1. Simple modulation testing above UHF.
First I get straight on with monitoring typical modulation sidebands, I’m at 2,320MHz with DSA1030A-TG3 - a performance spectrum analyser from RIGOL incorporating a 3GHz tracking generator. As we’ll see, the user has a wealth of advanced measurement capabilities at their disposal.
Fig 1. DSA1030A-TG under 13cm band RF test. The RIGOL-UK team at Telonic Ltd. let me hook up to a DSG3060-IQ Pro RF source1 .
Rigol headlines their 3GHz model like this:
• 3GHz measurement Bandwidth (from 9kHz up)
• Achieve low noise floors down to DANL of -148 dBm
• Low phase noise -88dBc/10kHz
• All digital IF Design
• Easy to read bright screen, plus a VGA-out socket you can display on large lab monitor.
• 10 Hz Minimum Resolution Bandwidth
So what does this mean when you look closer at real 13cm signals? Can I easily see 40dBc s/n ratio? Well here’s my RF carrier at 2,302.2 MHz:
RIGOL’s DSG3060-IQ is rather more than a ‘sig gen’: it can push the envelope of many an RF test e.g. modulating a 5.6 GHz carrier at 100 million 256QAM characters per second! Here we stick to a modest 2,320.2MHz test!)
Fig. 2 - This is quite a demanding plot: the resolution is down at 10Hz, with quite a low noise floor yet my sweep time is a nifty 40-odd seconds. (I mistakenly left in 10dB input attenuation, so it’s really 10dB better!). I don’t often need 10Hz resolution, so sweeps can be sub 1 second.
Now I add my modulation. For no particular reason, I chose Phase Modulating with a 1.1 kHz sine with fairly wide phase deviation setting on the signal generator (Fig 3.):
The aim of this test is to explore the analysers ability to quickly measure occupied bandwidth of my modulated signal. The AMK (Advanced Measurement Kit) purport to have many automated functions. One is Occupied Bandwidth OBW) - the analyser automatically puts internal marker around your transmission, analyses it and calculates and reports the bandwidth.
So let’s now take a look at how the DSA1030A-TG gets on displaying this modulated spectrum, and how the AMK feature (standard with –A model) reports the OBW results at the bottom of the screen in fig 4.:
Fig 4. - You can immediately see the carrier power now spread over a number of frequencies in the expected comb pattern for a phase modulated signal. As per theory the tines of this comb are 1.1 kHz apart (each of the ten horizontal squares is ~2.4kHz of the span).
I get an accurate measure of bandwidth occupied by sidebands with just a push of the button.
2. Inside the technology - for our RF instrument designers
– the Inside track on RIGOL’s approach to designing-in advanced performance and stability
RIGOL adopts a proven digital IF approach, citing reliability and performance success of its little brother, the long-established DSA815-TG. The more capable DSA1030A-TG is aimed at the most demanding and widest range of RF applications but this time up to a full 3GHz.
Stability and precision is the primary design goal hinging around its all-digital IF core with 10Hz resolution bandwidth, -88 dBc/Hz phase noise (typical) at 10 kHz offset. The front end is impressive with up to -148dBm displayed average noise level (10 Hz RBW, standard preamplifier on) and less than 1.0 dB total amplitude error, affording high precision measurements whether your application calls for low noise or narrow resolution.
Rigol boils down the advantages you get from their design approach like this:
1. The ability to measure smaller signals: Digital IF design approach and filter technology enables smaller bandwidth settings, greatly reducing the displayed average noise level.
2. The ability to distinguish small signals by frequency: using the IF filter with the smallest bandwidth setting it is possible to make out signals with a frequency difference of only 10 Hz.
3. High precision amplitude readings: this technology almost eliminates the errors generated in traditional spectrum analyser design filter switching, reference level uncertainty, scale distortion, as well as errors produced in the process of switching between logarithmic and linear display of amplitude arising from a traditional analogue IF design.
4. Higher reliability: compared with traditional analogue designs, the digital IF greatly reduces the complexity of the hardware, the system instability caused by channel aging, and the temperature sensitivity that can contribute to parts failure.
5. Your measurements, faster: digital IF technology improves bandwidth precision and selectivity of the filter, minimizing the scanning time to display your results a lot faster.
