cheap battery-operated sine generator?

to

on

My version of the MiniDDS should do what you want, with an amplifier on the output:

Leon

--
High frequency end?

JF

Precision? Stability? THD?

--
Paul Hovnanian	paul@hovnanian.com
-----------------------------------------------------------------------
Have gnu, will travel.

Look into adjustable phase shift oscillators.

There's the ancient XR2206

Down to 0.01Hz Split supply +/-5V up +/-13V See page 10 about DC offset control.

But dunno a distributor that has it..

D from BC British Columbia Canada

"starfire"

** See:

If you change the two 68nF caps to 3.3 uF, the lowest frequency will then be below 1 Hz while increasing R6 will allow more output level - long as the battery voltage is sufficient.

The circuit uses a common quad op-amp ( TL064) and draws only 1.7 mA with a

9 volt supply.

Injecting a small current into the negative input of the final op-amp will provide DC offsets.

...... Phil

High frequency end should only have to go to about 10Hz or 20Hz or so. It is really needed for fairly low frequencies (< ~ 5Hz).

Dave

It doesn't have to be very precise or stable. Maybe +/- 3% - 5%. THD <

5%...

It just has to be a test-aid quality sine source. It will not be used for calibrated measurements.

Dave

Thanks.

I had forgotten about the XR2206. I thought I still had one in my junk drawer but apparently I got rid of it a while back. I do have an ICL8038 but according to the spec sheet, it looks like it needs +/- 12VDC or so. I wanted to reduce the battery size a little...

Dave

Thanks very much!

I will check it out.

Dave

My version of the MiniDDS should do what you want, with an amplifier on the output:

Leon

Thanks very much.

I will check it out.

Dave

--almighty snip--

The 8038 data sheet

says 'minimum single supply +10 V'. I've had no difficulty running it off 12 V and it will work on 9 V, but the point of making reference to the 8038 is to illustrate how simple it would be to adapt the circuit provided in the data sheet to run on any arbitrary supply voltage in excess of six or more times Vbe, that is 3.6 V. I've used it as a voltage-controlled low-frequency oscillator in a number of analogue attempts at simulating the sound of a Leslie speaker (e.g. the 147) - how can it be that no-one has actually managed to do that yet?

Chris

e k 038 a s

What do you mean, to simulate a Leslie cabinet? With a single speaker it will be impossible IMHO. The impression of the doppler modulation together with the wall reflections need at least 4 speakers and even then nothing beats the mechanical movement. Ban Apricale, Italy

What do you mean, to simulate a Leslie cabinet? With a single speaker it will be impossible IMHO. The impression of the doppler modulation together with the wall reflections need at least 4 speakers and even then nothing beats the mechanical movement. Ban Apricale, Italy

I mean exactly that. If I listen with one ear to the sound my Leslie 147 makes, it has a strong sonic 'fingerprint' - a sound that is practically unique, being nothing like the sound made by other arrangements of speaker(s) in cabinet(s). Consequently the thing has a strange transfer function and it must be possible to simulate this in electronics, but no-one has done it yet. So far, everyone that's tried (e.g. in guitar and keyboard foot pedals like the Boss RT20, Korg G4, H&K Rotosphere, Line 6 MM4 and Rotomachine, etc.) has missed out something very important but I don't know what it is - nor do they, so it would seem.

Try listening, with one ear, to an electric guitar played through one of the things and then tell me if your 'humble' opinion has been adjusted at all! Try listening to the lead guitar in Badge by Cream, The Idol by Procol Harum, Wishing Well by Free, The Air That I Breathe by The Hollies ... the potential list is huge.

Or, to put it in your own terms - why is it that 'nothing beats the mechanical movement'? - what's the physical explanation for this difference?

Chris UK

PS: dynamic convolution will allow accurate simulation by blending through a cycle of measured impulse responses, but the processing power to do this in real time isn't quite yet available in OTC PCs.

The Leslie response is not a transfer function, because the Doppler shifting due to the horn motion is not a linear, time invariant operation, and hence can't be represented as a convolution.

Cheers,

Phil Hobbs

Acoustic effects. Visual the standing waves inside the box with one of them twirling around, stirring up the air, getting in the way of its own reflections, etc., etc... [...google] And they only spin the treble horn:

Good Luck! Rich

Has anyone suggested a small micro? (I haven't really bothered following the thread, until I spotted 8038. It's too bad about the supply; but I guess these days with $.95 micros (+dev. system, +learning s/w), it's not worth making a specialized chip like that.

Good Luck! Rich

the

l!

he

nce?

gh a

in

Excellent argument, or see it this way: the loudspeaker radiates a different signal into each direction, not only is the transfer function (frequency response) different, which would be possible to simulate, but also the frequency shift. So the reflections from walls/ objects close to the source have another pitch then the direct signal and all this continously changing with the angular movement. We are not even able to electronically create the sound of a Hammond-Organ or a tube-amp, let alone this complex Leslie stuff. In the 70s I've been working with Rock-groups and in a recording studio and we tried in a couple of ways to eliminate the bulky Leslies, but to no avail. Ciao Ban Apricale, IT

I've built a very simple sine generator some years ago:

The output has to be amplified for +/-6V. You can adjust the frequency with R1, if you don't need an exact frequency. Otherwise Leon's MiniDDS would be better.

--
Frank Buss, fb@frank-buss.de
http://www.frank-buss.de, http://www.it4-systems.de