Forum Replies Created
- Posts
The GPIO is driven with two, SN74AVC4T774 chips. 10k pulldown, 100 ohm series resistor. The 100 ohm resistor will limit your switching rate as you add capacitive loading, and you might have to worry about ground bounce, etc. with long cables. But this should give you enough to model it if you have concerns.
A fan-out buffer board could always be the fall-back position. I’d use something like a Max 10, so that you could get creative if you needed to drive all 60+ devices independently (e.g. one nibble is data, the other nibble is address for 64 independent outputs)
It looks like some strong VSWR effects. I would throw a 6 or 10 dB attenuator on the SA and TG and repeat the test. If this does not solve the problem, I would try different cables.
The most basic checks are:
1) Did you click “RF ON” to turn the RF output on?
2) Did you click “MOD ON” to enable modulation and sweeps?Let me know if this does not work.
Flo,
The image response will be 21.4 MHz above the 20-100 kHz signal. If your configuration is measuring flicker noise or similar, and you have a low frequency amplifier with, say, less than 10 MHz of bandwidth, between your DUT and the SA44B, you would not subtract 3 dB, as the noise at 21.4 MHz would be negligible compared to the noise at 20-100 kHz. You could turn off spur rejection, deal with the spurs, and not worry about VBW settings. Or you could set VBW to 1/10 RBW with spur reject on, and still get a valid reading.This has to do with the SA-series software-based image/spur rejection algorithm. Behind the scenes, it is injecting LO high side, making a sweep, then injecting low side, making a sweep, and picking the lower measurement. For this reason, the SA44B is not recommended for broadband measurements (such as broadband noise).
If you need to make a noise measurement with the image/spur rejection enabled, set your VBW to 1/10 the RBW. You will need to compensate for energy at the image frequencies if your noise bandwidth is >42 MHz. Subtract 3 dB from the measurement for these situations.
The alternative is a BB60C or an SM200A, which have excellent hardware-based image rejection and much lower spurious. These will provide good noise measurements without any tricks.
Idan,
You can record IQ data in Spike’s zero span mode, in binary .iq files, or text-formatted CSV files. See the Spike software manual for more information.
AlexH,
On some devices, the mixer’s 0 Hz feed-through brings up the noise floor close in. It really depends on how well the mixer is balanced. It sounds like you got a well balanced mixer. Your readings are typical, but we can’t guarantee all SA44Bs will perform that well.
This is a difficult question to answer. My first question is, what antenna distance are you using? Is it the same distance in both cases (certified test and pre-compliance test)? If not, there is a 20 log (d2/d1) dB correction to apply.
Obviously your chamber is not anechoic, which will change results, and VSWR on your antenna and amplifier can change things. I can see these stacking to produce maybe a 6-8 dB variation, but 20 dB seems a bit excessive.
Do you have two of the antennas? If so, you could calibrate your system in position using a signal generator instead of your DUT…Quadsat,
We do not specify relative amplitude accuracy, but as an FFT-based spectrum analyzer it is typically very good. After the device has warmed up for an hour, as long as you avoid switching bands which can add a tenth or two (dB) to the uncertainty, you should typically see a few hundredths of a dB relative amplitude accuracy until you approach DANL or the device changes temperature.
April 25, 2019 at 9:28 am in reply to: BB60C, EMC measurements:How to lower the noise floor of my setup ? Karlist,
With the BB60C, maximum sensitivity is achieved at 55 dBuV maximum input level, or -50 dBm maximum input level.
As far as measurements being 20 dB off… is the antenna the same distance as in your test site, or did you compensate readings for distance to DUT? Did you factor in any preamplifiers or cables used? Using a chamber without absorber will skew your results as well, but not by 20 dB… however, using a test distance of 3 meters instead of 30 meters would give you a 20 dB offset.Unfortunately, no. Due to the small pattern buffer, a test GPS signal cannot be generated.
With our VSG60A (coming later this year), you can play very long patterns. If you have software that generates a GPS test signal, the VSG60A could play it.Flo,
1) If I read that correctly, the gain of your amplifier is 1000 volts per amp. So I would think you would convert the SA44B noise measurement to volts per rt Hz, and then use this as your noise measurement (mA/rt Hz because amp output is 1V per mA input).
2) You are using power averaging, so there is not a 2.5 dB correction to apply.Also, please note that if your gain was 1000V / V, this is actually 60 dB of gain, not 30 dB. (20 dB gain = 100x power increase = 10x voltage gain)
Jared,
Because the SA series uses a software algorithm for image rejection, it is not a good candidate for EMC precompliance (except possibly for measuring fixed clock harmonics).We will be adding a preselector to the BB60D, which should be out later this year.
In the short run, assuming you are focused on accurate harmonic measurements, a high pass filter to remove the fundamental but pass the harmonics is a good solution. Mini Circuits has a collection of low cost SMA high pass filters.
The ENBW is going to be very close to the selected bandwidth. My guess would be at most a 5% difference, or maybe ~0.2 dB error.
Qswitch,
If you use “high dynamic range” scalar network analysis, the output power is both -10 dBm and -30 dBm in both cases. The damage level for the SA44B is +20 dBm regardless of attenuator setting (it could probably handle a bit more if the attenuator was set to 10-15 dB), but you can’t accurately measure signals above +10 dBm.
For passive high dynamic range, the pair of settings used is -30 dBm TG & -30 dBm SA (to 0 dB insertion loss), and -10 dBm TG & -50 dBm SA (for >40 dB insertion loss)
For “active” measurements, the SA reference level is moved up 20 dB I believe. The included 20 dB pad can be used to measure up to 40 dB of gain. To measure more gain than this, additional attenuation would be required.
If you are measuring an amplifier capable of more than +20 dBm output, I recommend using the 20 dB external attenuator. Or, more specifically, use enough attenuation so that if your amplifier saturates, it won’t blow the SA44B.You are correct that scalar network analysis units should be dB, not dBm, but our software’s control panel (at least when the manual was written) required you to enter a dBm reference level, and used this as dB. This may have been fixed since then.
marvistamike,
Do you have a torque wrench for tightening the SMA? Quite often, simply cleaning with a bit of isopropyl alcohol if needed and tightening (but not over-tightening) the SMA to ~7 inch-pounds can fix the problem. If not, we can set up an RMA for you.
I guess I should add that an isolation transformer could, in theory, work, but one that performs well down to 10 Hz might be very hard to find. As for an opto-isolator, my primary concerns would be linearity and noise, and you would likely still have to DC block.
Last time I needed to AC couple down to 10 Hz, I used a 220 uF aluminum polymer cap for DC blocking with a 1k to ground after, using two SMA connectors and a soldering iron. It wasn’t pretty, but it worked.
If your lowest frequency is 10 Hz, and your highest frequency is 100 kHz, and electrolytic will probably work fine. If you need to go up to a MHz or beyond, a polymer cap will work. For 10 Hz, you would be looking at something like 200 uF – 470 uF.
A 1 kohm resistor to ground after the cap will help discharge the cap AFTER powering on your device and BEFORE connecting the SA44B. A 50 ohm series resistor somewhere is desirable if you have a low impedance output.Hopefully this helps.
Maksym,
I and Q values are voltage rather than power, so you have to square them to get a power unit.