Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

AD7008 can be used for SSB modulation?

Status
Not open for further replies.

neazoi

Advanced Member level 6
Joined
Jan 5, 2008
Messages
4,119
Helped
13
Reputation
26
Reaction score
15
Trophy points
1,318
Location
Greece
Activity points
36,918
Hi, I have some old AD7008 and I wonder if they can be used as SSB modulators and how?

Also, any software that can control them from the LPT or through a micro will be greatly appreciated, so that I do not have to write my own.

Thanks
 

It's been said that is obsolete, it'll be hard to find.Another improved version is recommended
https://www.analog.com/en/products/...igital-synthesis/ad9856.html#product-overview
There are many AN, for isntance
https://www.analog.com/media/en/technical-documentation/application-notes/AN-924.pdf
Read carefully, you'll find SSB modulation technique.
A software can be requested from AD but I warn you this IC produces pretty dirty spectrum.It's not very convenient for classic SSB communication.

Note that AD ICs are too expensive..
 
  • Like
Reactions: neazoi

    neazoi

    Points: 2
    Helpful Answer Positive Rating
It's been said that is obsolete, it'll be hard to find.Another improved version is recommended
https://www.analog.com/en/products/...igital-synthesis/ad9856.html#product-overview
There are many AN, for isntance
https://www.analog.com/media/en/technical-documentation/application-notes/AN-924.pdf
Read carefully, you'll find SSB modulation technique.
A software can be requested from AD but I warn you this IC produces pretty dirty spectrum.It's not very convenient for classic SSB communication.

Note that AD ICs are too expensive..

The obsolete ad7008 costs 5$ on ebay and if you order 10 or so it can go up to $3!

Is the noise, the reason why these are usually locked to a PLL in communications receivers?
I was wondering about why is the PLL needed since a DDS is used.

I has a look at the AN you posted and there is a similar modulator to what the AD7008 uses. But I am not sure if this can produce ssb and how.
 
Last edited:

The obsolete ad7008 costs 5$ on ebay and if you order 10 or so it can go up to $3!
Good, old one cheaper one..

Is the noise, the reason why these are usually locked to a PLL in communications receivers?
Using a DDS for a PLL is different than using for transmission.DDSs are pretty dirty signal sources in term of spurious and their application field is very limited in transmission medium.
I was wondering about why is the PLL needed since a DDS is used.
Because PLL is naturally analog system and it synthesize clean signals.

I has a look at the AN you posted and there is a similar modulator to what the AD7008 uses. But I am not sure if this can produce ssb and how.
Read carefully mathematical analysis in this AN, you will understand how I Q modulators create SSB signals.
 
  • Like
Reactions: neazoi

    neazoi

    Points: 2
    Helpful Answer Positive Rating
Read carefully mathematical analysis in this AN, you will understand how I Q modulators create SSB signals.
I have tried to, but it is too complicated for me :(

I do not see any baseband input (eg audio input) pin on the chip, so that this signal can be SSB modulated, so how can the quadrature modulator be controlled?
I guess I miss many things...
 

I have tried to, but it is too complicated for me :(

I do not see any baseband input (eg audio input) pin on the chip, so that this signal can be SSB modulated, so how can the quadrature modulator be controlled?
I guess I miss many things...
Look at this ...
https://www.linear.com/product/LTC5598

It's direct I/Q Modulator.
 
  • Like
Reactions: neazoi

    neazoi

    Points: 2
    Helpful Answer Positive Rating
Is the noise, the reason why these are usually locked to a PLL in communications receivers?
I was wondering about why is the PLL needed since a DDS is used.
As you will know DDS outputs a series of values from a DAC.

The higher the frequency division ratio within the DDS, the more output points per cycle you get, and the closer to a sinewave output you get.
As you get up closer to the Nyquist limit of only two output points per cycle, the more crappy the output waveform from a DDS looks.
Its just full of spurs, harmonics and general wideband crap.
Look at the raw unfiltered output of a DDS on an oscilloscope and you will be truly horrified. Low pass filtering is mandatory, bandpass filtering much better, but not always possible.

The way to really clean it up is by phase locking a VCO to the output of a DDS.
You then get a nice sine wave with just some harmonic distortion, but all the spurs and wideband noise will be gone.

Basically the PLL far more accurately interpolates the individual output points of a DDS thus cleaning it up.

Raw DDS output is barely good enough for a receiver. For a transmitter, the multiple wideband spurious outputs are just completely unacceptable.
 
  • Like
Reactions: neazoi

    neazoi

    Points: 2
    Helpful Answer Positive Rating
As you will know DDS outputs a series of values from a DAC.

