30 MHZ ad9850 DDS Signal Generator in 12$

[Kedar264]

An Precession Signal generator is very easy and affordable make using an arduino and dds synthesizer (ad9850) .

You can make decent 0 -30 MHZ frequency Signal generator only in 12$ . If you are pro over clocker then 40MHZ in same price .

See the video
  https://youtu.be/1SgcjxacGlQ

You will need frequency generator if you are a amateur radio guy or hobbyist or professional electronic guy.

 Here's  Part list you will need
1. Arduino Pro mini
2.AD9850  (DDS Synthesizer)
3.16×2 LCD Display  ( Hitachi hd 44780 )
4.Rotary Encoder


#CODE LIBRARY AND SCHEMATICS

http://www.mediafire.com/?vn3xfn956k12g

Here's  Few scope shots with 1,10,20 and 30 mhz

[moffy]

Thanks.

[Kedar264]

 

[Skimask]

And when you get it up to around 30Mhz, look at the actual output on an o'scope and see how much it really looks like shit.

[coppice]

And when you get it up to around 30Mhz, look at the actual output on an o'scope and see how much it really looks like shit.

How the signal looks at 30MHz will depend entirely on the quality of filtering used at the output of the DDS chip. The OP's schematic doesn't show any filtering.

[Rerouter]

the conventional ebay modules have a poor filter set at the wrong cutoff frequency, doesnt bother me in the slightest considering i am using it for up to 25Khz with fractional hz adjustment,

[Skimask]

Even with filtering, point is that those modules generally use a 125Mhz xtal.  30Mhz sine wave with a 125Mhz MCLK will give a guy, what, 4.x bits of DAC resolution, give or take...
Sine wave ends up looking like sticky spaghetti thrown on a wall.

[niekvs]

Even with filtering, point is that those modules generally use a 125Mhz xtal.  30Mhz sine wave with a 125Mhz MCLK will give a guy, what, 4.x bits of DAC resolution, give or take...
Sine wave ends up looking like sticky spaghetti thrown on a wall.

Can you elaborate? Why wouldn't this DDS chip be able to generate a 30MHz sine wave when fed by a 125MHz crystal? From the datasheet: "The AD9850’s circuit architecture allows the generation of output frequencies of up to one-half the reference clock frequency (or 62.5 MHz)". Even a square wave will be a perfect sine wave after proper low-pass filtering. Look at page 9 of the datasheet to see some of the aliasing that occurs, which is spaced out far enough for easy low-pass filtering.

Admittedly, the original post has a schematic without any filtering, but even that would probably show up as a perfect sine at 30MHz on most people's Rigol DS1052 limited to 50MHz

[Kleinstein]

The AD9850 has a 10 Bit DAC. So Distortion is not that bad. The principle limit of the DDS is at half the sampling frequency. With a real filter, something like 1/3 of the sampling frequency is a realistic upper limit. However this at least needs the filter components to be chosen correctly and a reasonable HF layout.

So if the filter is correctly build, a 30 MHz Signal can still look good. Still it's a good idea to check it - it may be OK, but it may also show quite some aliases. It depends on the type of module used.

[Kedar264]

Soon i will upload a video sshowing its response, but for a beginner its useful for such budget considering other similar products.

[Kedar264]

Soon i will upload a video showing its response, but you can get idea from these scope shots ,for a beginner its useful for such budget considering other similar products.

[niekvs]

Nice. I hadn't expected it to look bad. Not sure what the other guy meant with his comments about that it would look like "shit" or "spaghetti". I guess perhaps he wouldn't know any better considering he's a guy who prefers not to take things apart, but turn them on

[Kleinstein]

The waveform looks OK, but the amplitude is going down quite a lot. So it looks like the filter is setting in to low. The way the output is connected may influence the filter as well. How are the data measured ?

Without an appropriate output filter, the signal may look rather strange, possibly like a more or less random up and down close to the upper frequency limit, as one has a mixture of the frequency one wants and something like the sampling frequency +- the set frequency.

There are two thing missing to make a useful function generator: A slightly more powerful output amplifier, so you can get something like a few volts into 50 Ohm would be very useful. Also a defined attenuator to set the amplitude to lower levels if needed is commonly expected.

[sync]

Nice. I hadn't expected it to look bad. Not sure what the other guy meant with his comments about that it would look like "shit" or "spaghetti". I guess perhaps he wouldn't know any better considering he's a guy who prefers not to take things apart, but turn them on

The system (generator + cabling + scope) has a little bit more than 10MHz bandwidth. At 1MHz the peak-to-peak voltage is 1.06V. At 10MHz only 784mV. That's near the -3dB point. With such a low bandwidth (low pass filter) every waveform at 30MHz looks like a good sine wave.

I looked at one of these Ebay AD9850 modules with 100+ MHz bandwidth. At 30MHz the waveform didn't look nice anymore.

[Skimask]

Nice. I hadn't expected it to look bad. Not sure what the other guy meant with his comments about that it would look like "shit" or "spaghetti". I guess perhaps he wouldn't know any better considering he's a guy who prefers not to take things apart, but turn them on

Nice...try again.
Run those waveforms thru a spec-an and see what numbers you come up with.
30Mhz looks more of a triangle.

