# Transfer function

Hi, i’m an Italian student.

I’m using an Oscilloscope with Basic Tx and finding a strange behaviour
sending
a costant value of 20000. I see a sinusoidal wave generated from Nco
(fine
modulator) + coarse modulator in this way :

32 MHz -> only coarse -> 670 mVpp
16 MHz -> only coarse -> 730 mVpp

0.5 MHz -> only fine -> 680 mVpp
0.1 MHz -> only fine -> 570 mVpp
After this frequency, Tension gets down:
12.5 KHz -> only fine -> 148 mVpp
1 KHz -> only fine -> 13.3 mVpp

The transfer function on the datasheet shows that with interpolation
(I’m using Interpolation value of 124 -> 31 (FPGA) * 4 (AD9862) ) it
should not
be multiplied by the sinc function, so I can’t explain the Tension drop
after
100kHz.
Thanks a lot.
Giuseppe Pilla

Hello Giuseppe!

[email protected] wrote on 2007-05-21 12:57:

Hi, i’m an Italian student.

Not far from Austria!

I’m using an Oscilloscope with Basic Tx and finding a strange behaviour sending
a costant value of 20000. I see a sinusoidal wave generated from Nco (fine
modulator) + coarse modulator in this way :

You output a constant value and get ripple?

32 MHz → only coarse → 670 mVpp
16 MHz → only coarse → 730 mVpp

0.5 MHz → only fine → 680 mVpp
0.1 MHz → only fine → 570 mVpp
After this frequency, Tension gets down:

I interpret Tension as Voltage.[1]

12.5 KHz → only fine → 148 mVpp
1 KHz → only fine → 13.3 mVpp

[…]
I can’t explain the Tension drop after
100kHz.

That one’s easy:
The Basic-daughterboards are transformer-coupled [0]. This means you
won’t get DC and low frequency (f < 1MHz) over these boards. If you need
that frequency range, the LF{RT}X-boards are the right ones for you.
Note: The Basic-Boards do not have aliasing filters, so you should be
able to receive signals from 1MHz to about 250MHz with aliasing. The
LF-Boards have a 30Mhz-aliasing filter, you get the range 0Hz < f
<30MHz.

Patrick