically a phaser type of thing that does it's effect only after a certain level threshold has been reached and then uses typical compressor style attack/release/knee or adsr type of circuitry to adjust the mix control for the phaser effect or whatever.

I WANT ONE. Maybe I should try to build one. if only I had enough time, I'm having trouble finishing the projects I already have...

Cheers,
Don

Welcome to the forum! As you can tell, we've all been trying to guess what you were up to all of those years ago. Adding some truth to our speculations would be very much appreciated!!

Mike

Really good to meet you Keith!

I hope you didn't read my probably wrong post above before I edited it LoL. The more I think about frequency modification the more I'm certain it could be done as someone described earlier.

Anyway, was it sort of an envelope-following phaser? that's my guess :-)

Cheers,
Don

Hi folks, thanks for your interest, are you all mostly in the USA? Pleased to meet you all, I am in the UK. Before I joined EMI in London in 1969 I worked as a development engineer in radio communication systems and the frequency translator borrowed from that experience. It took about 6 months or so to design and build during quiet times often late into the night whilst being on call within the studios as an engineer providing technical support to studio sessions. I negotiated a total budget of GBP 100 with the studio manager, plus my time. That was the total cost. The device used thermionic valve (tube) technology, I remember that around 10 thermionic valves were used in the final device. As far as I am aware no photographs exist of the device and it had fallen into disuse and disrepair by 1975 which is the year I left the studios. It worked like this. It shifted the audio spectrum in a linear fashion, not harmonic. eg 1000Hz becomes 1010Hz, 2000Hz becomes 2010Hz, 3000Hz becomes 3010Hz etc. The audio baseband signal modulated a 10.7MHz carrier producing a resultant double sideband amplitude modulated radio signal. The carrier and one sideband were then filtered out producing a 10.7 MHz single sideband (SSB) radio signal with suppressed carrier. And this is where the fun starts. You then re-introduce the carrier at a slightly different frequency and demodulate the resulting signal producing a reconstituted audio passband that now has the necessary linear shift. If the re-inserted carrier is itself modulated at very low frequencies then the shift itself becomes variable and a multitude of various weird sounds can be produced. Feeding the output back to the input enabled even more possibilities and the device was exploited very effectively by AP on DSOTM. It was used on a number of other albums and continued in use as long as I was available to maintain it. It survived for quite a reasonable time in a very busy studio operation especially considering that it was constructed in prototype fashion on an unprotected open aluminium chassis with no case. At first the project had no name at all and was simply referred to as "my" project (the Keith Adkins project) but it was named as the Frequency Translator on completion. Any other questions I'll do my best to answer, hope this helps.

Keith, welcome to the forum. Can you share some of the other Albums/songs that your Frequency Translator was used?

Keith,

That is fantastic. Thanks so much for the information.

I'm curious, have you ever considered producing these as a commercial project? Or would EMI own the intellectual property since you were working for them?

There is certainly a booming business in reproductions of 70s audio gear, much of which I think is much less spectacular than your frequency translator!

Mike

Could anybody please upload some examples from DSOTM where this has been used?
I'm dying to listen to this piece of history! :D
Thank you Keith for making this forum even more interesting than what it already is!

Bas

Keith,
Thank you so much for the lowdown on a piece of musical history.
What were you thinking that made you craft this device?
Just curious, because I know a lot of the neat effects we have came from funny "what if?" thinking sessions.
I am honored to meet you by way of the web... What a great thing!

pultech

If you can find a communications receiver (not everyone will have one!!) and receive an SSB voice transmission you will be listening to the original audio after the modulation and filtering processes mentioned above, plus of course any effects created by the radio propagation. Most communications receivers will have a control called a BFO (beat frequency oscillator). I believe some CB radios that are SSB capable will also have this control, and it may be called a "clarifier". Adjust this control and you will be making a small adjustment to the re-inserted carrier mentioned above. This mis-tuning will give you the linear frequency translation effect, admittedly over a small passband, typically 300Hz to 3KHz. My original thought was what would happen if you created the same effect deliberately using the full audio spectrum and so I set out to do just that.

