Forums › BB Series Discussions › BB60 for EMC
- This topic has 16 replies, 7 voices, and was last updated 6 years, 10 months ago by Cory Allen.
- julian April 11, 2015 at 3:53 am
Another reason I really am interested in the BB60 over most other things in my budget, is that I think the RTSA aspect of it will be awesome for precompliance testing.
One thing I haven’t seen, though, is quasi peak detection.. can it be done in the BB60 or the software used?? if not, could it be done with new software or 3rd party software, or is it something that would have to happen in the actual hardware?Andrew April 14, 2015 at 10:11 am
Currently we have not implemented quasi-peak detectors in the BB60. Expanding our EMC functionality is something we are planning on doing the second half of this year. We have a number of features requested by users and that we want to implement ourselves that fall within the realm of EMC precompliance.
I am not familiar with any 3rd party software that uses our unit for EMC pre-compliance.
The real-time functionality of the BB60 is great for EMI testing.
A.J.julian April 16, 2015 at 5:16 am
aaah great! yeah – it *seems* that quasi peak could be relatively easy, on top of all the other stuff you already have working. Though I was not sure if it’s something that needs to run in the actual capture unit itself, or something that can happen in the PC software… if it’s relatively easily doable in software I’d really appreciate it. (in fact – if that BBApp repo on github is your latest work on Spike, and it is a matter of finding the peak detect code in there, I’d be tempted to try and have a go at it)
Seems like, for an RTSA, whose advantage is picking up all the instantansous blips scanning analysers can miss, this would be a very nice addition.
Not a lot of general purpose and hobbyist affordable analysers seem to do it (rigol DSA series has it as an option) and it seems to me that more and more engineers are caring about precompliance these days. So having a specific EMC tool like that, I think, would get you some good attention. (Then you could also market a kit of calibrated antennas for 9k – 30MHz, and at the price those go for you’ll be laughing all the way to the bank… 🙂Jared April 23, 2015 at 5:14 pm
As possibly the person who is the culprit of the list of features AJ now has on his to-do list, here’s how I used my SA44B (oh for a BB60C!).
You can get around QP by using both Peak and Average. QP is proportional to signal duty/waveforms/etc so will always be bracketed by either Pk or Avg.
More important is the RBW, as this needs to be the same or close to that defined in the appropriate standard. I just left everything set to 9kHz.
I found the most useful feature for EMC debugging is the demodulation to audio for signal identification. Especially when in real-time mode.
For EMC debug you dont necessarily need calibrated antennas as you are looking for relative changes. However most RF compliance labs will charge to calibrate and to provide antenna factors vs frequencies.
Precompliance is slightly different as its typically the first test so you wont have a benchmark. Again, a local lab may be able to rent to you a calibrated antenna (with factor table) or calibrate your antenna against a known standard.
By way of an example, I have made a whole bunch of near-field probes, made a couple of PCB-based log periodic antennas and have almost finished an open TEM cell.
You might want to have a read of my other posts (on the Spike sub-forum) about feature requests. They all stemmed from an EMC debugging exercise.
Which, I have just been informed, now passes.
JaredJared April 23, 2015 at 5:18 pm
http://www.wa5vjb.com/products1.html (PCB antennas)
http://www.ewh.ieee.org/soc/emcs/acstrial/newsletters/summer08/pp2.pdf (Tem cell)
http://beehive-electronics.com/probes.html (Near Field probes – I really should get around to ordering some of these…)julian April 30, 2015 at 5:21 am
Good point about peak and average – not so hard to do and nothing new to be developed for that..
Well.. I ummed and ahhed, and went for the lowest expense buy available to me, so now I have a BB60A (2nd hand, ebay… a bit less money than a new SA44 is over here, and about 1/4 the price of the new bb60c… I’m excited to play with the RTSA capabilities, and, well.. when I eventually feel the need to upgrade to a nice flat noisefloor and reduced spurious and better specs, I’m hoping this rumoured 20G unit will be out) Unfortunately I can’t run it yet as my laptop has no USB3 ports.. (d’oh!) looking at getting an expresscard adapter ASAP to run it, but need to know about current draw… came back here to ask that, actually.
Those PCB antennas look like a lot of fun. I have been sort of wanting to start learning enough to design my own for a while, but like to tell myself that a lack of simulation software is holding me back… That said if I’d just started with making them to basic theory and tweaking and testing them a year and a half ago I’d probably have some decent understanding and experience by now! oh well..
