Hi all, this is my first post here, but I've been reading the forums for some time. I've recently succesfully built an EPE theremin using the information posted here.
Now I'm triying to build another theremin, this time an EW theremin. I've only built the variable and fixed pitch oscillators to make sure it worked right before going on. The main differences with the original is that I dont use the toko coils, instead I've rewound a couple variable inductors from an old radio. Also I couldn't get the antenna coils, so I wound a 40mH one using an old ferrite core. The problem is that with this configuration I get barely 15-20cm range, so I guess I'm not hitting the resonance point.
After this I tried to find the resonance point using a mesurement procedure similar to the one used when tuning tesla coils (more info here http://www.hvtesla.com/tuning.html) and started removing turns of the coil until reaching resonance around 285 kHz. The odd thing is that at this point the coil measures about 12mH, and the antena (12mm OD and 45cm lenght) is very similar in size to the one used in the instructions. Using this configuration the range is increased a bit but the notes are still very compressed near the antenna.
Also the oscilators behave strangely, fist while I'm closing my hand to the antenna the pitch increases, but close to the antenna the pitch goes down to zero and increases again. I dont know what else to do next. Dou you have any ideas? Thanks in advance.
Bob Moog had his reasons to use 4 x 10mH coil instead of a single 40mH coil. Don't try to be smarter than he was. A quick use of the search function here in the forum would have pointed you to sources where you can order the needed coils.
A big 40mH coil will have ways too much parasitic capacitance and thus have a much too low self resonant frequency. In order to obtain a decent linearity and pitch range with this circuit design, the resonant frequency of the series tank circuit formed by the linearization coils and the static capacitance of the pitch antenna vs its environment (without the player's hand) has to be around 285kHz. At the same time the resonant frequency of the linearization coil assembly and the oscillator's tank capacitance (3.3nF) has to be between 10 and 12kHz (somewhat more than the double of the desired pitch range). The tone spacing very close to the pitch antenna will always remain somewhat compressed but be best with linearization coils having a SRF around 700kHz. In this differential pair oscillator design the output amplitude depends strongly from the coil's DC resistance. The working point of the transistors has been chosen to give a maximal but undistorted amplitude with each of the 10mH coils having a DC resistance of about 30 Ohms.
You'll obtain that by applying the same modifications which Bob Moog developed during the transition from the old EM design (which had a smaller pitch range and worse linearity) to the newer Etherwave design: Using a 40cm long and 10mm thick pitch antenna and only 3 of these 10mH coils (Hammond 1535G). The oscillator's tank circuit should be formed by a 3.3nF silver Mica capacitor in parallel with 94uH (a fixed inductor and a variable inductor of 47uH each in series).
My advice to all theremin builders: Never build a circuit without deep analyse and understanding of it and only after you can confirm with your own computations that there is neither error nor potential for opimizations in the initial design. Don't imitate, invent!
"My advice to all theremin builders: Never build a circuit without deep analyse and understanding of it and only after you can confirm with your own computations that there is neither error nor potential for opimizations in the initial design. Don't imitate, invent!"
I think there is tremendous pedagogical value in just taking a function generator, a scope, some capacitors, inductors, and resistors, and just dicking aroud on a breadboard to see what happens with various simple arrangements. This is how I happened upon my most recent hybrid (analog & digital) oscillator. I do understand how it works, and there are undoubtedly optimizations that can improve its sensitivity, but that's where deep analyis and simulation come into the design effort for me. Everyone has a different approach to these things.
Ok, my approach is different... Things have to be understood and solved by modelling all aspects of the circuit in my mind (assisted sometimes by simple pencil and paper) first. Only when I think that I am able to follow the trace of each electron in my imagination, I go to computer aided circuit simulation (just for verification because I permanently mistrust myself ...and circuit simulators).
And now comes finally the moment when I most times decide to not to build the circuit because I had already my intellectual pleasure and practice seems often trivial. ;-)
dewster my man:
You have the same western ideology as I, just jump in the coral and get bucked off a few times. Start with little circuits on plug in proto-boards and burn a few things up. Some capacitors pop like firecrackers, this teaches what that neg black stripe means. My own original theremins were built only using Radio Shack parts, so the name RS Theremin. My first tool choice is my oscilloscope and second my parts substitution box, I purchased years ago at Olson Electronics. The discipline is to develop an understanding of how each basic circuit works and why the value of component was selected. A eureka moment for me years ago was understanding proper transistor bias. Op Amps have really simplified many constructions. I find electronics to be individual circuits stacked together in a harmonious relationship. I have never done any computer analysis; a slide rule can get me close enough. Anal tight measurements are rarely needed.
In theremin building parasitic capacitance, stray magnetic coupling or a poor power supply are the theremin killers. EMF from an oscillator is the most fascinating phenomenon to explore for a first time project.
