I need to generate a signal of arbitrary shape on the AD9910, for the samples recorded in the RAM. Tell me please what mode for this should be used. Modes of direct switch and ramp up to no avail.
Have you programmed the "RAM Playback Destination" bits for AMPLITUDE?
Did you store the samples in RAM as 32-bit words with the amplitude information contained in the 14 most significant bit positions? This is important, because the RAM always expects 32-bit words as data regardless of the playback destination.
Did you program the appropriate RAM start/end address in the desired Profile register?
Are the Profile pins set to the appropriate profile?
1) I would suggest setting the RAM playback operation as "continuous recirculate" so that the device repeatedly generates the waveform. This will help the troubleshooting process.
2) Set the RAM playback rate sufficiently slow so that changes at the output are more readily visible (especially if you are using an oscilloscope to observe the waveform).
Is it possible to generate a signal of arbitrary shape on the AD9910?
The AD9910 is not intended to be an arbitrary waveform generator. Its many features, however, provide a wide range of possible waveforms. If you could be more specific about the exact type of waveform you are interested in generating, we may be able to help provide a solution.
I need to generate Gaussian signal
Would that be Gaussian in amplitude, phase, frequency, or some other combination?
Because you only need to control one of the three DDS parameters (frequency/phase/amplitude) you can generate completely arbitrary waveforms with the AD9910. There are two ways to accomplish this depending on the number of samples needed to generate the waveform.
Sample sets of 1024 or less can be stored in the internal RAM of the AD9910 and played back at any time. Sample sets in excess of 1024, however, require the use of the AD9910's parallel data port. In this case, you write the amplitude samples to the device in real time.
In both cases, you set the output frequency via a separate frequency control register. The amplitude varies according to the contents of the RAM or the data delivered to the parallel data port, depending on which method you use to control the amplitude.
So I'm using RAM, and recorded data there. But I can not control it amplitude. Since the frequency is obtained. What do you advise, is to generate a form of amplitude
Basically I do everything as you wrote. Is that the recording is 14 bits in RAM. But I will correct it. Tomorrow I will write a response. Thank you very much
KennyG, I did everything as you said, but the results are not received. And counts rechecked. Do not even know what to do
Before we begin the troubleshooting process, I need to know the device environment.
Are you using an AD9910 Evaluation Board/Software or an AD9910 chip installed on a custom board with a controller and your own control software?
I use the AD9910 Evaluation Board with the incoming with the software. Writing to carry out by using the RAM board STK500
If you're using the AD9910 evaluation board but you're programming the AD9910 externally via the pin header on the evaluation board, you must avoid bus contention from the evaluation board ICs also driving the interface pins of the AD9910. So, you must either tri-state these ICs on the evaluation board and control all appropriate inputs externally. Note, there's a couple of good discussions on the already on the DDS forum of the steps involved to program the RAM in the AD9910.
My friends, do everything you say. Generation occurs, but only until the RAM Playback Destination = 00. as soon as I record there in 10 to control the amplitude at the exit, nothing happens.
Is it possible you have OSK enabled? OSK will take priority over RAM in controlling the output amplitude if enabled. OSK Enable is bit 9 in CFR1.
no, OSK is not included.
Forgive me for my persistence, but I decided to put the results in graphic form. And then I've been head does not work. One archieve contains the results of the choice of amplitude for the control of another frequency. So he decided to bring clarity to the content of registers. Waiting for help, thanks in advance
no new posts in 2 months
Retrieving data ...