Friday, May 29, 2015

The Incredible Mutation (And Restoration) Of The 360 Systems Frequency Shifter

The 360 Systems frequency shifter is a really interesting and fairly rare piece of analog gear, and well suited for use with synthesizers. There are at least two different versions of this device, with the earliest ones housed in a green case...

... and the later versions, sporting some circuit upgrades, housed in a blue case.

I remember first reading about this as part of the equipment listed on the back of the excellent first album by Larry Fast/Synergy "Electronic Realizations For Rock Orchestra". You can see one of these sitting on the top of the Moog Modular 15 in his studio at that time...

 

I had no idea what it actually was, but MAN did it sound cool! FREQUENCY SHIFTER!

Larry not only helmed the many wonderful and inspirational Synergy releases, but also added his distinctive magic to albums and tours by Peter Gabriel, Nectar, Foreigner, and others. A true pioneer.

Thanks, Larry!

My unit is a very early one, I think. A look at the serial number would suggest that it might be made in early 1975, perhaps serial number 8? Maybe, maybe not.

There's quite a bit of rework going on underneath the board. This is often the case with very early boards, as mistakes are discovered and repaired to make the early units sellable. A close examination of this board showed that entire power supply was laid out backwards (!) so all the diodes and capacitors had to be installed backwards to make it work correctly. The blue jumper wires make the proper connections here and there, and parts that were either forgotten or added later to make things work correctly are soldered on back here as well. I really like the 360 logo etched into the board. If you look closely, you can see that they actually used it to connect two traces!

 

The print ad for this (thanks, Riley!!) at the time made some very interesting suggestions about the capabilities of this unit, in typical 1970's flair....

Turn a violin into flannel bass??!??!? What the what? Good God, Y'all!

Those with sharp eyes will note that the case shown in this ad isn't the same case used in either the green or the blue production versions. This picture was probably the original proto unit.

As I mentioned earlier in my blog, I acquired one of the earlier green ones along with my Moog 55, and after a few simple mods and some recalibration, I was able to use it on various projects for a number of years. I remember calling the good people at 360 Systems at the time I got it to see if I could get a schematic, and it took a while for them to actually locate one at the factory! I eventually got a huge blueprint-size schematic in the mail, free of charge. Impressive customer service!

The most useful mod to it at the time was replacing some of the single turn trimmer pots in the balanced modulator sections with 12 turn trimmer pots, to allow for a more accurate calibration (the four green rectangles at the bottom left quadrant).

 

 

Adusting these pots allows you to reduce the amount of background noise and signal bleedthrough, and with the new more accurate trimmers in place, calibrating it was much easier, and improved the performance by a noticeable amount. I also at that time upgraded the generic opamps in the audio path with quieter devices.

In use, there are some pretty interesting effects to be had from this device. As you turn the shift knob, one output shifts things upward by the frequency set on the shift knob, while the other out shifts things down by the same amount. Things can get crazy pretty quickly!

With just a small amount of shift, and the up and down outputs panned hard left and right in a mix, a beautiful chorus sound is created, different from any others. It's said that the swirling "oooh" and "aaah" background vocals on Pink Floyd's song "The Great Gig In The Sky" are treated with a prototype frequency shifter, and that's what it sounds like to me...a sort of growly, swirly wonderfulness.

 

There's also a Ring Modulator output, to create all sorts of clangy, metallic timbres. I read somewhere that John McLaughlin used one of these units in the mid 70's with a footpedal to sweep the oscillator, and that seems likely to me as well...imagine the hyper-speed metallic goodness. Go Johnny, Go!

 

 

Ring modulators were starting to enjoy some well-deserved popularity in the early 70's. Oberheim built a stand-alone effects box type device (dig those crazy graphics!!!)...

 

...and also built a slightly modified version for sale by Maestro (they look quite a bit alike, eh?).

 

Artists like Jon Lord, Joe Zawinul and Jan Hammer were soon using them on their electric pianos and organs to produce a more interesting synth-like texture and create a pitch bend effect for solos.

