On/Off whine

Posted: 6/9/2010 4:48:02 PM
electrostatic fish

Joined: 1/20/2010

Whenever I turn on or off my etherwave standard, it makes a whining sound, something to do with the charging and discharging of capacitors. How can I get rid of this? I don't mean silence; I want to know where to put a component to squelch this effect.
Posted: 6/9/2010 5:23:37 PM

From: Colmar, France

Joined: 12/31/2007

[i]I want to know where to put a component to squelch this effect[/i]

The component is your left hand and you have to put it on the volume loop.

Other solution: Switch your Etherwave on first, hang its power cable over the volume loop and only then switch the amp on.

Or, still another variant: Make it like me. I never switch my Etherwave off but only the amplifier. So I'm sure that the Theremin is always warmed up and stable. And the 80mW standby power consumption (Etherwave muted with its power cable hanging over the volume loop) is negligible. You'll need more than 7 weeks in order to see a single kWh on your invoice.

There is IMHO no way to add a squelch circuit since the VCA needs stable -12V in order to mute the output and just those aren't available in the first moment after switching on.
Posted: 6/18/2010 11:18:23 AM

From: Escondido, CA

Joined: 2/6/2008

You can build a muting circuit using a relay easily enough.
Posted: 6/20/2010 12:51:32 AM
Art Harrison

Joined: 3/17/2010

This may help:


-Art Harrison
Posted: 6/20/2010 3:14:46 PM

From: Colmar, France

Joined: 12/31/2007

I'm not sure if this will help. Muting through the LM13700 VCA needs stable -12V. And that's just missing in the moment of switching on or off...
Posted: 6/21/2010 3:10:04 AM

From: Eastleigh, Hampshire, U.K. ................................... Fred Mundell. ................................... Electronics Engineer. (Primarily Analogue) .. CV Synths 1974-1980 .. Theremin developer 2007 to present .. soon to be Developing / Trading as WaveCrafter.com . ...................................

Joined: 12/7/2007

Below is an untried quick idea.. using a vactrol or similar part which has a LED and photo-resistive element coupled together so that as LED current increases, the photo-resistor's resistance decreases.

Shown below is a circuit with R1 (330k) charging C1 (47u) and a voltage follower (Q1 - any NPN with reasonable hfe and VCEO >30V, simulation uses BC337) driving the vactrol LED via a 15V zener (D2) and 510R resistor (R2).. Value of R2 set for max 2.4mA
(2.4mA with 24V - set to taste, absolute maximum allowed is 40mA, using VTL5C7 you get about 1k resistance for 2mA, and about 1M resistance for 0mA, - Using VTL5C6 you get about 10k resistance for 10mA, and about 100M resistance for 0mA.. Other vactrols have different charactaristics - important ones are the on resistance, you want as low as possible - and the off resistance, you want as high as possible)

As C1 charges,there is a delay until the voltage on C1 exceeds the zener voltage + LED Vf + Vbe.. This gives about 10 seconds of maximum muting (zero LED current) thereafter current through vactrol LED increases , and the vactrol resistance IN SERIES with the signal output will decrease, giving an increase in volume. assuming a 10k load, the 1k resistance of the fully on VTL5C7 will only slightly attenuate the signal.. At about 30 seconds the vactrol resistance will be about as low as it gets.
Note that the capacitors charge is exponential, while our sensitivity to loudness is logrithmic, so the sound will appear to reach nearly its maximum long before 30 seconds - I think it will be silent for 9 seconds after power up, then be heard at about 10secs, and sound close to maximum at about 15 secs... with probably +/- 20% variation due to tolerances (particularly C1).

Increasing C1 or R1 will increase the time - R1 is selected for the transistors hfe, it should not be increased, and could be reduced to say 150k for more reliable operation with lower hfe transistors.. I would increase C1 to 100uF and decrease R1 to 150k if I wished to retain the same time constant and ensure reliability.

Also, dont forget that C1 sees 24V, so make it at least 35V rated. At 2.4mA there should be no dissipation problems, but beware if you need to drive a higher current LED.


Vactrol resistance connected in series with output.

Vactrols (http://uk.farnell.com/jsp/search/browse.jsp?N=1004335+336735+336736&No=0&getResults=true&appliedparametrics=true&locale=en_UK&catalogId=&prevNValues=1004335+336589+336583+336593+336628&filtersHidden=false&appliedHidden=false&originalQueryURL=%2Fjsp%2Fsearch%2Fbrowse.jsp%3FN%3D1004335%26No%3D0%26getResults%3Dtrue%26appliedparametrics%3Dtrue%26locale%3Den_UK%26catalogId%3D%26prevNValues%3D1004335)

NSL 32 (http://uk.farnell.com/silonex/nsl-32/optocoupler-resistive-o-p/dp/3168773)

NSLR32 SR2 (http://uk.farnell.com/silonex/nsl-32sr2/optocoupler-resistive-o-p/dp/3692206)

This is probably the best cost / performance vactrol for this application:
NSLSR32 SR3 (http://uk.farnell.com/silonex/nsl-32sr3/optocoupler-resistive-o-p/dp/3692218)
Reduce R2 to 220R to use this vactrol, and reduce R1 to 120k, increase C1 as required to lengthen time (make it 100uF)

You must be logged in to post a reply. Please log in or register for a new account.