ADL5391 multiplier problem

Hello,

I am testing my new ADL5391-Evaluationboard, and coming across with the following problem. As you can see on the figures, the multiplication output is somehow disturbed. The output is not a proper sine-function anymore. Some peaks are higher, some smaller.  It seems that the two input sin/cos-functions internally get some DC before multiplication. I use the board in single-ended mode, so I added an 56Ohm resistor for R2. Any ideas how to fix this problem?

Thanks!

Best regards,

Sam

Squaring result of sine-function  

Squaring result of rect-function

  • 0
    •  Analog Employees 
    on Jan 23, 2015 3:16 AM

    What is the GADJ pin biased to?  Is it floating?

    Did you check all the signal paths, making sure everything is electrically connected as it should?

    Also, how much current is it drawing?  It should be about 135 mA at 5V.  I would also check all the bias levels on the pins, which should be 2.5V.

  • I do not connect anything to GADJ (J3), but it is biased to 1.122V.

    Xp,Xm,Yp,Ym,Z,p,Zm are all biased to 2.49V. The signal pathes are all correct.

    Power supply at 5V gives 126mA.  I put this 5V on the TP13(Vpos), and ground on TP12 (COMM).

    Any ideas? Thanks!

    Best regard,
    Sam

  • 0
    •  Analog Employees 
    on Jan 26, 2015 11:47 PM

    Hi Sam,

    I'm wondering if what you are seeing is the fundamental bleedthrough deconstructively interfering with the 2nd harmonic (see page 11 in the datasheet for the case of squaring a sine wave).  Unfortunately, due to the part having internal DC offsets, this provides a mechanism for the fundamental to bleedthrough to the output. 

    I have not gone through the trigonometric math for the "sine * cosine" case, but I have a feeling the interference could be worse. 

    You could try a couple of things:

    1)  Bandpass filter the output at the frequency of interest (the second harmonic)

    2)  Put a DC offset on the inputs to null out the internal DC offsets that are causing the bleedthrough

    If indeed the problem is due to the bleedthough, the output filter might be the easiest experiment to confirm this.

    Hope this helps,

    Joel

  • 0
    •  Analog Employees 
    on Aug 2, 2018 2:53 PM
    This question has been assumed as answered either offline via email or with a multi-part answer. This question has now been closed out. If you have an inquiry related to this topic please post a new question in the applicable product forum.

    Thank you,
    EZ Admin