3. Using DSA1030A-TG for filter / cavity testing and tuning
Fig 5. A quick photo that speaks for itself: showing a black band pass filter under test.
The RIGOL reveals this filter under test is good below 1.2GHz, but it does a poor-ish job of rejecting some frequencies above 1300MHz – with rejection as bad as 20dB at some points e.g. 2 GHz and 2.9 GHz.
The picture on the left shows the dB scale better.
4. Keeping tabs on harmonics with DSA1030A – courtesy of G3WKZ
Colin Bayliss (G3WKZ) has also used this setup to test the purity of emissions from his own implementation of a 70MHz Transceiver. This is Colin’s comment on his build and test:
“Here’s my solid state 4m PA unit that I built earlier this year from modules …Performance is good I’ve achieved over 60 IARU locator squares on 4m this summer from both normal ground wave / tropo and from sporadic E propagation. It’s built into a self-contained PA unit with associated power supplies.
Fig 6. In the above test the third harmonic is good and low. We set ourselves a goal of up to 20dB of clear water between carrier (left) and 2nd harmonic (about 2.5 squares from the left). A tough target because it measures harmonics directly at the transmitter output where it’s worst. Antenna systems obviously give a massively preferential gain at the fundamental, with low gain at 2nd. Colin concludes: “My PA underwent spectrum analyser testing using a RIGOL 3GHz spectrum analyser and showed very satisfactory results. The spectrum above is a test of my single BLF177 MOSFET Power Amplifier (PA) and associated 100W throughput, 9th Order, Low Pass Filter (LPF) to give a clean RF output.”
5. Using DSA1030A markers to make 3rd Order Intercept IP3 measurements
I can’t escape without attempting a serious receiver pre-amp measurement, so I opt for one of the most important tests in professional RF design. First I remind myself: why do we 3rd order Intercept Point (IP3 / TOI) of an amplifier?
• IP3 is an important parameter for nonlinear systems like mixers or amplifiers which helps to verify the quality / linearity of the device.
• It is a measure of the receiver performance in the presence of strong nearby signals.
• It can be used to define the upper limit of the dynamic range of an amplifier.
Any devices with nonlinear transfer functions generate harmonics. In the presence of two sine wave signals with a small frequency distance intermodulation products are generated.
The maths looks like this (Fig 7. for f1, f2 etc.):
Pout = A1f1 + A2f2 + Fundamentals @f1 and f2
A3(f1+f2) + A3(f1-f2)+ 2nd order IM
A4(2*f1+f2) + A4(2*f1-f2) +
A5(f1+2*f2) + A5(f1-2+f2) + 3rd order IM
A6 (2*f1+2*f2) + A6(2*f1+2*f2) + 4th order IM
We can see that worryingly, the 3rd order Inter-Modulation (IM) products are located directly beside the two fundamentals all the other are far away and therefore less likely to interfere. The 3rd order products may well lie within e.g. a receiver pass band, or exactly at frequencies which might trouble adjacent communication channels. Certainly this is one of the main reasons why we look on IP3. Another one is that the two-tone approach has the advantage that it is not restricted to broadband devices and is commonly used for radio receivers.
Why the two-tone approach is commonly used to test radio receivers? The main advantage is that can also be used for measurements on devices with smaller bandwidth, the method itself is simple as long as you set up capable test equipment with good quality output.
But in practice how do we determine this theoretical intercept point of an amplifier? The problem is that the value cannot be measured directly as the amplifier compresses or limits before the IP3 point is reached. In other words, it is an extrapolated convergence of intermodulation distortion products in the desired output. It can be done graphically by expanding the transfer curve (as if the amplifier exhibited ideal linear behaviour). The cross point between the two lines of fundamental and 3rd order product of intermodulation (i.e. two-tones) is the IP3 point.
Many of you may also know, the relation between the 1 dB compression point and the IP3 point can be summarised neatly with the rule of thumb that the third-order intercept point is roughly 10 dB above the 1dB compression point.
How to select the right parameters for frequency and power?