For many artists at the time, this often served as a gateway (or at least a sideways entrance) to using synthesizers like the Minimoog or the Arp 2600, as they became more readily available.

 

 

I loaned my 360 Systems Frequency Shifter out to a friend, who at some point apparently decided that it was now his. When I finally got the unit back, he had proceeded to remove the circuit board, discard the case, and mount it into a single rack space casing. Like most of his (always very nicely made) devices, it was basic black, and all knobs, jacks and switches were unmarked. Feh.

After a number of phone calls, I was luckily able to track down the original case languishing in the back rooms of a recording studio (thanks, Roger!), and decided that when I got the time, I would restore it to its former 70's glory. The case had faded and discolored somewhat with time, but I still loved the sheer weirdness and funky mojo of the overall package.

GREEN! It's really an underused color.

My original schematic had gone missing, but after some sleuthing, I found a copy on the internet. I'll publish it here for those who might be interested. This would seem to be the scematic for the later model blue Frequency Shifter, as it differs in some areas from the components used in my unit, and the date on the print is from 1976. It was however plenty close enough to allow me to identify all of the necessary wiring locations on the circuit board to successfully transplant the board back into the original enclosure.

This looks remarkably similar to a schematic I saw in Electronotes back then for this type of device. It's certainly possible that one might have inspired the other...

 

Some important parts were missing. The original power transformer was gone, as was the cool green illuminated power switch and original potentiometers.

This also seemed like a good time to replace the electrolytic capacitors, as they were now 40+ years old, and the rubber seals were dried and cracked. The opamps in the audio path had already been upgraded a number of years ago, and I also wanted to extend the frequencies of the shift oscillator, and provide better CV control of the shift frequencies so I could easily control it from the Moog 55 and other devices.

The original unit had line level and instrument level outputs ( just the line level outputs scaled down with a resistor divider). I thought that the panel space for the instrument level outs would be better served by making them the inverted phase versions of the line level jacks. This could make for some interesting stereo effects during shifting.

I found a transformer from Mouser Electronics that fit the same mounting holes, and was just a bit larger voltage-wise than the original. I also found a functional green illuminated power switch...nowhere near as cool looking as the original, but it would certainly do until a better replacement could be found. These were fitted, along with a new power cable and strain relief.

 

The front panel was cleaned, and the new power switch, jacks, knobs and pots were installed. You can see some of the discoloration around the screw holes, but it still looks pretty good, I think.

 

 

Adding the parts for the output inverters and the improved CV scaling would require the addition of a small board to hold the necessary circuitry, so I built everything on a small perfboard and mounted it to the inside front panel of the unit. This made wiring up the various outputs easy and kept the lead lengths short.

When I fired it up initially to take a look at everything, I found that that the transformer voltage was indeed too high. This made the voltage regulators on the main board overheat since they now had to dissipate the extra power to bring this higher voltage down to the required level at the main board.

I added a couple of small 5 watt resistors to each side of the transformer outputs to soak up the extra voltage, and secured them to the top of the transformer with some silicon rubber adhesive. This reduced the power dissipation of the regulator, and now things were running nice and cool again. I also upgraded the 12 turn trimmers to grey 20 turn trimmers, since one had gone intermittant after all these years.

 

 

 

 

To extend the frequencies of the shift oscillator, I bought a 5 position rotary switch to replace the original 3 position switch. Now I had a position below and a position above the original panel marking. I added the appropriate capacitors to the shift oscillator circuit for the two new positions, and that gave me a lower and higher shift range, as compared to the original circuit.

After a bit of wiring to connect all of the front panel controls, and some tesing to verify that all of the various circuits were functioning correctly, it was time to reassemble the board into the original case, and recalibrate the system.

 

The unit is still missing its unusual gigantic radio tuner dial to set the shift amount, but I believe that it's merely biding its time hiding in my storage space. Until it returns, I did manage to find a smaller version to keep the front panel from looking too barren.

And so here it is, in all of its Mighty Green 70's glory, its Mojo and Mighty Shifting Powers intact!

Good God, Y'all!