I ended up getting tekbox near field probes like this but minus the wideband amp and the wooden box – now I kind of wish I’d got the amp and box too… http://www.tekbox.net/test-equipment/tbps01-emc-near-field-probes-tbwa2-wideband-amplifier They seem quite like the beehive ones. Recommended, if you are interested. Today they did a great job at showing where a local oscillator in a configurable radio was dropping out of operation. At some point I’m going to try making my own using RG405 rigid and compare response to these.
That diy TEM cell looks like it might be too much fun to pass up. You have to give us a report on it here once it’s up and running, and you’ve had a chance to use it.
julian May 14, 2015 at 7:07 pm
- This reply was modified 7 years, 10 months ago by julian. Reason: attach picture of right file size
PS, extra thanks for the heads up on the PCB antennas.. I just went to get a log periodic, and then my curiosity got the better of me.
here’s my haul that just arrived.
Any pointers to good wideband LNAs that are buyable or makeable for a nice price would be appreciated.
Attachments:You must be logged in to view attached files.Jared May 24, 2015 at 4:16 pm
I just got the ones on the left hand side.
I’ve made some NF probes with RG405 rigid, but havent had a chance to characterise them yet.
Minicircuits make a bunch of good kit that isnt too expensive and probably suitable for the wideband amp.
I have a loan BB60C and VSG25A (thanks to Bruce and Ken!!) to play with and now I have to say I want one of each of these now too!
TEM cell is 99% finished (just needs some platen braces) and is looking pretty swish!
BruceModeratorBruce July 23, 2015 at 6:15 pm
I don’t know if you ever got an answer about the BB60A current draw. It draws around 1.2A. We shipped it with a USB3.0 Y-cable so users could spread the current draw over 2 USB3.0 ports.
hansrosenbrandParticipanthansrosenbrand August 26, 2015 at 7:48 am
Could you please give us an update about what kind of EMC pre-compliance functionality your are planning to add?
Regards, HansAndrew August 26, 2015 at 8:46 am
We have not started development on this yet, and we don’t have a time frame for any of this, but the types of functionality we have discussed here are log frequency sweeps, quasi-peak detectors, 6dB bandwidth filters, more support for correction tables, spectrum masks, and numerous other requests from customers.
This is what we would like to see in the future. When and what actually gets developed is still to be determined.
A.J.Jared September 8, 2015 at 5:13 pm
An update on EMC debugging (with BB60C).
We had a particularly troublesome HDMI port and due to the frequencies involved in measuring it, a lack of a suitable BW oscilloscope for signal ringing, eye-pattern and jitter analysis. Not to mention suitable differential probes…
I was looking at making some coupling transformers with a couple of secondaries such that I could connect them in-phase and out-of-phase to get differential/common mode signal levels. Again the bandwidth of the signal proved problematic for suitable magnetics.
However I recalled an article I found some time ago, which uses a hybrid coupler to transform a differential pair input into two single ended outputs – one differential mode, the other common mode. So after hacking some short SMA pig-tails onto an HDMI connector (also made a USB and Ethernet one) I then just connected up to the BB60C and had the HDMI display a fixed image. This gave me a defined spectral content signal. Then I just displayed that test signal on the two iterations of PCB layout and observed the drastic reduction a carefully considered layout makes…
I’m interested in the BB60 and one of the planned applications is EMC precompliance testing.
I’ve seen that you recently added the 6db RBW filters with Spike 3.0.22.
Do you plan to add quasi-peak detectors soon?Andrew April 8, 2016 at 12:15 pm
Hi Electronic Eel,
Yes, in the next release we will be adding quasi-peak detectors. They will be single frequency measurements, not QP sweeps. If you have used pre-compliance software, you will recognize the detector bar meter plots. For now, the BB60C will be supported for these measurements.
thanks for your answer.
If you do not offer QP sweeps, how do you recommend to use this? Do a classic peak sweep and afterwards do the QP measurement manually on just the frequencies which are near/over the limit?
Is a true QP sweep that much more work to implement for you? Maybe it is easier to implement if you don’t utilize the full bandwidth of the BB60 for the QP sweep at first?
Andrew April 8, 2016 at 2:05 pm
- This reply was modified 6 years, 11 months ago by Andrew.
Hi Electronic Eel,
Yes, the process you described is the standard method of testing on an EMI receiver.
QP sweeps may come at a later date.
Cory AllenKeymasterCory Allen June 1, 2016 at 3:04 pm
Hi Julian. Once upon a time you asked about precompliance testing with the BB60. Spike 3.1.0 now includes this functionality.
Check out the blog post here: https://signalhound.com/news/spike-3-1-0-now-includes-emc-pre-compliance-analysis-mode/
Just figured I’d let you know.
You must be logged in to reply to this topic.