Antoher Jaycar/SC question. Sorry.

Hello all.

I've just built a jaycar/SC theremin and it actually works! Seeing as it's my first time working with PCBs or electronics i'm quite happy.
I have fiddled enough with it to competently tune both antennas but i'm having a problem with the quality of tone.

The 'sine wave' tone seems to have something else with it. Kind of like distortion. The sound is not smooth, a kind of roughness to it. I think if you were to look at it on a scope there would be small undulations within the wave itself. It is there throughout the volume range but is more noticable when louder and on a low note. Sorry, this is the best i can describe it.

I read this possible fix here:

If You have a distortion on the sound when you increase the volume just put a 1 k resistor between pin 6 of LM358 and Ground, that remove completly the buzz.

Do you think this will help or am i expecting too much from a cheap unit.

Great site by the way!

cheers

jimmy

Hi Jimmy.

I haven't tried that mod yet but i soonwill. I have the same problem as you and i think that will work. So if your confident with moding the circuit have a go.

This article is also very helpful ifyou haven't read it.

http://www.thereminworld.com/silicon_chip_theremin_modifications.html

Tatems

Actually, with this comment:

If You have a distortion on the sound when you increase the volume just put a 1 k resistor between pin 6 of LM358 and Ground, that remove completly the buzz.

I am unsure if you just solder the resistor to pin six and ground and leave the connection from pin six to the original connection or if pin six just attaches to the resitor to ground and there are no other connections.

Get that?

I tried the first way round and it didn't really help i'll have another go but if anyone can tell me that would be great!

Tatems

All

There's more than source of noise in this theremin design, but for the price, man, hard to complain. One source of noise is simply the fact that it's a single layer PCB which funamentally limits grounding. Another factor is that the regulator (7805) is being pretty heavily loaded, there is mod to address this which does help. DC biasing the negative input of the line out amplifier (pin 6 of the LM358 to ground through a 1K resistor) might help reduce the noise by reducing the gain (gain = 1 + R2/R1 where R2 is the resistance of the wire/trace from pin 6 to pin 7 and R1 is the 1K resistor you are adding to ground) but I doubt it will eliminate it entirely and if you look at the design, pin 5 (the positve input terminal) is DC biased at VCC/2 (~3V). So with the mod there's a 3V DC offset between positive and negative inputs of the amplifier.

One way to help lower the noise floor is to use shielded wires to the antenna's and speaker. Also be sure to lower the resistor value between pin 1 and pin 2 of the LM358 (the mod on this site suggests 220K) and should be coupled with upping the Re value on the 1496 to 2.3K from 1K. The MC1496 (see http://www.onsemi.com/pub/Collateral/MC1496-D.PDF) is what's known as a Gilbert Cell(named after Barrie Gilbert; an icon of analog IC design) or Balanced Modulator. Here's a nice application note (http://www.onsemi.com/pub/Collateral/AN531-D.PDF), if you care to dig into the theory of opperation. Anyway, the signal out of this amplifier is going to control the bias of the 1496 and with the feedback resistance at 1M on the LM358, the amp has too much gain and the volume circuit is a little sensitive which can cause the 1496 to overdrive the line out amplifier or distort its own output. Upping the Re value of the 1496 gain adjust resistor further reduces the gain of the 1496 (see equation 12 of the app note).

To answer a previous post, the connection from pin 6 to pin 7 must be left in tact. This amplifier is in a unity gain configuration; removing that connection would completely disconnect the output path.

To address the pure sinusoidal tone there are many mods that one could use, the mods on this site were improvements for the most part. Wave shaping is another. Look up saw toothe wave shaping and you might find some nice circuit mods.

Overall, I would have to say for the price this a nice kit. It works, is well documented, generates some cool sounds and includes a speaker and amplifier. And now that Tim's gone and done a redesign, I'm betting the Mark II fixes many of the noise issues.

Happy hacking,
Bourland