 

 

 

17 comments:

  1. hi, interesting post; I've always wanted to try one of these 360s out. I never see them for sale anymore. It looks like a relatively modest circuit for a frequency shifter. You didn't happen to scan the PCB by any chance? I'd love to etch a board if I had the layout. synthbay at gmail

    ReplyDelete
  2. Hi!
    Sorry, I didn't scan the board, but I doubt that the layout is too critical for this. Big project, though...
    I'm curious just how many of these were made...the one I've got was clearly an early one with lots of rework involved to get it going. The Blue ones were probably much revised, but I've sure never seen many of either over the years.
    If you've got access to Electronotes, be sure to check out the one published there in the early 70's...as I remember it looked pretty close to what I saw in the 360, and their schematic included all the cap values for the filter networks, which the 360 schematic does not, for some reason. I spent a bit of time trying to resolve these values by comparing them to my board, but gave up after a bit with more pressing things to do.
    These are really cool units...it's a shame there aren't more of them around.

    ReplyDelete
  3. Hello. If you could scan the board, you would make two people happy!

    ReplyDelete
  4. I'd say as late as the mid-2000s these would come up for auction regularly, maybe 3 or 4 times a year. Always the blue-face model. But even back then, they were upwards of $400-$500. More than I wanted to spend (if only I knew how reasonable that would seem just a few years later). Just like everything else rare, ebay has turned it into a highly coveted commodity. I haven't seen one for sale in years now.
    There's enough parts so that it wouldn't be real easy to breadboard (not for me anyway). But compared to other frequency shifters, it's downright light. I'm only seeing 5 dual op amps in the photo(plus a 1495?), on a single sided PCB! It'd be pretty DIY friendly if there were a PCB layout to etch.
    I guess it would be a pain to take it all the way out of the case, to get it onto a scanner. If you ever have it out for repair or something, it'd be cool to have the layout artwork up on the web.

    ReplyDelete
  5. A friend of mine has a more or less complete archive of the early Electronotes newsletters from the old Emu Systems library, so I'll try to remember to ask him if he can dredge up the issue with the frequency shifter (sometime between '73 to '75 or so IIRC). Getting those cap values in the input stage would be a key part in getting a clone running. These are all the polystyrene caps that you see in the upper left hand side of the board.

    As a DIY project, if I had to recreate this nowadays, I'd probably opt for building it in sections to make testing and troubleshooting easier, and then joining things together. Just this restoration project took several months of spare evening and weekend time to get it up and running again.

    None of the circuit elements are particularly complicated, especially if tackled separately, and certainly the power supply could be greatly simplified, given modern advances. The original layout on mine is huge, and could probably be breadboarded to about a quarter of the original size, maybe even less by a dedicated builder with time and patience.

    ReplyDelete
  6. Hi there!
    Great blog! Since I’ve been on the lookout for one of these for many years and since I’m an electronics guy myself I took the liberty to use your pictures to make a drawing of the PCB layout. It took quite a while and a bit into the project I had to face that I wasn’t going to be able to figure out all obscured component values even though I have a copy of the same schematic that you’ve got. Your early version differs quite a bit from the final schematic in the input section, the Dome filter implementation, the PSU and in the CV control section.
    Now the question. Would you Please consider opening up your unit and using my provided component layout with designations filling in the unknown component values? I know it is quite a lot to ask for but you don’t need to detach the PCB, just maybe use a loupe and a small mirror on some components. In the end you’ll get a complete schematic and PCB layout for your 20/20. I think you’ll also make many nerds like me happy.

    Here’s a link to my Component Layout:
    https://drive.google.com/file/d/1a4S34qQCbXpMBu77jc91ItbLNmLGqmC9/view?usp=sharing

    I’ve marked the unknown/uncertain component values in red. Of course the most critical component values are for the styroflex capacitors and 1% dale resitors in the Dome filter.

    Eagerly awaiting your reply…

    ReplyDelete
    Replies
    1. I'd be happy to help!
      Give me a bit, and I'll compile a list of the values for your layout.
      Nice job, by the way...;)

      Delete
  7. Ahh!! That is great news! You make me very happy! THANKS !!!