Deuce 225 & mikelevitt

I will try to answer your questions within the next few days.

Thank you.

Hi Keith, thanks for chiming in!
This is fascinating...

mikelevitt

It had crossed my mind. I don't think i'd really want to start a manufacturing operation but I might be interested in working in some advisory capacity with someone who was. On the question of who owns the intellectual property I'm not sure there would be any actual ownership of the processes involved as the techniques are well established in the field of RF design. It is the unusual use and application which may or may not result in some intellectual ownership. But I am not a lawyer. Of more importance perhaps would be the commercial use of the name "frequency translator" (which is not completely unique I agree as frequency translation itself is a recognised term). But in the specific field of popular recorded music the name is now obviously linked with the Pink Floyd DSOTM album. This possible means there might have to be some discussion involving EMI, the artists and myself were someone to become interested in manufacturing a contemporary implementation of the device. I wouldn't imagine this to be a major problem though if everyone behaved reasonably. On the relevant question of my employment contract, well it was a long time ago and I am not sure if a copy of this would still exist, so that particular angle may not be of any great relevance now. But all that is very hypothetical of course!

Keith, Excellent... Yes! I know what you are talking about, I use SSB on boats sometimes. The old ICOM I used had the "clarifier". I always thought that was cool sound.
So I guess the lot of us might be prone to twiddling knobs in real time during a mix to achieve some nifty sound warping.
Great story about a super rare (dare I say, extinct), piece of kit.
Thank you for letting us all hear it!

EDIT: IT's the song called TIME I was thinking off. the female chorus background vocals do that ooh ahh thing.

dude, just listen the the song TIME (corrected now!), the female choir background vocals in the bridge parts (or one could call them chorus parts although imho there isn't a chorus in that song).

go to listen.grooveshark.com if you don't own one of the greatest albums in history :-)

cheers,
Don

So Keith, the controls would have been completely manually operated during the mixdown (or maybe submixdown) of the female bg vocals in DSOTM, is that right? there wasn't any kind of auto-level trigger in place that would make your effect stronger with louder input or any such side-chain type of trickery?

another sort of related question: Does the sound they got in DSOTM equate pretty well with the sound your device created in one of it's set it and forget it kind of settings or do you feel that the operator was turning knobs in real time during each single 4 second long "Ahhhhhhh" section to get that flowing watery musical sound?

Some effects just sound a certain way and of course the operator is always of great importance to how well it's used, but if it's fully manual control then I'm basically asking if there would have been knob turning and tweaking in real time or would they likely have adjusted it until they liked the sound and then brought it in and out of the mix by riding faders or a blend control of some kind?

Since I don't have that type of device around, I like to use phasers etc for that kind of effect (which I do use on occasion, thanks to your hard work for making it possible!). I'm trying to decide how much of the magic was the operator (and therefore more easily replicated by someone like me with realtime tweaking) or the device (in which case I'll have to compromise).

Cheers,
Don

It's been a bit, but it sounded like there was an LFO controlling the frequency of the re-inserted carrier, and the rate control was being tweaked during mixdown. To these ears anyhow

yea, and only very subtly was the rate changed during mixdown I think from memory here (it's all pretty slow if I recall).

That's what I'm figuring, but worth asking him anyway.

cheers,
Don

dkelley

The frequency translator had only two controls, power switch and “tuning”. The tuning control operated a 180 degree travel variable capacitor via a hefty reduction gear. This variable capacitor, with a maximum value of about only 50 pF, enabled small changes to the frequency of a 10.7 MHz quartz crystal oscillator. which provided the re-inserted carrier. The reduction gear enabled the “offset” to be controlled within a fraction of 1Hz, that is less than 1 part in approx 10 million. In SSB radio design the oscillator constraints would be less stringent because the intermediate frequency was typically 465 KHz not 10.7 MHz - and also because full intelligibility would not require such a precise alignment of the re-inserted carrier. 10.7 MHz was chosen as the IF for the frequency translator because the higher audio passband requires a higher carrier frequency with respect to filter design and because, not least, 10.7 MHz components were easily available as this was the IF used in domestic VHF FM broadcast technology. Remember my budget of 100 GBP!!

