Building EM Theremin / rebuilding EtherWave circuit

Hi,

just new to Theremins and ThereminWorld :)

Recently I (electronic technician) discussed with a friend (musician) about the ability of a electronic technician to make music instruments.

So I told him about the Theremin. On that point, I only knewed it from hearsay... He was interested on it, and I tried to get more information for him. The more information I collect the more I got interested in it myself.

I saw the video of Jon Bernhardt on YouTube, and I loved it! :)
Now, it's time to really build two Theremins, one for my friend - and one for me!

Okay, long preliminary(sorry), the short main part is, I think I could get all needed electronic parts from the local distributor, but the coils are giving me headache, at least the TOKO coils. After searching through the forum, I found out that DigiKey has had them in the past, but it seems that they are no more available.

It would be great if anyone could lead me to alternative (DigiKey) parts.
Any tipps/tricks/hints, like improvement of antenna linearity or PCB layout for building the Theremin are welcome, too :)

Regards,

Scotty

P.S. Sorry for my bad english, I'm from germany ;)

Welcome to Theremin World.

Are you near Lippstadt? There is a theremin building weekend in two day's time.

http://www.theremin.musikschule-lippstadt.de/

Hi Gordon,

Lippstadt is not behind the next hill :)
It's approximately 600km away from my location (black forrest).
Anyway, thank you for the info, I'll keep eyes open for Theremin events. Maybe there'll be a event near my location in the future.

Regards,

Scotty

In the EM theremin there are variable inductors of 100uH. These have been replaced in the EW design by a variable inductor of 47uH with a fixed 47uH inductor in series which makes 94uH. That's why some capacitor values have changed too, so that the pitch oscillators keep working around 285kHz. The suggested TOKO coils (ancient 10mm versions) are obsolete but you may use any other (5mm or 7mm) type as long as the inductance value is correct. Other parameters as Q factor or DC resistance aren't much important, the oscillator circuit is very stable. You should only take care with the antenna loading coils (5, 10, 20 mH) which should not be replaced by cheap ferrite bead suppressor coils. Farnell sells fine coils for that purpose but with additional shipping costs since they have them only in the U.S. stock.

I'm not far away from the black forest, on the other side of the Rhine river near the Vosges mountains. You may contact me by email if you want: theremin(at)tfrenkel(dot)com. I happen to understand and to speak the black forest language AND some German too ;-)

I have an unofficial theremin repair and design shop and would be glad to help you building and/or debugging, although my current activities are somewhat reduced since I have a broken and plastered leg actually.

Hi Thierry,

I appreciate your offer to help, thank you. I hope your leg doesn't hurt :(

I checked distance from my location to Colmar, it's about 120km. So if needed, we could arrange a meeting :)

Yes, black forrest language is a language by itself :)

A few questions:
1) Do you know if the schematic in the "hot rodding your EtherWave" document from Moog is up to date? There's no part list. I use the part list from the EM Theremin as starting point. As you pointed out there are differences in values of some parts. E.g. is L14 really 22µH? Or can I use a 47µH coil as used in the pitch oscillators?

2) It seems that Farnell now has some TOKO coils with 47µH available even in europe, Q factor is 40, Farnell order number is 1516081. Do you think
these coils will fulfill the requirements?

3) For fixed value inductors, do you think a coil like Farnell # 1186791 is a good choice? 47µH, Q-Factor is 70.

4) As you said, the antenna coils are only available through Farnell US stock. But Mouser stocks them. Maybe I can get these coils via mouser.

5) Can you give layout tipps? My general approach would be to group the function components, building "islands" whith a GND plane under the components. The GND planes are connected together with a track with appropriate width, following the signal tracks from one island to the next island. Or is a GND plane over the whole board better?

6) The connection from the antenna rods to the PCB could be made with simple wire? I think shielded wire to GND should be avoided, since it would disturb antenna functionality.

Regards,

Scotty

Hi Scotty,

1) The schematics and component values in the Hot rodding manual are up to date (out of the antenna coils, but this later). OK, there is no parts list, but component values are indicated in the schematics which should allow you to check against the older EM component list. NEVER mix up values from both schematics. Take the EW ones! And yes, L14 has to be 22uH so that the volume oscillator will work on a higher frequency and thus interfere less with both pitch oscillators.

2) The variable Farnell 1516081 coil will be very fine for that purpose.

3) No, don't use the Farnell 1186791 coils since they are designed for much higher frequencies and risk to have saturation problems in the EW circuit. I'd rather consider using the Farnell #516545 in the pitch oscillators and the #516533 in the volume oscillator.

4) There are lots of different mouser coils with the same inductance. They differ mostly in self- and stray capacitance (not always indicated in the data sheets) which affects the linearity and the pitch range. That's why Moog changed them during the years and uses actually 10+25+5mH on the pitch side (instead of 4x10mH) and 2.5+10+2.5mH on the volume side instead of 2.5+2.5+5+5mH). Hammond series 1535 3 Pi type coils which have very low capacitance would be a good choice.

5) Keep near the original layout which is a good compromise between short connections, stray capacitance and locking problems between the oscillators. A common ground plane will be ok.

6) Simple silver coated copper wire will do the job. No shielded cables as they add too much capacitance. But don't forget the grounded aluminum foil below the pitch antenna wire. It is needed for fine tuning of range and linearity.

