Help with Rigol 1102E problem

Thread Starter

Buzz25W

Joined Apr 9, 2013
3
Hi, I'm a new member of the forum and I noticed a thread about Rigol scopes. I'm having a problem that I can't seem to resolve. I've called Rigol, but have not received an answer.

Problem: with a 3.875 MHz + - few hundred Hz unmodulated clean carrier at 1.0ms/div horizontal and 500mV/div vertical, I observe the pattern in the link/attachment Rather than a clean unmodulated carrier, this is what I see. I am told that this is an aliasing problem. Is there any fix/adjustment for this? I would like to maintain the current sweep rate of 1.0 ms/div and must maintain the frequency 3.875 MHz. I am using the scope to view amplitude modulation on the carrier.

Buzz

http://forum.allaboutcircuits.com/attachment.php?attachmentid=54387&stc=1&d=1365548245
 

Attachments

MrChips

Joined Oct 2, 2009
30,711
The Rigol DS1102E is a 100MHz scope with a sampling rate of 1Gs/s.
The problem is when you select the HOR SWEEP of 1ms/div you are asking the scope to display 10ms worth of data. In order for the scope to do that it has to reduce the sampling rate and that is why you see aliasing, i.e. the sampling rate is too slow.

You have to change your HOR SWEEP setting. There is no other solution that I can think of.
 

YokoTsuno

Joined Jan 1, 2013
43
There is no other solution that I can think of.
According to the manual on the Rigol website this instrument has two memory sizes 16K and 1MSamples called Long Mem and Normal.

10ms worth of data in a 16K memory is equivalent to a sample every 625ns or a rate of 1.6 Msamples/s which is indeed too low. Shouldn’t Long Mem mode resolve this problem?
 

crutschow

Joined Mar 14, 2008
34,283
In the olden days they used a probe with a built-in rectifier to view the modulation envelope of a signal. You could do the same with a fast diode and a simple RC filter (R and C in parallel to ground). You want the filter corner frequency to be somewhat above your highest modulation frequency.
 

vk6zgo

Joined Jul 21, 2012
677
AM Modulation Monitors as used at Broadcast stations are a more sophisticated version of the rectified probe.

You could read the DC voltage out for an unmodulated carrier, & compare it with that of the modulated signal & calculate the mod % that way.

Usually,instead of all the Maths,we just calibrated them using a Trapezoid pattern on an analog Oscilloscope,so if you could borrow one,you could set up your "Mod Monitor",using the DSO as the display.

Of course,if you can borrow an analog 'scope,you can just look at the modulation with that!:D

There is another way--- you can find out what frequency carrier is not affected by the aliasing problem,& convert your 3.875MHz signal down to that.
Back in the day,it would have been easy,as there were a lot of Shortwave Radio Receivers around with 455kHz IF frequencies,& you could have just connected a 'scope probe to some convenient point & tuned the Radio to your original HF signal.
 

Thread Starter

Buzz25W

Joined Apr 9, 2013
3
I sincerely thank all of you have replied to my post. May I add the following information and comments:

1. The artifact disappears, or at least is not visible becasue of the high frequency of the artifact, if the signal is only a few kHZ away from 3.875 MHz. It re-appears in a different form at other fequencies (the form posted resembles the "beating" of two close frequencies)

2. I have not tried either of the remedies suggested (different sweep frequency, changing the memory settings) and can not do so for about one week becasue of lack of access to the facility

3. With assymetric amplitude modulation as is my case, neither the classical trapazoid display, nor envelope detection provide the dynamic information that a direct dispay of the modulated carrier provides.

4. Currently the modulated carrier is directly sampled at the carrier frequency, and down or up conversion to avoid 3.875 MHz is not desirable because other artifacts may be introduced.

5. The 1.0ms/div sweep rate is chosen becasue it provides the best view/undertstanding of the modulated signal. If at all possible, I would like to maintain that sweep rate.

