08-08-2018 12:32 PM - edited 08-08-2018 12:33 PM
Hi all
I am trying to drive RF Pulses Amplifier (Tomco BT02000-AlphaA) from PXIe-5413 waveform generator in pulsed mode with 50 ohm load at output of RF Amplifier but mismatch LED starts to blink as soon as i increase the PXIe-5413 output voltage from above 100 mVpp.
With the same setup, if i drive Amplifier with function generator in pulse mode, the Amplifier works fine with function generator output up to 1 Vpp.
I am working in frequency range of 80 KHz to 500 KHz and with duty cycle less than 1%. If we talk about specifications, PXIe-5413 has almost same specifications as the function generator i used.
What can be possible reason for this behavior and what remedies NI can suggest?Is there any isolation required b/w PXIe-5413 and RF Amplifier input although output impedance of PXIe-5413 matches with input impedance of RF amplifier i.e. 50 ohm?
Thanks
08-09-2018 11:39 AM
Hey there,
I checked that the mismatch LED is on the amplifier and not the NI hardware but was not able to find anymore information on the expected behavior of the LED. Do you have a manual that we can refer to and post the information on the post?
Additionally, I can't think of anything else that might cause that since the impedance is matched on both ends. I haven't been able to find anything in the manuals or specification guide on whether the device is isolated, making me think that it's not.
Something of note is that we want to make sure that the 5413 is working as expected. Can you verify the output on a scope of some sort?
Thanks.
08-09-2018 04:23 PM
Asim,
I have some other questions that you could answer to help us understand your application of the 5413.
08-13-2018 10:42 AM
Thanks for help. My problem has been solved. Actually i was giving a very wide bandwidth pulse to Tomco Amplifier RF IN through PXIe-5413 and the load attached at output of Amplifier was of narrow bandwidth. That was the reason i was getting mismatch. As i reduced the bandwidth of pulse, it was working fine.
Regards
08-13-2018 01:43 PM
That's good news. I'm glad you got things working.
Let me point out another potential problem I see with your design. You said:
I am working in frequency range of 80 KHz to 500 KHz and with duty cycle less than 1%.
The 5413 will not be able to achieve a duty cycle less than 1% with a 500 kHz square wave, because it has a rise time of 17 ns.
The pulse width of a square wave with 1% duty cycle and 500 kHz frequency would be
pulse width = 1/frequency * duty cycle
pulse width = 1/500,000 * 0.01
pulse width = 20 ns
It would be impossible to have a pulse width of 20 ns, given that the rise time is 17 ns. NI-FGEN will give you a error if you try to program the device with those numbers.
The minimum pulse width for square wave is going to be roughly 2.2 * the rise time. This gives the output enough time to rise, settle, fall, and settle.
Note that other NI signal generators in the same family have faster rise times.
5413 - 17 ns
5423 - 8 ns
5433 - 3.75 ns
So, you might need to use a 5423 or 5433 to get shorter pulses with sharper rising / falling edges.
08-13-2018 11:40 PM
Hi Balexand
Thanks for advice. Here in my case, i transmit a frequency modulated pulse of 1 ms duration and there is a 500 ms delay (off time) before i transmit the next 1 ms pulse. The duty cycle calculated will be (1/500)*100.
Regards
Asim