ad834

Hello

I want to use analog multiplier to multiply the two amplitudes and output the products, which is its theory of operation. unfortunately I do not see it working the way it should. I do not see any variation of output with the change in input amplitudes supplied using a DC supply source. i have attached the schematic of the board which was designed according to the data sheet suggestions. Can you please suggest me with any solution ?

Thank you !

  • Hello,

    Sorry to hear you're not achieving the expected results.

    Are you using a physical circuit? If so, it's always useful to probe around the circuit with a DMM. Make yourself a list of expected results by pin number starting with pin one.

    Your schematic just shows the device with terminating resistors but there must be some sort of connections to the device. If this is the case, what type of I/O are you using?

    Are you using a SPICE model for your testing? if so, trouble shooting can be similar - most SPICE engines permit pin-voltages to be displayed. This can help if you check the pin voltages against expected conditions.

    Best of luck,

    js

  • Thank you for the reply. I am measuring it on the board. The I/O interface is with SMA connectors as I have square wave signals at the input in my final application.

    I will try it on SPICE, but I am looking for someone who has already used AD834 for their experiments. 

  • OK, good luck. There are certainly a lot of these out there.

  • Hello jstaley

    (Please find the attachment for the schematic)

    Thank you. But no one replies to this thread. Anyway, I have used the Spice models and simulated the whole circuit as I am testing it on PCB. The spice models seems to give me the results I am looking for. But unfortunately not the physical circuit. Here I will explain my basic setup.

    The inputs to the analog multiplier are two square wave signals (NRZ) of different amplitudes.So when these two signals are multiplied, I expect the analog multiplier to give me the output which is the product of the amplitudes of the two inputs (Example: V1= 0.2 Vp-p, V2= 1 Vp-p then DC output of AD834 Vout=0.2*K Vp-p (lets assume K is some factor)).

    There are two major issues I am facing.

    1. There is zero variation of DC output of AD834 with the variation in input amplitudes.

    2. The voltage feedback opamp AD8397 takes in large currents then specified, especially the negative pin.

    Can you see any issue with the circuit ?

    regards

  • Hi Eshwar,

    Thanks for the schematic, I can add some comments.

    Which data sheet circuit figure are you using for a reference?

    The op-amp U2 looks like a SOT-23, but the p/n AD8397 is for a pair of low-frequency opamps packaged in an 8-pin SOIC. Not sure which part you have in the circuit but if it is really an AD8397 you'll have to change it to something else. Fact is, I'm surprised you see a response from the SPICE model, the circuit should not work at all, neither in a physical sense nor in a simulation.

    The AD834 is actually a gm device where the input is voltage and the output is current. The output looks like a current source, which has infinite impedance, and is equivalent to a voltage source with an impedance of 50 ohms. You must connect the circuit configured as shown in figure 9 of the data sheet where the W1 and W2 outputs of the AD834 are connected to the inverting mode inputs of two HF op-amps. The op-amps should have decent frequency responses, such as the AD8001 which is characterized as a current-mode amplifier.

    You said the circuit is DC, and it is indeed dc as far as pulse signals are dc. The intent of the circuit of figure 9 is truly dc, (same as a battery). When the input is pulsed, it is not the same as a continual dc from a battery and any amplifiers must be capable of handling the transitions with no distortion. You can use the circuit shown in figure 9, just change the op-amps to AD8001 (or AD8009) and let me know what you see.

    Best,

    James