DIY Signal Generator
- Mike H
- Amstrad Tower of Power
- Posts: 20157
- Joined: Sat Oct 04, 2008 5:38 pm
- Location: The Fens
- Contact:
#1 DIY Signal Generator
Due to the enthusiasm re 2nd hand oscilloscopes, it has subsequently also been revealed that some folks are without a usable sig gen. There may be 2nd hand ones about you could pick up, but if not, I have (on and off) been applying my bwain to the pickle of build-yer-own. Cos it can be a bit of a pickle....
Traced the circuit of my SG-2601A the other day but TBH would be very complex to build; all discrete trannies (actually it's basically a low power p-p amp but with a quite unusual FET input arrangement), plus, would be very difficult to match the thermistor used with anything "off the shelf". Plus, it doesn't simulate well in LTspice, which doesn't bode well, unless there's something I've overlooked....
OTOH there is a LTspice "educational" example for a wien null oscillator based around an op-amp. Also, I discovered that Rapid stock tiny weeny little eensy weensy light bulbs that they describe as "grain of rice lamp" (literally, the size of a grain of rice), turns out this might be perfect as the low power, temperature dependent, resistance compensation device.
This simulates very well and would be easy to make, i.e. not a lot of components. It would use the NE5532 audio quality op-amp IC, coupled with a class A emitter follower buffer for low impedance output.
Diagrams later....
Traced the circuit of my SG-2601A the other day but TBH would be very complex to build; all discrete trannies (actually it's basically a low power p-p amp but with a quite unusual FET input arrangement), plus, would be very difficult to match the thermistor used with anything "off the shelf". Plus, it doesn't simulate well in LTspice, which doesn't bode well, unless there's something I've overlooked....
OTOH there is a LTspice "educational" example for a wien null oscillator based around an op-amp. Also, I discovered that Rapid stock tiny weeny little eensy weensy light bulbs that they describe as "grain of rice lamp" (literally, the size of a grain of rice), turns out this might be perfect as the low power, temperature dependent, resistance compensation device.
This simulates very well and would be easy to make, i.e. not a lot of components. It would use the NE5532 audio quality op-amp IC, coupled with a class A emitter follower buffer for low impedance output.
Diagrams later....
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
- shane
- Social outcast
- Posts: 3403
- Joined: Sun Sep 16, 2007 12:09 pm
- Location: Kept in a cool dry place.
#3
At the risk of spoiling the fun, you could try something like this:
http://3d2f.com/tags/test/tone/wav/files/
http://3d2f.com/tags/test/tone/wav/files/
The world looks so different after learning science. For example, trees are made of air, primarily. When they are burned, they go back to air, and in their flaming heat is released the flaming heat of the Sun which was bound in to convert air into tree.
- Mike H
- Amstrad Tower of Power
- Posts: 20157
- Joined: Sat Oct 04, 2008 5:38 pm
- Location: The Fens
- Contact:
#4
Oh yeah had a look at some of those before, weren't too impressed TBH and they all seem to what money before you get anything that's really useful.
I can do virtually the same with Cool Edit-Pro, except, nothing above 19 kHz, exceeds computer's audio bandwidth.
Besides spinning a real knob is amazingly quicker and easier
I can do virtually the same with Cool Edit-Pro, except, nothing above 19 kHz, exceeds computer's audio bandwidth.
Besides spinning a real knob is amazingly quicker and easier
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
- Mike H
- Amstrad Tower of Power
- Posts: 20157
- Joined: Sat Oct 04, 2008 5:38 pm
- Location: The Fens
- Contact:
#5
OK well here's something ~ I balls'd up the attenuator bit so have to edit it, and haven't done the squarer section yet, but here's the oscillator section.
Not sure about the 22k pot TBH because another circuit I've got uses a 10k pot with the same capacitor values for 10 Hz to 1.5 MHz, but then again, a reduction drive would be desirable otherwise very awkward getting exact adjustment. Or, Rapid still do 10 turn potentiometers:
http://www.rapidonline.com/Electronic-C ... ters/29588
Otherwise, a vernier dial reduction drive could be used but where rotation is limited to 180°, hence if 22k pot you are at least guaranteed to get >half the total value.
You can get a new vernier dial from Mouser:
http://uk.mouser.com/ProductDetail/Eagl ... GieA%3d%3d
For a description of the principles see here:
http://en.wikipedia.org/wiki/Wien_bridge_oscillator
Including use of a bulb for the stabilising element!
I designed this one to output 15V peak (because my own sig gen does); gain is set slightly greater than X3 by R5 & R6, but at 15V the lamp is taking 1mA and the idea is it heats up until its filament >=500 Ohms. At which point the ratio of R5 to R6 and the bulb is reduced to 2:1 (gain of 3).