    ReplyDelete
  8. Here's an initial list of values for you to enter in your document. After you've updated it, let me know and I'll recheck it for any mistakes or missed values.

    C4 100n
    C5 8n2
    C6 3n3
    C7 16n
    C8 1n3
    C9, C18 C21, C22, C35. 47uf
    C11 100n
    C12 620pf
    C13 13n
    C14 6n8
    C15 270pf
    C19
    C38, C40 10uf
    D1, D2. 1N751
    Q1, Q2 2N3904
    Q12 2N2219A
    Q4, Q5, Q6, Q8, Q10 Fets 2N3820
    R2 2K21
    R3 18K2
    R4 18K
    R5 4K75
    R8 84K5
    R9 4K64
    R10 287K
    R11 3K32
    R12 23K7
    R13 69K8
    R22 6K8
    R26 73K2
    R27 383R
    R28 5K11
    R30 8K2
    R34 332K
    R38 221K
    R39 332K
    R40 47K5
    R42 332K
    R45,46,47 56K
    R48,51 18K
    R57,60,61,68 330R
    VR 7, 8, 9. 10K

    ReplyDelete
  9. Many Thanks!! I have now gone over your preliminary list and realized that I have used the same designations twice on resistors R22 – R40 on the layout, and used the R24 designation three times!. My ignorance. I apologise.

    When using your preliminary list I have tried to make qualified guesses on which component in the doubled R22 – 40 range that you relate to in your list. These assumption are based on what I’m able to glean from your pictures and from functionality based on my preliminary schematic, I’ve tried to keep the designations unchanged for the components that I think you relate to in your list and given new designations for the doublettes (R83 – 102)

    All updated information has been added to a revised layout:

    https://drive.google.com/file/d/1v1go98TDSc8syp5G0-LTTvJq9qQ8AMN8/view?usp=sharing


    Key to component values:
    Green = assumed correct from the start (adequate visability in the pictures)
    Blue = values from your preliminary list.
    Red = Still unknown/uncertain values

    If you’re having trouble reading the characters on R24 and R25 you can just measure them with a DC ohm meter since the are isolated by capacitors that will not shunt the DC ohm meter

    R6, R7, R84 and R85 should be identical so if you only manage to read one of them that would be sufficient.

    Of course C16 is vital to the function of the Dome filter.

    Again Many Thanks for your generous help. Looking forward to cracking this nut!

    ReplyDelete
  10. Here's the next round of values from your notes;
    C19 100N
    C20 22uf
    C27, C30 120NF
    IC2 CC20066 Raytheon
    Q3 2N3904
    R6, R7, R84, R85 681R
    R16, R32, R96 10K
    R24 16K9
    R25 549R
    R29 18K2
    R41 220K
    R50 10R
    R56 68K
    R59 22K
    R71, R72 100R
    R73 5R1
    R74 1R
    R83 6K8
    R87, R89, R90, R91 2K2
    R88 20K
    Q7 and Q9 were moved from the board to the small add-on Perfboard I made that is mounted on the front panel, along with the connected ranging and scaling circuits. The schematic as shown is correct for the connections.
    IC2 is a mystery...I couldn't fins a spec sheet, but it looks to be a nicer dual opamp of some sort with a standard pinout.
    Hope this helps!

    ReplyDelete
    Replies
    1. Thanks !!!

      I’ve added the latest info to the component layout v3: https://drive.google.com/file/d/1bhZBdIMSKHY-2vdB6QOrEtMzMxM24CMT/view?usp=sharing

      I’ve also made a preliminary unverified schematic: https://drive.google.com/file/d/10LGctAAtw4TCAF8Kb_81m57pbnGfOXWn/view?usp=sharing

      As you can see I’m still missing the value of critical component C16 in the Bode filter. If it’s not possible to read the value of C16 I think it would be possible for me to approximate the value and using an oscilloscope in X-Y mode to adjust C16 for as much of a perfect circle as possible over the audio range (meaning more or less 90 degree constant phase shift between the two signals driving the 1495 quatrature multipliers). It may be a simple task to calculate the value for someone familiar with the math (Hilbert transform) but that person sure ain’t me…

      Still missing is also 6 other caps and 1 resistor but these are only important for the sake of completion so Please feel free to disregard these. You’ve already done so much out of the ordinary to help out and I’m very greatful.