1) The used power settings should be far enough below the 1dB compression point of your device
2) Use frequencies which are near enough together to reduce the influence of different transmission function. Usually, the test frequencies are about 20 to 30 kHz apart
3) Select frequencies of the two test frequencies that fall within your real application values.
Theoretical Measurement and IP3 determination – I find it helpful to revisit the example in the graph below:
We use an Amplifier with 20 dB gain. The graph shows the input power and the output power. The red line shows the corresponding curve of our PA (incl. compression). The blue line shows the 3rd order IM product. It can be seen that if the input power is changed by on 1dB the out power of 3rd order IM product is changed by 3 dB, compared to the fundamentals with have a 1dB/1dB slope.
IP3 point is located at the input power level where the output power of the fundamental and the output power of the 3rd order IM product is the same. So it can be calculated when you have the measurement result of the output power of both signals fundamental and 3rd order IM product.
Fig 8. IP3 point in theoretical transfer plane.
IP3 out = P1 + ½ (P1 – P3)
My real test setup using DSA1030 and its markers to measure IP3:
To create my two-tone sine wave test signal I used RIGOL’s DG4000 (or DG5000) with the use of Ultra Station software at frequencies @ 10.000 MHz and 10.001MHz (1kHz spacing). Prerequisites: high quality two-tone generator (levelled) such as a RIGOL DG4062 Arbitrary Waveform Generator; Ultra Sigma Ultra Station software and about 10 minutes on the PC. This creates a two-tone waveform file in my DG4062 Generator and with 1 kHz spacing.
I’ve first connected my spectrum analyser directly connected to my DG4062 generator first, to check if the two-tone arbitrary file works as I wanted: my DG4062 can be set conveniently to drive a 50 Ohm Load, I set frequency and output level to -20dBm and the required (straight-through) test signal on the DSA1030A-TG can be seen in the picture below.
Fig 9. Amp input on RIGOL-UK’s DSA1030A-TG. Now I set up the test with the amplifier under test in between:
Hey presto, the DSA1030A-TG now reveals the in-band mod products (note narrow span):
On the DSA I set the corresponding centre frequency and span. Activate two markers and set one on the fundamental (upper or lower) and the second one on the corresponding 3rd order IM product.
IP3 = -32.66 dBm + ½ * (-32.66 dBm – (-60.67 dBm)) = -32.66 dBm + 14.01 dB = -18.7dBm
What a result!
Hopefully our long-winded description above has given some practical insights into using markers for a very relevant and necessary test – the RIGOL DSA1030A is well up to the job of many professional RF measurements and there are many like this we’d all benefit from.
6. The commercial story.
The back-story is RIGOL priced its 3GHz range to enter the market at a very attractive price. So it has enjoyed quite rapid adoption and is already in wide use in RF, Broadcast EMC/EMI electronics manufacturing, failure analysis, and R&D. The commercial spiel from RIGOL is all about speed and efficiency: ‘Up to date and more transportable, the DSA1030A-TG is easier on your pocket with self-calibration, the professional lab enjoys a lower overall total cost of ownership. A series of automatic setting functions such as Auto Tune, Auto Range, Auto Scale and Auto Couple, enabling the analyser to acquire signals and explore parameters automatically, instead of with a lengthy manual process’. (I can précis this as: ‘It’s a lot faster and easier on the back, fingers and eyes than on my traditional analyser!’)
Likewise on practical note, at power-on, I do indeed find it’s nice the DSA1030A-TG can fire up with either Last settings, my choice of User settings, or Factory Default settings (Preset) – with this I can quickly and easily recall previous or favourite measurement set-up. Even as a professional, before I prise open my wallet, I’m conscious of working to a budget and it’s worth giving cost-saving choices offered by RIGOL special consideration:
• Models like DSA1030A-TG3 include a 3GHz Tracking Generator
Note: this also serves as a lab-grade RF CW source, which can save you money on a microwave sig gen with its built-in 1dB step attenuator (up to 0dBm into 50 Ohm)
• A-series has a standard preamp for high performance (-148dBm typ. DANL, 10Hz res.). Note: whilst A-models mean a bit more outlay, you get all the present special offers like Advanced Measurement with –A’s too, which may offer you largest savings (see end).
