Searching \ for '[OT] Wavefrom generator IC' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=wavefrom+generator
Search entire site for: 'Wavefrom generator IC'.

Exact match. Not showing close matches.
PICList Thread
'[OT] Wavefrom generator IC'
2000\03\29@182235 by Ben Langhals

flavicon
face
I'm working on an application where I need to generate a sine wave and
sweep its frequency.  I want to use a pic to control the process of the
frequency sweep (most of the chips I've seen take an analog control voltage
to adjust the frequency).  The range of frequencies I'm interested in is
from 1 to about 35 kHz.  I want a low distortion wave (no more than .5%,
esp in the lower part of the range) Two of the best contenders I've come up
with are the exar 2206 and max 038 that can be found at


http://www.exar.com/products/xr2206.html

http://dbserv.maxim-ic.com/quick_view2.cfm?pdf_num=1257

Anyway, I was just getting a feel for if anyone has used these guys in a
similar application, or knew of a better option.  Whats easiest to
interface with a PIC?

Ben Langhals

2000\03\29@201932 by Brian Aase

flavicon
face
This isn't really a reply, but I am interested in using DDS
chips for this purpose rather than analog waveform
generator IC's such as Ben has mentioned.

The ones from Harris and Analog Devices seem to be
intended more for RF signals rather than audio.  Plus
they are rather expensive.  Someone (Micrel?) makes
an audio-frequency DDS chip, but its distortion spec is
pretty awful.  If anyone can recommend  a better IC for
audio-frequency use, I'd like to hear about it either on the
list or by email if you prefer.

Brian Aase


{Quote hidden}

2000\03\29@203352 by Brandon, Tom

flavicon
picon face
I too am looking for a similar IC so please count me in on any offlist
emails. Looking for audible range (20Hz-20kHz odd) sine, square (w\ PWM),
saw (up\down), triangle. Looking at 0.01-0.001 Hz accuracy across the whole
range ideally.

I had also considered deigning my own with a Scenix. The sine wold be a bit
of a trick to get smooth. A 50ms cycle (20Hz) is pretty long. But with a
nice sine table it'd be possible.

As part of this I'm looking for a cheap, high accuracy, multichannel
(looking at up to 32 oscilators) DAC system. Best I've seen so far is the
Analog Devices Octal Mux'ed SHAs. Except there nonlinearity only allows
10-12bit and I ideally want 16. AD also have a 16bit rated SHA but it's a
single package. Havn't even looked at pricing yet.

Tom.
{Original Message removed}

2000\03\29@220425 by Dan Michaels

flavicon
face
Ben Langhals wrote:
>
>I'm working on an application where I need to generate a sine wave and
>sweep its frequency.  I want to use a pic to control the process of the
>frequency sweep (most of the chips I've seen take an analog control voltage
>to adjust the frequency).  The range of frequencies I'm interested in is
>from 1 to about 35 kHz.  I want a low distortion wave (no more than .5%,
>esp in the lower part of the range) Two of the best contenders I've come up
>with are the exar 2206 and max 038 that can be found at
>

I looked at both, and ended up using the 2206. Seemed to me to be
easier to interface to control circuitry than the 8038. Distortion
is rated as 0.5% nominal. And also lots cheaper than digital
synthesis chips. [plus it's analog - so it's certainly more
difficult to get it to work right].

You can control the frequency using a PIC's PWM output feeding into an
NPN inverter, with 40K to base to 3.3K to gnd, in parallel with a cap
for filtering. 390 ohm emitter to gnd. Collector to pin 7 of 2206,
and 5-10M from collector to gnd.

You can actually get up to 500:1 frequency sweep with this kind of ckt.
You have to tradeoff the size of the cap across the 3.3K base resistor.
Too small, the frequency wobbles, too big, the sweep is slow. 3.3K
and 10uF gives 33 msec time constant, and the total sweep time would
be somewhat less than that.

BTW, power to the 2206 was +/-6v, and the NPN inverter ckt was
referenced to -6v. If you use 0-10v, say, on the 2206, you would
have to fiddle with the values I gave.

Now for the fun part. Take the squarewave out of the 2206, measure
its frequency with the PIC, and close the loop back to the PWM to
stabilize the frequency. [the problem, and funnest part, with analog
is drift]. For frequency-setting, use a good cap, like COG, 30 PPM.