[Mechatrommer]

it may not run perfectly fine at >30MHz for some people, but at $12 say 10MHz BW limited, it may be the best bang/buck for most poormen approach. and the OP hasnt yet put any filtering and front end output just like other more robust and expensive DDS units. so sure there are alot to improve on this one. otoh i like the idea of push button to switch between freq increment range.

[Skimask]

Don't get me wrong.  I've got 6 of these AD9850 modules laying around in various forms, in use as variable square wave makers, hacked up as signal generators, etc.
But I also don't expect them to make decent high freq waveforms either.
Like most other things on chinabay, to say that they can do 30Mhz is quite a stretch of the imagination.

[Mechatrommer]

they maybe overspecify it, but say from testing it got 10MHz BW in reality... still alot cheaper than a standard 10MHz FG made in "not china". granted they have all built in, but some people just dont have the bill.

[sync]

The modules have much more bandwidth than 10MHz. In reality the modules I got have too much bandwidth (filter for AD9851 at ~70MHz). The 10MHz bandwidth is caused by other things.
These modules are nice. But don't expect wonders.

[niekvs]

Nice...try again.
Run those waveforms thru a spec-an and see what numbers you come up with.
30Mhz looks more of a triangle.

That probably means the filtering wasn't done properly on these modules (or by issues in the way it's measured, as mentioned above), which is also evident by the attenuation seen in the plots.

Even with filtering, point is that those modules generally use a 125Mhz xtal.  30Mhz sine wave with a 125Mhz MCLK will give a guy, what, 4.x bits of DAC resolution, give or take...

I still don't see what you're trying to say with the above statement, about 4.x of DAC resolution? The AD9850 has a 10 bit DAC. And proper filtering should remove any alias images or harmonics, resulting in a sine wave. Again, if you look at page 9 of the datasheet (http://www.analog.com/media/en/technical-documentation/data-sheets/AD9850.pdf), you'll see that the images should be low pass filtered, after which only the fundamental (sine) remains.

[Skimask]

I still don't see what you're trying to say with the above statement, about 4.x of DAC resolution? The AD9850 has a 10 bit DAC. And proper filtering should remove any alias images or harmonics, resulting in a sine wave. Again, if you look at page 9 of the datasheet (http://www.analog.com/media/en/technical-documentation/data-sheets/AD9850.pdf), you'll see that the images should be low pass filtered, after which only the fundamental (sine) remains.

Works nice...in theory...

4.x BITS of DAC resolution, call it 4 bits if the X bothers you.  (Note...I actually meant 2.x)
MCLK runs at 125Mhz.  You want 30Mhz output.  You have to add to the accumulator a number that'll swing the DAC fast enough to give you that 30Mhz output, and it has to be a large number.  As the desired output frequency gets higher, the number of 'steps' in your output drops.
Say you want a 62.5Mhz sine wave output.  Well, it ain't gonna happen.  125Mhz / 62.5Mhz = 2 steps, high and low.  1 bit sine wave = square wave.  Filter that...
Want a 31.25 sine wave output?  Ain't gonna happen again. 125Mhz / 31.25Mhz = 4 steps, high, 1/2 high, 1/2 low, and low.  2 bit sine wave?  No...
You want the full 10 bit DAC resolution, you have to divide your MCLK by 1024...125Mhz / 1024 = 122,075.3125 Hz.  Double the Hz for every bit you can stand to lose in DAC resolution.

Simple math.

As far as filtering goes...  Sure...design all the filters you want.  Try it at 30Mhz and you'll end up with garbage.  Even if you do get a decent sine wave output that's free of distortion, it'll likely only work well within a very narrow set of parameters.
Sure...go ahead and buffer the output of that filter.  Sounds good.  By the time you get done with all of that, may as well have gone and bought a proper 30Mhz sine wave generator.

[retrolefty]

Too much elitism going on here. This is a great project for a DIY beginner offering access to a very basic RF generator that so many would not be able to or wanting to make the investment needed to get started on serious RF learning.

 I so wish such tech was available when I started in ham radio in the 70s, it would have helped many of my early 'learning' RF projects. It would be cool to see several examples built by different people, and any changes or additions offered up.  I would think an external amplifier with AGC limiting would be a good candidate?

 

[sync]

An amplifier and AGC circuit would be nice. For >20MHz I only have these modules.
btw: I fixed the 30MHz issue by overclocking (200MHz).

[paulie]

Well, it ain't gonna happen.  125Mhz / 62.5Mhz = 2 steps, high and low.  1 bit sine wave = square wave.  Filter that...

I need to tell you a secret. Inside every square wave there is a 100% pure sine wave struggling to overcome social stereotypes and break out of the closet. Only 10 cents of components required to do so. The real problem is unless those numbers in your example are exact it does not result in actual square wave due to alias and other factors.

[Skimask]

Look...I'm not saying the damn things won't do 30Mhz.  I'm saying they won't do a nice, distortion free 30Mhz.  Plain and simple.  Nor will those $12 modules get anywhere near it.
Overclocking...ok, sure, maybe you can get away with it.  Maybe not.  Iffy at best.  The AD9851 is rated for 180Mhz, same thing with an extra 55Mhz of headroom on the Xtal input.  Probably a binned AD9850 anyways.  Might be able to overclock that too.  Yay!
Yes, square wave composed of sine waves...ok fine.  No kidding.  Also assumes you have a nice distortion free square wave to begin with...which you won't...which will give you nice distorted sine waves at the output...which you will.