If you'll have finished building it feel free to come to Colmar for debugging/tuning/fine adjustment.

Hi Thierry,

sorry for the late reply. What about your leg? I hope it's getting better.

For the antenna coils, mouser also has the JW Miller coils. So, does it matter which manufacturer, JW Miller or Hammond, I choose?

I'll make the layout on base of the original layout from the hot rodding article as you suggested. If anyone is interested, I can maybe post the board files.

You mentioned a grounded aluminium foil under the wire connecting pitch antenna with PCB. Unfortunately, neither the EM article nor the hot rodding article mention this aluminium foil (only a short hint). Can you provide a photo?

Thanks for your offer for a visit. I've just begun with schematic capture, and I assume it will take at least 2-3 weeks until I order the PCBs (delivery time approximately 1 1/2 weeks), so it will take a minimum of 4-5 weeks until I get a real piece of hardware. I'll order the parts as soon as possible to make sure that the footprints are correct.

My local electronic distributor is interested to sell my PCB layout (if it works :) ), but we're not sure if this would be allowed (copyright of circuit by moog, etc.). What do you think?

Regards,

Scotty

[i]"My local electronic distributor is interested to sell my PCB layout (if it works :) ), but we're not sure if this would be allowed (copyright of circuit by moog, etc.). What do you think?" - Scotty [/i]

Ignoring any ethical issues..

If your PCB layout is not a copy of the Moog PCB, you should have no problem regarding copyright on this..

If the circuit is essentially a Moog EW circuit, then you cannot duplicate and / or publish (as in, you cannot supply a copy of this circuit) without Moog's written permission to do so.. However, if component allocations are the same as the Moog, you could provide a link to an on-line Moog schematic.

The above is the 'technical' aspects - In practice, unless you go "above the radar" and your sales bothered Moog, I think it unlikely you will have any problems.

Where you may have real problems (again, only if you are "above the radar" is with regard to compliance with European law.. Technically, you need CE certification for any electronics you manufacture - And your distributor (particularly if they are a "real" distributor with conformance to normal standards required by distributors of electronics parts and builds) will need to be sure that anything they stock is compliant with the minimum CE and RoHS standards.. Then there is also the cursed WEEE directives, which force a supplier to pay up-front for the end-of-life disposal of their products.

Changing a PCB layout (unless you can PROVE that such changes are irrelevant) invalidates any prior certifications obtained for the product.. This is particularly true for any board where RF radiation is a deleberate function.

Getting CE certification is not too dificult - One can do self-certification based on established principles for the majority of the design.. But one does need a quick final check by an EMC test lab to confirm that radiated signals are within statutory requirements - you may be able to get this done for about 300 Euros.. But be damn sure you have it right first time.. Also, once you get a 'pass' you cannot change anything in the design which may alter the radiated RF without getting it re-certified.

Other requirements are easier to meet - The LVD (Low Voltage Directive) does not apply if you use a certified external supply.. One also does not need to worry about EM susceptability, or ESD, as failure of a Theremin is not going to put the operator or environment in any danger.

My personal advice is this..

If you are going to all the trouble of putting a Theremin kit together, do something better than what is allready on the market - Some simple changes to the EW design can give big improvements.. For example, use CA3083 Transistor arrays for the oscillators - These have a matched pair of NPN transistors and 3 other seperate transistors which are still more closely matched than 'off the shelf' 2N3904's.. one could either use two CA3083's using the matched pairs for Q1+Q2 and Q3+Q4, or you could use one CA3083 with the matched pair replacing Q1+Q2 and the other 3 transistors replacing Q3,4+5..

Just doing the above requires no design skills.. And your schematic would probably be free from any Moog copyright.. And the product would be improved.

I have a personal disdain for the idea of taking someone elses design, which is still in production, and effectively ripping it off - If one takes a design as the basis and improves upon it, that MAY be a different matter.

Hi FredM,

the PCB layout wouldn't be a copy, only the component positions should be near the "original".

The compliance testing etc. is irrelevant for me, since I only want to build a device for myself. So, if it's really needed, it's part of the distributor (I think). Is the Moog EtherWave certified?

From the ethical side, the distributor asked for my layout, because he's mainly distributing kits for hobby electronics. And many hobby electronics could not afford the pricing even for the theremin self made kit.

Regards,

Scotty

I think that the Etherwave is too complex to be distributed as a kit. Even if someone could get the board running, there are still lots of other difficulties to overcome until you have a theremin which deserves the name "Moog". That's most probably why Moog decided to sell a pre-built kit - they wanted to avoid a bad reputation. During the years I have seen lots of people who head already difficulties with much easier designs such as the EPE/SC theremin or the Theremax. The latter can be transformed into a fine playable instrument as I demonstrated some months ago on o8o8's PT03. But most theremax out there are either used only for oooooooeeeeooooo-sound effects or people aren't rather satisfied with it.
It's like building violins: Lots of people have done 100% identical copies of Stradivari or Guarneri instuments. But the result hasn't been the same.

I suggest to you to get at first your EW working and compare it then with a reference instrument (mine could serve for this). After seeing how many blood, sweat and tears had to be invested you may decide over further steps.

Please don't feel discouraged by this posting but building theremins is not at all the same as building a TOS based electronic organ. So proceed and learn step by step.