Tie a small cap, .003 uf or so, directly on the chip output (600
ohms Rout) to filter out the little nibs on the top/bottom of
the sinewave. Total price, $3 and change.

best regards,
- Dan Michaels
Oricom Technologies
http://www.sni.net/~oricom
==========================

2000\03\29@223012 by l.allen

picon face
Tom Wrote
> I too am looking for a similar IC so please count me in on any offlist
> emails. Looking for audible range (20Hz-20kHz odd) sine, square (w\ PWM),
> saw (up\down), triangle. Looking at 0.01-0.001 Hz accuracy across the whole
> range ideally.
>
> I had also considered deigning my own with a Scenix. The sine wold be a bit
> of a trick to get smooth. A 50ms cycle (20Hz) is pretty long. But with a
> nice sine table it'd be possible.
>
I have used a number of DDS applications where very low
distortion wasnt as critical as frequency accuracy and
agility.

However, I am probably going to have a go in the the near
future at seeing how low I can get the distortion using a
tracking lowpass filter after a DDS generator, since I have
had some very pleasing results using tracking filters and
sampled waves before. Should work well at the lower
frequencies.

Anyone done this?
_____________________________

Lance Allen
Technical Officer
Uni of Auckland
Psych Dept
New Zealand

http://www.psych.auckland.ac.nz

_____________________________

2000\03\29@233605 by rottosen

flavicon
face
I put together a couple of versions of DDS using the Scenix.
For details on this see:

http://www.rhoent.com/examples.htm

There is enough information there to build it yourself to play with
:-)

-- Rich


"Brandon, Tom" wrote:
{Quote hidden}

> {Original Message removed}

2000\03\29@235708 by rottosen

flavicon
face
I have played with several of the signal generator chips: ICL8038,
XR-2206 and the MAX038.

I think all of them are going to be marginal generating less than 0.5%
distortion even with some tweaking.

For instance the XR-2206 is specified as having 0.4% distortion
TYPICAL with adjustment and 1.5% maximum when adjusted!

The ICL8038 is even worse 0.8% typical when adjusted. (There is no
maximum spec.)

The MAX038 is also only specified for typical distortion. Its spec.
is 0.75% adjusted.

I think that the MAX038 is the easiest to work with although I think it
is also much more expensive than the other two.



Ben Langhals wrote:
{Quote hidden}

2000\03\30@060143 by Tom Handley
picon face
  Ben, take a look at Micro Linear's ML2036 Programmable Sinewave
Generator. I've been looking into this as part of my PIC-based DAQ system.
I'm currently testing the ML2036, MAX038, and looking at Analog Devices' DDS
family.

  For your application, the ML2036 seems ideal. All you need is a crystal
(literally, no capacitors). It requires +/- 5V and has a 3-wire SPI-style
interface. You clock-in 16 Bits, LSB first, and set a Latch Enable line to
set the frequency from 1 to 50KHz (I tested it to 65,535KHz). By using a
8.388608MHz crystal (Digi-Key PN: SE3415 / Epson CA-301 family), you get a
direct one-to-one program data to frequency setting with a 1Hz resolution.
The chip provides Osc/2 and Osc/8 outputs which may be of use if you want
to clock a PIC at 4.194304MHz.

  There is a voltage reference input. Tying this to +5V gives a maximum
output swing of +/- 2.5Vp-p. You can reduce this via an external reference.
Harmonic distortion is -45db max. The device comes in a 14-pin DIP or 16-pin
SOIC package. For more info:

     http://www.microlinear.com/

  - Tom

At 06:19 PM 3/29/00 -0500, Ben Langhals wrote:
{Quote hidden}

------------------------------------------------------------------------
Tom Handley
New Age Communications
Since '75 before "New Age" and no one around here is waiting for UFOs ;-)

2000\03\30@084249 by Mitch Stein

picon face
HI,

Try looking at Analog Devices AD9831 or AD9832 DDS chips they both have
available demo boards, and I am using the demo boards down to 10 KHz.

Mitch

spam_OUTmitch.steinTakeThisOuTspamphilips.com

More... (looser matching)
- Last day of these posts
- In 2000 , 2001 only
- Today
- New search...