If o/p Voltage increases then so does the filament temperature and it's resistance, so gain is reduced to <X3. Thus the o/p is stabilised at 15V peak. (In theory.... )
The o/p is class A using an emitter follower with a CCS, and is supposed to maintain full Voltage into a 600 Ohms load. The controlling element is a high quality audio op-amp IC.
To get the full 30V peak-peak Voltage swing without clipping needs a 36V supply, which is max. for the NE5532.
The NE5532 could do it all by itself, but there is likely a bandwidth limitation problem due to the max. Voltage swing (for this device) into a load like 600 Ohms, hence the addition of an o/p stage.
Edit ~ circuit changed!
Not sure about the 22k pot TBH because another circuit I've got uses a 10k pot with the same capacitor values for 10 Hz to 1.5 MHz, but then again, a reduction drive would be desirable otherwise very awkward getting exact adjustment. Or, Rapid still do 10 turn potentiometers:
http://www.rapidonline.com/Electronic-C ... ters/29588
Otherwise, a vernier dial reduction drive could be used but where rotation is limited to 180°, hence if 22k pot you are at least guaranteed to get >half the total value.
You can get a new vernier dial from Mouser:
http://uk.mouser.com/ProductDetail/Eagl ... GieA%3d%3d
For a description of the principles see here:
http://en.wikipedia.org/wiki/Wien_bridge_oscillator
Including use of a bulb for the stabilising element!
I designed this one to output 15V peak (because my own sig gen does); gain is set slightly greater than X3 by R5 & R6, but at 15V the lamp is taking 1mA and the idea is it heats up until its filament >=500 Ohms. At which point the ratio of R5 to R6 and the bulb is reduced to 2:1 (gain of 3).
If o/p Voltage increases then so does the filament temperature and it's resistance, so gain is reduced to <X3. Thus the o/p is stabilised at 15V peak. (In theory.... )
The o/p is class A using an emitter follower with a CCS, and is supposed to maintain full Voltage into a 600 Ohms load. The controlling element is a high quality audio op-amp IC.
To get the full 30V peak-peak Voltage swing without clipping needs a 36V supply, which is max. for the NE5532.
The NE5532 could do it all by itself, but there is likely a bandwidth limitation problem due to the max. Voltage swing (for this device) into a load like 600 Ohms, hence the addition of an o/p stage.
Edit ~ circuit changed!
Last edited by Mike H on Mon Sep 13, 2010 1:01 pm, edited 1 time in total.
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
- Mike H
- Amstrad Tower of Power
- Posts: 20157
- Joined: Sat Oct 04, 2008 5:38 pm
- Location: The Fens
- Contact:
#6
Another balls up, my sig gen is not 15V after all, it's 10V
Memory is not what it was
Memory is not what it was
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
#7
Why not just use the other half of the op amp as an output buffer, I would think it would provide enough drive for most uses?
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
- andrew Ivimey
- Social Sevices have been notified
- Posts: 8307
- Joined: Mon Jun 11, 2007 8:33 am
- Location: Bedford
#10
I like 8038s; voltage controllable but they are about £8 each if I remember.
- Mike H
- Amstrad Tower of Power
- Posts: 20157
- Joined: Sat Oct 04, 2008 5:38 pm
- Location: The Fens
- Contact:
#11
Thanks, however I particularly wanted to do it the "old fashioned way" as all you can get nowadays (and for some time past) are function generators, and I have an irrational dislike of function generators, if probably unfounded....
Back shortly ....
TBH I think that's probably the simplest option really specially as there's 2 op-amps on the same chip, I was mainly concerned about increased distortion doing that but probably not much in it in the long runNick wrote:Why not just use the other half of the op amp as an output buffer, I would think it would provide enough drive for most uses?
Back shortly ....
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
#12
In the case of that chip, I would say the 1% THD would be a strong reason why not. Depends on what you want to do with the signal I guess.Neal wrote:Why not use a signal generator chip
http://courses.cit.cornell.edu/bionb440 ... r8038a.pdf
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
- pre65
- Amstrad Tower of Power
- Posts: 21373
- Joined: Wed Aug 22, 2007 11:13 pm
- Location: North Essex/Suffolk border.
#13
So, what is the difference between a "function generator" and a "signal generator" ?
The only thing necessary for the triumph of evil is for good men to do nothing.
Edmund Burke
G-Popz THE easy listening connoisseur. (Philip)
Edmund Burke
G-Popz THE easy listening connoisseur. (Philip)
- Dave the bass
- Amstrad Tower of Power
- Posts: 12273
- Joined: Tue May 22, 2007 4:36 pm
- Location: NW Kent, Darn Sarf innit.
#15
It's Sine and square that we need mostly though isn't it WRT amp testing?
DTB
DTB
"The fat bourgeois and his doppelganger"