AnsweredAssumed Answered

33500 Seamless transfer from constant frequency to sweep

Question asked by john on Mar 28, 2013
Latest reply on Apr 15, 2013 by lhornburg

Like • Show 0 Likes0
Comment • 5

Hi!

The situation in more detail:
The 33500 generates a sinus with 1000Hz. I want to start a sweep (1000->2000Hz in 2s), but I want it to start "in phase" with the current waveform. I don't want it to immediately switch to sweep when I send the SCPI-command. Instead the sweep shall start when the current sinus period is finished (seamless transfer).
Is it possible? How?
Note: arbitrary is not an option since I don't know in advance when the need for the sweep appears...

Many thanks!

No one else has this question

Outcomes

5 Replies

lhornburg Mar 28, 2013 3:39 PM
Mark Correct
Correct Answer
Hello John,

Yes the 33500 can do this. When you use the sweep mode of the function generator, and you set the trigger source to BUS. The function generator will sit at the start frequency until you send a *TRG. I believe it will finish it's current cycle and begin sweeping upon receipt of the trigger. There is no disconitinuity I could see when I just tried this out.

Let me know if you have more questions.
Like • Show 0 Likes0
Actions
- john @ lhornburg on Apr 9, 2013 1:00 AM
  Mark Correct
  Correct Answer
  Thanks for the input!
  
  Your suggestion kind of pushes the problem with the discontinuity to another point in time. As soon as the sweep function (SOURce1:FREQuency:MODE SWE) is sent over scpi the amplitude makes a jump regardless of trigger type. This even if the current frequency of the generated signal is the same as the start frequency of the sweep. But you are right, if you choose BUS as suggested the actual start of the sweep when *TRG occurs is smooth.
  
  Background:
  What I really want to do is to simulate a speed sensor which measures impulses on a rotating shaft. I want to be able to ramp between different shaft speeds and mix that with constant speeds segment (a speed profile with linear segments). Everything smooth, without discontinuties as it would be from a real rotating shaft.
  
  Maybe FM is a solution? But I really don't want to use an external modulating source since it is not available to me. And I have tried to do it using internal FM source without success :(
  Like • Show 0 Likes0
  Actions
  - lhornburg @ john on Apr 11, 2013 2:04 PM
    Mark Correct
    Correct Answer
    Hello John, The amplitude is jumping???
    
    What firmware rev are you running?
    
    Maybe you can send a scope shot of the jump?
    
    FM could work, but I think you will run into problems with the triggering aspect. If recollection serves me correctly, you can't trigger FM.
    
    I think there could be a few different ways to achieve the performance your looking for, but sweep is the most straightforward method in my opinion. Let's see if we can figure out why this thing is jumping in amplitude.
    
    Regards,
    
    Luke
    Like • Show 0 Likes0
    Actions
john Apr 12, 2013 1:38 AM
Mark Correct
Correct Answer
Luke, thanks for your interest!

I'm not native English so maybe "jumping" is not the correct word. Anyway, the firmware is (*IDN?)
Agilent Technologies,33512B,MY52400235,2.03-1.19-2.00-52-00.

I've attached a scope shot showing the behavior when I send these scpi commands.

*CLS
*RST
OUTPut1:LOAD INFinity
OUTPut2:LOAD INFinity
SOURce1:APPLy:SINusoid 2, 6.5, 4.75
OUTPut1 1

sleep for 1s

SOURce1:FREQuency 4

sleep for 1s

SOURce1:FREQuency:STARt 4
SOURce1:FREQuency:STOP 8
SOURce1:SWEep:TIME 1.5
SOURce1:SWEep:HTIMe 1000
SOURce1:SWEep:RTIMe 1000
TRIGger1:SOURce BUS
SOURce1:FREQuency:MODE SWE

sleep for 1.5s

*TRG

That starts with a low frequency which is increased where the first vertical cursor is. Here the transition to the new frequency is smooth. A little bit later the sweep command is sent and that is where the amplitude "jumps". A while later (second vertical cursor) I trigger the sweep and it starts smoothly without "jump".

What I hoped for is that the "jump" where the sweep command (SOURce1:FREQuency:MODE SWE) is sent didn't occur.

Any idea how to avoid it?
Attachments
- print_01.bmp2.3 MB
Like • Show 0 Likes0
Actions
- lhornburg @ john on Apr 15, 2013 2:05 PM
  Mark Correct
  Correct Answer
  Well you could have fooled me with your English :)
  
  The issue I believe you are describing is the discontinuity when switching from Constant Wave to sweep mode. Sweep has to do a lot of hardware config, and the switch was never meant to be phase-continuous.
  
  However, I would suggest leaving your output off until you have finished configuring the box for your application.
  
  Basically wait until after you have sent SOURce1:FREQuency:MODE SWE, to turn on the outputs. This way you will avoid the discontinuity. If the sweep is bus or manual triggered then it will park at the start frequency just like CW mode. The trigger then causes the sweep to start and is phase-continuous, as you saw previously.
  
  If you hadn't discovered it already you can use the sweep hold time parameter to dwell at the end frequency if you'd like, and the return time parameter to smoothly sweep back to the start frequency.
  
  If this doesn't satisfy your needs let me know. There is another way to do this using sequencing, but it is a little more difficult.
  
  Regards,
  
  Luke
  Like • Show 0 Likes0
  Actions

33500 Seamless transfer from constant frequency to sweep

Outcomes

Attachments

Related Content

Recommended Content