HFU HF Underground
Technical Topics => Equipment => Topic started by: ChrisSmolinski on April 17, 2023, 1311 UTC
-
I reconfigured some antenna connections here. Now the 900 foot sky loop is feeding the Airspy Discovery that I use to record MW and above the MW band for pirates (it used to share the 500 ft beverage with one of the KiwiSDRs) in addition to the netSDR (for 43m) and one of the KiwiSDRs.
One interesting thing, when I connected the Discovery to the splitter, the noise floor on the KiwiSDR went up on MW. More like RFI than noise. I swapped in a coax cable wrapped around a large toroid core in place of the plain coax from the splitter to Discovery, and the noise floor went back down to normal. Seems like some RFI was flowing on the coax shield when the Discovery was connected. I think the noise floor on the Discovery was higher as well.
Isn't RF fun? :P
-
Yes, there are some mysterious situations.
Once I described how I accidentally touched my bare leg to the splitter of antennas and felt a prick of electric charge. The voltage of several volts is present on the cable screen when the laptop is working on the power adapter.
I didn't do anything to prevent it, and maybe this is the reason for the noise level of the power supply, about 0.5S. I now have to try an antenna splitter with an electrically isolated input so I'll check if there is an improvement.
The splitters have about 2dB of attenuation, but this is not a noticeable difference with the very high signal levels from my long Skyloops.
-
The Airspy HF+D is USB connected, so RFI or even a ground loop could be originating via the connected PC.
I have a massive ferrite choke on the USB cable of my Airspy HF+D, plus more CM choking up the feedline.
BTW, if receivers have significant impedance differences, just splitting a feedline can cause significant signal variances between them.
There are active HF multicouplers, though some have used old RCA (composite/component) video distribution amps often found for cheap on eBay and similar. Example:
http://gm1sxx.blogspot.com/2018/01/the-poor-mans-receiver-multicoupler.html
I have a smaller video distribution model somewhere around here rated to VHF frequencies. IIRC, it did slightly raise the noise floor, but the SNR difference was probably not too big a deal at HF and below.
-
I use AC line isolation transformers on all radio related equipment. Look for Series 30 Topaz Ultra-Isolator Isolation Transformers on the used market.
Topaz Low-Capacitance Isolation Transformers - for Affordably Clean Power
https://www.head-fi.org/threads/topaz-low-capacitance-isolation-transformers-for-affordably-clean-power.857448/ (https://www.head-fi.org/threads/topaz-low-capacitance-isolation-transformers-for-affordably-clean-power.857448/)
The magic that isolation transformer uses to suppress transients and noise
https://electrical-engineering-portal.com/isolation-transformer-suppress-transients-noise (https://electrical-engineering-portal.com/isolation-transformer-suppress-transients-noise)
Then ditch all those cheap power bars with surge protectors. They are a potential source of noise.
... and just in case you are curious, yes, I use a floating neutral, hence no MOVs (surge protectors) or surge protected power bars, no on/off switches with fancy LED lights etc ... Essentially, you don't want anything bridging across hot, neutral or ground --- just clean power.
-
If looking for something retail off-the-shelf, the APC Line-R AVR series is decent as well.
https://www.apc.com/us/en/product-range/61877-liner/
If one does not care about automatic voltage regulation, the ESP XG-PCS-15D power filter is another decent choice for powerline common-mode mitigation. They are popular for protecting ATMs, photo copiers, and similar. Actual branding could be ESP, Ricoh, SurgeX, etc. Working pulls often are under $30 shipped at eBay.
https://www.ametekesp.com/esp/next-gen/next-gen-120-208v
Normal Mode: >30dB 80kHz – 50MHz
Common Mode: >30dB 70kHz – 50MHz (XG-PCS-15D / XG-PCS-20D)
A couple of basic demos:
https://www.youtube.com/watch?v=cRurSc9QKLA
https://www.youtube.com/watch?v=UQkMoNlXbb4
-
Great videos Rob, thanks. I'm going to have to keep an eye out for those ESP SurgeX units. I pretty much exclusively use APC Line-R units as voltage regulators and they work great. Same with the UPSs, I use the APC Back-UPS Pro xxxx S series which produce a sine wave output.
-
Now, let's talk about those splitters. I'm assuming cable TV splitters here? As tempting as they may be they are a bad choice for HF and here are a few reasons why.
1) Low quality splitters lack circuitry and isolation in them. Have a look here: Coaxial Cable Splitters
https://www.flukenetworks.com/knowledge-base/cableiq/coaxial-cable-splitters-cableiq (https://www.flukenetworks.com/knowledge-base/cableiq/coaxial-cable-splitters-cableiq)
What's in your splitter?
2) Cable TV systems are designed with a frequency split:
a) Forward - 52 to 1000 MHz (or higher). This is what the customer receives; and
b) Return - 5 to 42 MHz. This is what the customer transmits back to the head-end.