Because this was an analogue device using thermionic valves it was much easier to set the inserted carrier accurately than it was actually to keep it where you set it. In other words the frequency would drift away from the setting, an effect that reduced but was not eliminated the longer the device was powered up. Not only that but because of the fairly rudimentary open chassis construction etc it was not easy to keep the waveform of the re-inserted carrier “clean” and this normally unwanted effect provided a kind of additional character to the sound you might say. In operation the tuning control would be set to an offset which provided the necessary effect. Setting it several Hz away from the centre would provide a kind of underwater effect noticeable on some of the guitar work on DSOTM whereas setting it to 1 Hz or less would provide a different kind of sound but equally unreal. The frequency drift and the “unclean” waveform provided an interesting lack of predictability and plenty of surprises no doubt, not all of them unpleasant judging by some of the enthusiastic remarks about this particular contribution to the album.

So the operator had only one control to play with and its use must have been very confusing when trying to take the frequency drift into account. The balance engineer did of course have all the usual possibilities provided by the mixing desk which included adding the translated sound to the original, feeding the output of the translator back to the input, adding reverberation. But I think mostly commonly the technique would be to set the “tuning” control to a suitably interesting sound, cross your fingers, and then simply insert the sound into the stereo image as required. Perhaps Alan Parsons would be able to remember exactly what he did on DSOTM, but you’d have to ask him that.

As regards replicating the device in 2009, I think it is quite possible that the exact sounds might be difficult to re-create with cleaner more stable silicon technology although how close you could actually get is not something I’d like to try and predict. Which leads to the interesting possibility of how easy it would be to build a true replica using thermionic valve technology. Having already done one of these I think I might leave that task to someone else. Any takers?

Thank you for helping me to re-live the design process Don, your questions were right on the nail.

Keith,

Thanks for sharing your engineering knowledge with us. It's been fascinating!

Darren Duke

Deuce 225

I don't have any actual list I'm afraid. I think that one of the most dramatic examples can be heard on the solo guitar work starting at 1:19 in track 7 (Any colour you like) of DSOTM. It is used quite differently on the female vocal accompaniment in track 3 (Time) from 3:00.

There is another example on Roy Harper's Valentine Track 4 Twelve Hours of Sunset where a kind of growling underwater guitar backing provides a certain mystery.

I know there are other examples but I don't recall them at the moment. The translator was only used for a few months. Because of it's non-robust prototype-like construction, really quite unsuitable for everyday work in a control room, it often required repair. Later in 1973 my role as an engineer changed within the studios and I was no longer available to repair it. The last time I saw it was in 1974 sitting in a cupboard probably awaiting some ignominious ending. But perhaps not, maybe it's still there somewhere, I shall have to enquire.

I just have to say this is pretty incredible having Keith chime in... Very interesting for a young kid like myself to hear some stories of the past... I am a huge fan of that era of music... #1 record and radio city are 2 of my favourite albums of all time... and to compasspnt, you contributed to some of the most influential vocal parts in rock history in my opinion... Would love to hear more stories if you care to share... Now more associated to this thread...

Hello Susie by The Move... Has kind of a similar effect... I guess its more of a phaser than a flange but it kind of reminds me of those background vocals on "Thirteen"... Maybe a little less wet though

Fantastic info Sir, thank you so very much for taking the time to join in our little discussion and for travelling down memory lane in such detail!

I'm strongly considering building a replica of your device based on what you've described, although I'll have to put a lot more work into the plan than I've got in my head at the moment heh heh

I've always loved throwing things together on a breadboard. Certainly never hurts!

Thanks again on so many levels for having created this device in the first place. You not only have given me some of the best musical memories during my adolescent life and later but also have now given me something concrete and human and current to go along with it.

This is why I love this website so much!

Please, if you have the time, stick around the website and pop in where you feel like the readers' knowledge level is high enough to truly benefit. I know most people here will appreciate your stories and ideas!

My best,
Don Kelley