7. The MEAS Button – the advanced -A series
- accesses some very powerful algorithms
An innocuous looking button named ‘MEAS’ is not the first one my hand would drift to. But one press makes it clear that RIGOL-UK.co.uk is offering some powerful stuff in their deal.
As if by magic I bring to life an armoury of RIGOL’s Advanced Measurement functions on my DSA1030A-TG. These are particularly relevant for those of us wanting easy, accurate measures on transmission/baseband or broadcast signals, without tiresome messing around with multiple settings.
The MEAS button’s one-press gives me a rich menu including these automated measurements:
- Time domain Power, Channel Power, Adjacent Channel Power;
- Occupied Bandwidth, Carrier to Noise Ratio;
- Harmonic Distortion, Intermodulation Distortion;
- Also in quick menus is Pass/Fail, Frequency Count, N dB marker measurements.
I move on to PC control: remote control is easy through USB (as a USB device), there are firmware versions to support LAN too, in fact I’m already dreaming of integrating my regular tests into my lab just by controlling the RIGOL DSA with some standard SCPI commands.
Making records of what you’re doing along the way is a good practice I’ve learned from experience. Saving screens is easiest – whether to PC-folder via USB cable, or to USB memory stick. Or if my memory stick eludes me I can save thousands of measurements and configuration settings using the built-in 1 GB of internal storage that comes as standard.
8. Interfaces and Optional extras.
Here’s the range of interfaces that come as standard on all DSA1030 models:
USB host is available to use a USB flash device to save the instrument settings and history data as well as for firmware updates but I’ve also 1 GB of internal storage and I find I can save thousands of measurements and all my common config settings without impediment.
USB device is available for printing with a PictBridge printer, or to connect as a TMC instrument.
LAN – connection and control via IP address is possible. LXI-C is standard and support for VISA control over Ethernet is included
VGA comes as standard on DSA1030 - Connection for extending screen to an external monitor is provided for demonstrations and training
Options (some are included in the DSA1030A-TG special offer) include Tracking Generator; EMI, faster Quasi-Peak scan processor, Advanced Measurement; Rack-kit; GPIB Optional - Add a GPIB port with a USB-GPIB module (option). Carry bag system option 1000 SCBA also available.
For VSWR tests, a Choice of VSWR bridges: the 1MHz - 2GHz model VB1020 is a very flexible and broadband hybrid design – this is the one for widest amateur use; or VB1040 is up to 4GHz for use above 800MHz. Both easy-fit to RIGOL DSAs. (Type 'VB' into site search bar at RIGOL-uk.co.uk).
In conclusion. RIGOL continues to release a succession of good quality test instruments offering outstanding performance for the price. The noise floor on DSA1030A-TG is exceptional and it’s very powerful with endless uses. I also find this model a relatively compact and rugged design with extra storage (non-volatile memory) enabling storing of field data swiftly to a USB flash device as we’ve already seen so could work very well in field applications requiring transportability. You can add a custom designed well-fitting easy carry system if you need it, so spot tests are easier than ever.
A rich range of measurement functions enhance value: with advanced features like Noise Marker, you can tell from its user-friendly menus RIGOL has aimed this to readily meet the requirements of a broad set of user's measurements. In addition, if I like programmed tests I can take advantage of the out-of-the-box drivers or optional Ultra Spectrum to control more analysis and display functions such as building waterfall curves to expand the measurement capabilities to my own tailor-made lab tests.
The RIGOL-UK.co.uk team and instrument web store has good customer reviews you can read from purchasers of RIGOL spectrum analysers from a wide variety of quarters including RF professionals, repeater groups and EMC teams and specialists all very pleased with DSA1030A-TG.
Dave Phillips MIET – Chartered UK Engineer
Telonic Instruments Ltd
0118 978 6911
When purchasing 3GHz DSA1030A-TG from RIGOL-UK.CO.UK, receive these additional benefits for a LIMITED PERIOD ONLY:
- Advanced Measurement Kit plus the Quasi-Peak+EMI filter options free;
- Special web-order package price
- (this package previously over £5,000 with options above costed-in).
Next steps: phone Telonic Instruments Ltd. for all latest spectrum analyser offers up to 1.5GHz, up to 3GHz, and above 3GHz… Or go to