Chances are that at best you can find a bi-directional 5 to 2400 MHz splitter. So where does that leave the 0 to 5 MHz portion of the HF band?
3) Except for a 2 way splitter, only 1 port will be of lower loss (approx 4 dB and all others will be approx 8 dB). That's just the way it works in a passive design.
4) Do you terminate the unused ports with 75 Ohm resistors? You should.
... food for thought.
-
Took a look at the old analog video distribution amp on the shelf. It is an Inday HDDA-1 marketed for component video back in the day, thus the extended frequency rating compared to basic composite-only models.
https://www.inday.com/hdda1.html
https://www.inday.com/assets/hdda-manual-inday-web.pdf
"The HDDA-1 is a very wideband analog Distribution Amplifier (DA) capable of buffering and splitting any nominal 1 to 2 Volt (peak to peak) signal that terminates into 75 ohms.
....
Frequency response is specified at DC-220 Mhz +/-.3dB."
Spec table mentions driving > 300' cables. So yeah, the outright gain is likely why I remember a somewhat increased noise floor. Dropping the gain at the PCB might be easy enough, or perhaps just toss some cheap CATV attenuators on the output cables. YMMV. I never really bothered much with it due to then having a larger antenna farm available.
All that said, sadly it looks like prices for analog video distribution amplifiers are up at eBay compared to a few years ago. These types of amps used to be a few dollars each. Oh well.
-
I reconfigured some antenna connections here. Now the 900 foot sky loop is feeding the Airspy Discovery that I use to record MW and above the MW band for pirates (it used to share the 500 ft beverage with one of the KiwiSDRs) in addition to the netSDR (for 43m) and one of the KiwiSDRs.
One interesting thing, when I connected the Discovery to the splitter, the noise floor on the KiwiSDR went up on MW. More like RFI than noise. I swapped in a coax cable wrapped around a large toroid core in place of the plain coax from the splitter to Discovery, and the noise floor went back down to normal. Seems like some RFI was flowing on the coax shield when the Discovery was connected. I think the noise floor on the Discovery was higher as well.
Isn't RF fun? :P
Still using these? I remember your recommendation from about 6 years ago. https://www.mouser.com/ProductDetail/Fair-Rite/2631803802
-
I tend to use mini-circuits splitters. Although the port loss is substantially higher (approx 6 dB per port) it is at least linear across all ports. Port-to-port isolation is high, typically 30 dB. For high gain SDR receivers, the 6 dB loss is marginal. For example, my RSPduo typically runs at +20dB IF input attenuation therefore the 6 dB loss of the splitter has no negative impact. If necessary, the loss can be mitigated by adding a good quality 6 dB amplifier at the input of the splitter thus keeping unity gain on the outputs. Just as the splitters, there are also suitable mini-circuits amplifiers available.
A ZSC-4-1 splitter gets you 4 ports, 0.1 to 200 MHz and a ZSC-4-3 gets you 4 ports, 0.25 to 250 MHz.
There are plenty of surplus mini-circuits splitters and amplifiers on eBay at a very reasonable price.
https://www.ebay.com/itm/385567484830 (https://www.ebay.com/itm/385567484830)
https://www.ebay.com/itm/264992215420 (https://www.ebay.com/itm/264992215420)
-
Still using these? I remember your recommendation from about 6 years ago. https://www.mouser.com/ProductDetail/Fair-Rite/2631803802
Richly and religiously.
-
I tend to use mini-circuits splitters. Although the port loss is substantially higher (approx 6 dB per port) it is at least linear across all ports. Port-to-port isolation is high, typically 30 dB. For high gain SDR receivers, the 6 dB loss is marginal. For example, my RSPduo typically runs at +20dB IF input attenuation therefore the 6 dB loss of the splitter has no negative impact. If necessary, the loss can be mitigated by adding a good quality 6 dB amplifier at the input of the splitter thus keeping unity gain on the outputs. Just as the splitters, there are also suitable mini-circuits amplifiers available.
A ZSC-4-1 splitter gets you 4 ports, 0.1 to 200 MHz and a ZSC-4-3 gets you 4 ports, 0.25 to 250 MHz.
There are plenty of surplus mini-circuits splitters and amplifiers on eBay at a very reasonable price.
https://www.ebay.com/itm/385567484830 (https://www.ebay.com/itm/385567484830)
https://www.ebay.com/itm/264992215420 (https://www.ebay.com/itm/264992215420)
Unlike VHF and UHF, signal (and noise) levels on HF (and MW to an even larger degree) are so high, and modern receivers so sensitive, that I've found losing even a few dB is trivial and unnoticed. And no need for an amplifier afterwards, which is just going to add more noise.
-
Yes, I live with the loss. It is actually beneficial and prevents the front-end of the SDR from overload since the majority of good SDRs have sensitive / high gain inputs. Most SDR receivers do not have band pre-selectors in the front-end so keeping overload to a minimum across the input is beneficial.