      I suppose the reason why they left out the Bode filter C and R values in the factory schematic could be either that the components needed to be factory matched to one another or the fact that the design and selection of components in a Bode filter was quite a big investment in time and effort in the 1970’s. Today there is completed designs widely available on the internet

      As you can see in the schematic it seems that your +/- 15V regulated power supply for the circuitry is supplied from ahead of the regulator. Considering that the rectifier diodes was hooked up the wrong way from the factory it makes me think that this ”fault” has been there all along. Correcting this would give you a much better regulated power supply. Then again given that the power supply changed for the later version it may be that the first design was simply not working? Maybe the regulator chip current limiting activated under certain conditions. The low ohm asymetrical value current sensing resistors R73 and R74 may point to that theory.

      Your version does not have the complete CV control circuitry on the main PCB so I guess either it used a much simplified CV circuitry driven directly from the Freq Shift potentiometer and Ext Control jack or there were an auxillary PCB used?. Q7 and Q9 is most likely some common PNP like 2N3905/06. Hopefully I’ll get some time in the future to make some practical experiments.

      Also notice the diffences in the VCO topology with 2 trimmers and 2 FETs

      When I drew the schematic there were two things that sprung to mind as a bit odd. In the later version factory schematic equivalent resitors of R57, R60, R61 and R68 is 330R, not 33R. Also equivalent caps of C9, C18, C21, C22 and C35 are 4u7, not 47u. But then the factory schematic strikes me as not completely trustworthy.

      R55, R65 and C10 are located on the solder side of the PCB. Looks like 27k, 27k and 33u from the one picture of the PCB solder side. 27k matches the factory schematic. C10 is an add-on to filter the negative supply for IC1 (maybe due to the bypassed voltage regulator previously mentioned).

      PCB track (as ”seen” thru the PCB) and component layout for confirmation of the circuit topology shown in my schematic: https://drive.google.com/file/d/1IGmlwtwHwMtrcQ2ezPnIVtIMPWBd00VC/view?usp=sharing

      English is not my native language so any strange phrasing in my messages is just due to this fact. Over here in Sweden the weather is fabuolus! Last weekend I took my first outside swim this year. It helped that I had a warm sauna waiting.

      Once again, Many Thanks !!!

      Delete
  11. C16 is 27NF...how did I keep missing that??

    Your english is quite good, Mr Grizzly tone!

    I moved the CV control circuitry to the little front panel board that I added, leaving just the 15k resistors on the main board. The circuitry is the same as shown in the schematic, with a trimmer to adjust the overall scaling.
    Thanks for your efforts in creating better documentation for this device. I appreciate it, and I'm sure there are others that will as well!

    ReplyDelete
  12. Great! Here's the updated schematic:
    https://drive.google.com/file/d/1U1nbDaa6HY_dq7c1tgQ-aYTMsgx6KJeI/view?usp=sharing
    Yes I hope this will be of use to others too. Please feel free to post my schematic on your blog.
    Thanks for your encouraging words and many Thanks for your generous cooperation!

    ReplyDelete
  13. This album by John McLaughlin’s Mahavishnu Orchestra uses the 360 System FS extensively. Around 1974.

    https://youtu.be/p3_VYkUKNYo

    ReplyDelete
  14. Ivon, thank you for the link to this wonderful performance. I love the raw, spacey intensity of the beginning, and John's trademark rapidfire interchanges with Walden's drumming certainly brought a smile to my face! Sorry it took a while to find this and thank you. Go Johnny, Go!!!

    ReplyDelete
  15. Hello !
    Thank you for sharing the documents. The 360 Freq.Shifter has been on my wish list for a long time and I would like to build it. I have already viewed many pictures and I am happy to have found the missing values (e.g. the DOME filter) here!
    Thanks
    Matthias (DIY-MMS)

    WE LOVE DIY :-)

    ReplyDelete