MSOX3054A with Spectrum Visualizer - Can I trigger RF with I2C decode

[jstarr]

I guess this question will really show how much of a novice I am but I got a MSOX3054A on loan from the job in the evening s for a few week to try to improve my capabilities.  I have been doing some basic I2C and RS232 decoding.  I have a 433Mhz wireless transceiver that sends data packets controlled by an I2C bus and cannot find the way to trigger the scope on the RF bursts.  I was wondering if Agilents Spectrum Visualizer   (ASV) option would allow me to trigger on the I2C decode and also trigger the RF burst that follows the proper write frame on the scope's 3rd channel with ASV?  I am being super-naive???

tia
john

[AndyC_772]

It's not entirely clear why you can't trigger the scope on the RF burst. How are you getting the RF into the scope? Do you have a direct (ie. wired) connection between the transmitting antenna and the scope, or are you using a receiver/decoder to convert the RF back into a digital waveform? Are you trying to directly observe the RF carrier or just the decoded output?

If you want to see the carrier itself, and to examine it in any meaningful detail, then you'll struggle to capture both that and the I2C signal together because of the large difference in time base. I2C is about 1000x slower than a 433 MHz carrier, so if you set the time base appropriately for one, it won't be right for the other.

[Neganur]

Edit: reading Andy's reply, yeah make sure you're not blasting the input of your scope with the power from the transmitter. Remember that the maximum input derates with increasing frequency.

When triggering on the RF burst, have you tried adjusting the trigger holdoff? Sometimes this works, but you might also not be able to trigger off the modulated signal in your case.

Not sure if this answers your question but the ASV trigger is either in free-run or trigger on power level (only the FFT window).
ASV configures the scope continuously while it takes control of the scope to pull data, so you cannot manually change the horizontal setting on the scope during ASV.

I can't duplicate your setup here at home but what I tried was (using the demo mode signals of the scope):
- start the ASV software and once it runs, click on 'Stop' (the application loves to reset your scope to default)
- enable the I2C training signal (demo 2 is SDATA, demo1 is SCLK)
- to feed the I2C training signals to the digital pod D0(SCL), D1(SDA)
- Disable all other serial channels with the digital menu, threshold 1.4V (the training signal is 2.8 Vpp)
- Serial decode menu set to serial 1, I2C mode, 7 bit address size,
- in the trigger menu set the trigger type to S1(IC2)
- set to trigger on write7 i.e. frame(start:addr7:write:ack:data) with Address set to 0x29 and data 0x01 so you can see the data cycle (20, 40, 60, 80 etc )
(adjust the trigger to whatever is meaningful in your application)

you can now use either the analog channel 1 or 3 to probe the RF signal, although 433 MHz burst on a 500 MHz BW scope... hmja, depends on your modulation type I guess.

- in the ASV software, click on 'Run' and adjust the start/stop frequency and RBW to what you need.
You should be able to at least see the carrier with your scope, not sure what the stop frequency can be set to with the MSOX-3054A

[jstarr]

Thanks all. Maybe I should start by first figuring out how to trigger on the RF burst.  Its a "Moteino" (like Arduino) but with an RFM69 transceiver at 433Mhz.  The bursts are on the order of the I2C bus speed since its turned on and off by the I2C bus and its FSK mod around 55 kb/s.   Output is about 20dBm at 50ohm so I have it connected directed from RF69 SMA output to Ch1 on the 3054A (set to 50 ohm imp).  I know that looking at 433Mhz on a 500Mhz scope is marginal but I was doing this not for quantitative accuracy but to qualitatively see how the scope works and whether my thinking on using the I2C write frame that turns on the RF would be the right trigger event.

Again any advice is greatly appreciated.

[Neganur]

It's hard to tell how your signal looks since the datasheet that I found of the RFM69 is very generic.

If the TX burst length is constant you should be able to trigger on it with plain edge trigger & trigger holdoff. (in the standard sketch it is 300 ms?)
I don't know what your I2C speed is (does that thing really use I2C and not SPI?) but I would think that it should be possible to trigger on the 'set tx mode' with the MSO pod listening to the SPI.
A last resort would be to have your code set a pin on the moteino every time you send data, which your scope can use to trigger.

Once you get a nice steady signal of the RF on the scope, ASV should be able to show the frequency shift in the spectrogram view with a narrow enough RBW (I could not figure out what your channel width is, is it 25 kHz?). Although I fear the update rate might be too sluggish.

[w2aew]

Thanks all. Maybe I should start by first figuring out how to trigger on the RF burst.  Its a "Moteino" (like Arduino) but with an RFM69 transceiver at 433Mhz.  The bursts are on the order of the I2C bus speed since its turned on and off by the I2C bus and its FSK mod around 55 kb/s.   Output is about 20dBm at 50ohm so I have it connected directed from RF69 SMA output to Ch1 on the 3054A (set to 50 ohm imp).  I know that looking at 433Mhz on a 500Mhz scope is marginal but I was doing this not for quantitative accuracy but to qualitatively see how the scope works and whether my thinking on using the I2C write frame that turns on the RF would be the right trigger event.

Again any advice is greatly appreciated.

I imagine you want to do something like what I show in this video. However, on your scope, you may have difficulty getting sufficient sample rate and memory depth for the I2C and RF simultaneously (as already stated).

[jstarr]

Thanks Alan, yes I was trying to see how far I could get without a spectrum analyzer.  I was hoping that the Agilent FFT Spectrum Visualizer software would allow me to trigger on the RF waveform while decoding the RS232 to PC and SPI code to the RF69 xcvr.  I wasn't looking for great resolution but more to learn how to squeeze the most out of the scope while familiarizing my with its functionality.  The scope has 4 GSa/s and 4Mpts plus segmented memory so I thought it should be able to trigger on the 433Mhz FSK waveform with a direct 50 ohm input (I have the 20dBm output version).  I am torn between the new Tek MSo scopes with the integrated SA (but no tracking generator)  as you have versus the Agilent MSOX3054A plus a Rigol DSA815 with a tracking generator as I squeeze the most out of my limited budget :-)

john

[kilohercas]

You can find RF detectors, they will generate TTL signal, if any power is generated, i guess that will work for you ?
http://cds.linear.com/docs/en/datasheet/5532f.pdf