6. Yes, an analog scope would solve the problem, but I found the price, size, and operting capabilites of the Rigol so attractive that I bought it without considering the downside of a digital sampling scope.

I would sincerely appreciate further thoughts and assistance.

James
 

vk6zgo

Joined Jul 21, 2012
677
Video modulation of a carrier as used for Television,is Assymetrical,in that
the negative modulation is not the same as that for the positive modulation.
In TV,DSB (&VSB) detection is commonly used to determine modulation levels.
100% negative modulation is indicated by zero output from the detector.

I'm not sure if my comment re frequency conversion was clear enough.

If you feed the 3.875 MHz signal into a frequency converter,using an appropriate local oscillator,the modulation envelope of the converted signal remains the same as that of the original signal.

For many years the ARRL Handbook carried an article for a just such a device,so that lower bandwidth analog Oscilloscopes could be used to examine higher frequency signals.
 

vk6zgo

Joined Jul 21, 2012
677
I sincerely thank all of you have replied to my post. May I add the following information and comments:

1. The artifact disappears, or at least is not visible becasue of the high frequency of the artifact, if the signal is only a few kHZ away from 3.875 MHz. It re-appears in a different form at other fequencies (the form posted resembles the "beating" of two close frequencies)

This is basically what it is,but under a different name!

2. I have not tried either of the remedies suggested (different sweep frequency, changing the memory settings) and can not do so for about one week becasue of lack of access to the facility

3. With assymetric amplitude modulation as is my case, neither the classical trapazoid display, nor envelope detection provide the dynamic information that a direct dispay of the modulated carrier provides.
See my previous comment.
4. Currently the modulated carrier is directly sampled at the carrier frequency, and down or up conversion to avoid 3.875 MHz is not desirable because other artifacts may be introduced.
A passive mixer will be unlikely to introduce significant artifacts.

5. The 1.0ms/div sweep rate is chosen becasue it provides the best view/undertstanding of the modulated signal. If at all possible, I would like to maintain that sweep rate.
From this,I assume that the modulation frequency is quite low,& you need to display one or two modulation cycles of the RF Envelope .

6. Yes, an analog scope would solve the problem, but I found the price, size, and operting capabilites of the Rigol so attractive that I bought it without considering the downside of a digital sampling scope.

I would sincerely appreciate further thoughts and assistance.

James
You are in a bit of a situation,here.
I don't have a Rigol,myself,or any DSO for that matter,but I thought the
DS1102E was more capable than you have found it to be.

There has been a lot of discussion of these 'scopes on another forum,
(eevblog.com),& some of the Rigol owners there may be able to suggest a workaround.
 

Thread Starter

Buzz25W

Joined Apr 9, 2013
3
Your comments and suggestions much appreciated. After speaking with Rigol tech support in the US, and reviewing the problem with them, I have not yet had their promised response. Other than this particular issue I have found the Rigol to be an excellent device, especially for the price. I will check the other forum per your suggestion

James/Buzz
W3EMD
see QRZ.com for info and picture of the Rigol in use with Collins KW-1
 

THE_RB

Joined Feb 11, 2008
5,438
Just "one-shot" record the data on a good samplerate to get enough detail, then after it has recorded change the timebase for display only.

This is all real basic digital scope stuff.

There's no problem DISPLAYING 3.8MHz at 1mS/div, but you have been trying to RECORD 3.8MHz at 1mS/div which just makes no sense!

You 'scope instruction book should state what recording samplerates are used for each timebase setting, it is up to you to make good decisions for recording samplerate vs the frequency you are recording.
 

THE_RB

Joined Feb 11, 2008
5,438
Why? I use one-shot all the time for adjusting things, it's a single button press to record and hold a waveform.

Another option is to change the input to "peak capture" or whatever it is called on that scope. That mode will capture the full range of voltages that occur between each displayed sample, so each sample will display as a vert line from min to max voltage. That will also fix the aliasing caused byt he low samplerate and let you see the amplitude modulation.
 
Top