The higher the frequency division ratio within the DDS, the more output points per cycle you get, and the closer to a sinewave output you get.
As you get up closer to the Nyquist limit of only two output points per cycle, the more crappy the output waveform from a DDS looks.
Its just full of spurs, harmonics and general wideband crap.
Look at the raw unfiltered output of a DDS on an oscilloscope and you will be truly horrified. Low pass filtering is mandatory, bandpass filtering much better, but not always possible.

The way to really clean it up is by phase locking a VCO to the output of a DDS.
You then get a nice sine wave with just some harmonic distortion, but all the spurs and wideband noise will be gone.

Basically the PLL far more accurately interpolates the individual output points of a DDS thus cleaning it up.

Raw DDS output is barely good enough for a receiver. For a transmitter, the multiple wideband spurious outputs are just completely unacceptable.

Thank you very much!
I see now, why even on first DDS-based communications transceivers (90's) lock the PLL to the DDS and they do not use a DDS alone.

- - - Updated - - -

So you say that newer DDS ICs like the AD9854 excibit these spurious when operated alone?
 

Early DDS had very limited internal clocking speeds and sometimes a DAC with not too many bits resolution.

So if your input clock were only a 100 Mhz crystal, and your output frequency say 25 Mhz, you only get four output points per cycle. If your DAC has eight bits (or less) the output may look pretty ragged.
It will work a lot better at 500Khz output, where there would be 200 data points per cycle.

DDS these days are now being clocked at several GHz with some having a sixteen bit DAC.
Say 2.5 Ghz clock and 25 Mhz output, you get one hundred data points per output cycle. It will look like a fairly decent sine wave with just barely discernible faint ripple on it.

On a spectrum analyser it will look very clean with most spurs below the noise floor. DDS has come a very long way over the last couple of decades.

Many early DDS designs were forced to use a VCO to clean up the output. Today with a top end DDS you could probably get away without that, But I suppose it depends on your specification goals (and budget).
 

Early DDS had very limited internal clocking speeds and sometimes a DAC with not too many bits resolution.

So if your input clock were only a 100 Mhz crystal, and your output frequency say 25 Mhz, you only get four output points per cycle. If your DAC has eight bits (or less) the output may look pretty ragged.
It will work a lot better at 500Khz output, where there would be 200 data points per cycle.

DDS these days are now being clocked at several GHz with some having a sixteen bit DAC.
Say 2.5 Ghz clock and 25 Mhz output, you get one hundred data points per output cycle. It will look like a fairly decent sine wave with just barely discernible faint ripple on it.

On a spectrum analyser it will look very clean with most spurs below the noise floor. DDS has come a very long way over the last couple of decades.

Many early DDS designs were forced to use a VCO to clean up the output. Today with a top end DDS you could probably get away without that, But I suppose it depends on your specification goals (and budget).

Some SDRs use switching mixers (74 series ICs), that are driven by square waves instead. In fact I think that these would not care too much about the quality of the sine wave, as long as the most level of the powerful components of the square wave (main carrier) is such that can change state from off to on, in the switching mixer.
Thus, despite a DDS alone generates spurious, I tend to think that they may not be able to switch the switching mixer. However, these spurious could present at the output, just like any other signal at the input of the mixer.
Is this right?
 

If your oscillator is very noisy with multiple unwanted frequency components it will add all that noise onto the incoming signal.

The big disadvantage of switching mixers is the very high level of local oscillator drive power required for clean switching. A high power drive source needs to have an exceptionally high carrier to phase noise ratio if its not going to add a pile of noise onto the incoming signal.

Even the slightest jitter will create phase noise components several Khz either side of your mixer injection. Any phase noise within the IF bandwidth will come roaring through.

Switching mixers are still highly desirable because of the very wide dynamic range.

If your switching mixer creates a great roaring noise, its probably not the mixer, but the oscillator feeding it creating that noise.
 

On a spectrum analyser it will look very clean with most spurs below the noise floor. DDS has come a very long way over the last couple of decades.
Agreed, direct synthesis of RF has taken huge leaps in the last five years alone. And AD does seem to have to best offering by a wide margin. It's hard to beat a fully digital modulator in a single chip, especially if you like frequency and phase hopping.
 

...$5 on ebay just the chipset !?
Here you can buy with $9 the entire functional DDS module, including the TCXO. And free shipping anywhere in the world! I think on ebay free shipping is only if you buy a Ferrari :)

**broken link removed**
 

I bought a couple of the AD9851 circuit boards from e-bay to experiment with.
**broken link removed**
These are a fairly primitive DDS system compared to the latest fastest top end DDS chips now available. But it will get you started.

I put one of these small demo boards in a box with some thumb wheels and a 40 bit parallel in/serial out shift register.
Its all very simple hardware (no processor).
Millihertz to tens of MHz output, jolly good fun and quite instructive.

DDS.jpg
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top