To keep this on topic, those big EW antenna chokes do improve things but were "probably" selected by trial and error. I believe there is a misunderstanding in this area of theremin design. “One day I will rise from the dead to say the sun is shining! “
This was written for the newbie, I owe my schooling to Forrest Mims III.
Batou: Your Tesla Coil tuning webpage is facinating, I wondered how to do this.
"And now comes finally the moment when I most times decide to not to build the circuit because I had already my intellectual pleasure and practice seems often trivial. ;-)"
My advisor said the best projects were the ones that stopped short of actually building anything as this neatly sidesteps all the pain. :)
"A eureka moment for me years ago was understanding proper transistor bias...Anal tight measurements are rarely needed."
This was a huge step for me too, and didn't happen until I was in college. We spent lots of time iterating the non-linear (exponential) equations until the transistor bias was known to many decimals, the time could have been better spent teaching rules of thumb that almost instantly get one close enough.
"To keep this on topic, those big EW antenna chokes do improve things but were "probably" selected by trial and error."
I wouldn't be surprised at all if that were true.
"This was written for the newbie, I owe my schooling to Forrest Mims III."
I owe much of mine to Don Lancaster's excellent CMOS Cookbook. The 4000 series is where real magic happens as it can do mixed design (analog & digital), has loose voltage and low current requirements, and isn't so fast that you get plagued with layout issues. Walt Jung's Op Amp Cookbook was another, and looking at thousands of schematics in Popular Electronics and similar was another.
In the digital music realm Musical Applications of Microprocessors by Hal Chamberlin must be seen to be believed. When it comes to HDLs though I haven't found anything that really teaches you how to think and organize your thoughts. Literally hundreds of authors blabbering on about verification methodologies and none on how to actually do it. It took me years to develop my own methods, only to find that the best digital designers all pretty much do the same thing (architect as clock domains on paper, then waveforms on paper, then write HDL & simulate).
dewster said: I owe much of mine to Don Lancaster's excellent CMOS Cookbook. The 4000 series
No first was TTL Cookbook and then "freedom" from 5v with Cmos Cookbook. Then the evolution from hex loader to assembly to Basic to C+ and the discovery of MPLabs free software.
All this was fun but I feel it is fading from the ability and interest of the modern home hobbyist. This is why for me personally the analog theremin is a last connection to era gone by. I am slowing and no longer do I work on broad home projects.
All of my theremin designs are completely original without looking at the work of others. This is why some of my theories do not match mainstream and the imitators. I do have a perfectly linear pitch field right up"next" to the antenna relative to the music scale and the sound (not voice) I have is Clara Rockmore. If this is true FredM would call it Luck. LOL I even provide the perfect freeware fast, large-display, theremin pitch note indicator to make this evaluation. I hope to find a local Master Thereminist one day to demonstrate my design. They are rare and why would they have interest.
On topic: I did purchase a new EtherWave Standard in the past year to compare against and found my research is not in vain, but it is better to give EW's away as gifts. It sells because it is built solid, has the name, is available and very resale-able.
Batou: If you have an oscilloscope clip on an 18" alligator lead and lay it about 6" from your theremin antenna. In my research I can see the EMF pitch field on the scope and can determine aspects of performance similar to your Tesla page. I will say no more on that subject so I don't get scolded by you know who. (-'
dewster if you have not already, click on my logo name and email me from my webpage, your page is all f--ked up, I have Verizon Fios but keep webpages elsewhere. I'll bring a tent & sleeping bag.
"If you have an oscilloscope clip on an 18" alligator lead and lay it about 6" from your theremin antenna. In my research I can see the EMF pitch field on the scope and can determine aspects of performance similar to your Tesla page."
This is how I finally got my EW Standard pitch section optimally adjusted. Placed an aligator clip in a loop a few inches from the pitch antenna, connected to a scope probe, and adjusted for maximum amplitude on the scope. Then tuned the local oscillator for the correct audible beat behavior. It's a completely different instrument now. I don't know how people who buy the kit and don't have a scope / haven't stumbled across this technique ever get their EW right, the written instructions that come with it are fairly inadequate IMO.
Using a 40cm long and 10mm thick pitch antenna and only 3 of these 10mH coils (Hammond 1535G). The oscillator's tank circuit should be formed by a 3.3nF silver Mica capacitor in parallel with 94uH (a fixed inductor and a variable inductor of 47uH each in series).
Hi Thierry. I have a couple of questions about this.
- Is it worth paying for the Hammond 1535G coils instead of the much cheaper Miller 6306?
- I guess the mica capacitor you mention is C1, but the PDF says it should be made of polystyrene. Does it mean that in the new EWs C1 (perhaps also C5) is a mica capacitor?