n6gn

I worked in an HP microwave lab and as I remember the production line spec for torque was surprisingly low, only 7 inch-pounds. That was with quality connectors on the highest quality vector network analyzers as well as on swept signal sources. As time went on other connector types good to above 18 GHz came to the fore but I don't think they were tightened more.

As Martin suggests, it's very easy to over-tighten these beyond that specification thinking it will make things better. It probably will only make things worse.

Glenn n6gn

Getting rid of differential noise is fine, though the power management IC regulators in the Kiwi's BB is so good that this is seldom to never a problem, in my experience. (Try deliberately introducing differential noise and check the Kiwi's response. I haven't done this for a while but I think even 100 mV tends to be pretty invisible)

But just looking the schematic of this typical linear regulator and wondering ...

What would happen if we built a PS, either linear or SMPS and then added another series pass device in the 'GND' leg of the output to raise the impedance and reduce CM current? Against typical CM source/load impedances, would it significantly improve the situation in real circumstances?

It likely might not since typical total CM impedance may be in the 100's of ohms territory and well above the effective resistance of the series pass devices, but it wouldn't be difficult to try.

I think the question was reasonable since it was "using down converters". For this reason using a frequency converter/extender such as the 0-2 GHz transverter I use here suggests the value of having many additional bands listed as options.

Amateurs have already made a great deal of use of narrow band modes above HF, particularly WSPR on 2m and shorter wavelengths.

6m FT8 is very active now in the Northern Hemisphere as we are in the midst of the sporadic E season. With the LO offset JKS has added, the KiwiSDR already natively supports all-modes on all-bands. Adding FT8 support and even all-extension support makes a lot of sense to me.

Why not support it all in a Kiwi?

Just 1:1 on small PCB, SMA male to Kiwi Completely isolated from SMA female to antenna feed line. This leaves only about 1 pF of common mode coupling between Kiwi and antenna line.

top side of board shown above, Center tap, if present T1-1T, is unused.

T1-1 from Mini-Circuits,

SMA edge connectors from eBay,

unzip & drop .brd file onto OSHPark.com. 3 boards postpaid in US in about 10 days, probably < US$5.

I've not examined recent products from the Ukraine but I can not recommend the earlier transverters. Perhaps the receive converters are different/better.

How much interest would there be in the creation of a fast-switching block down-converter for the Kiwi? This would be something with a high side, say 3 GHz IF preceded by a [20 MHz] stepping LO that went from, 3-6 GHz, mixer and LPF?

The result would be an "all band" down-converter one [20 MHz] swath at a time such that the Kiwi could use any of it's capabilities over 0-3 GHz? NF and IMD could be good and no additional filtering would be necessary as long as out-of-band signals were kept tolerably low compared to a mixer's spec. +20 dBm TOI should be fairly easy. Of course, filtering could be added at the input, as for a band-limited converter of the type most often used.

It would seem to me that a Kiwi extension to use I2C to control the LO/receive_frequency should not be difficult. The goal would be fast switching so that multiple 20MHz chunks might be stitched together to create a wider spectrogram.

Glenn n6gn

I'm attaching a write-up of what has worked well for me to improve Kiwi noise performance by reducing common-mode current paths through the Kiwi.

For a stock, wired-LAN Kiwi it is very common for noise and coherent signal current on the LAN cable to find a path through the Kiwi PCB ground and out either the Power Supply return or, more often, via common mode current on the antenna feed attached to the SMA input. This kind of QRM/QRN is insidious because it may mostly show up when an antenna is attached causing the user to believe the problem is elsewhere.

By operating the Kiwi with an inexpensive WiFi router, powered from the Kiwi and connected by only a single micro_USB <==> USB-A cable, the common mode current paths which can cause this degradation can be eliminated.

Because of the success of this scheme in both improving performance of previously wired-LAN installations and also the possibility of building a completely isolated, portable Kiwi system to use as adiagnostic tool for improving Kiwi and general site performance, I'm offering this description.

Write me for more information or for the 3D printable file for the USB clamp described.

Glenn n6gn

I too like it, even if only as a reminder.

Glenn n6gn

The family of lines in 6.5-7 MHz you suggest seem to me to be power line related, though this is true of many SMPS noise sources. When I detect in 10 kHz AM and analyze with Audacity I see astrong 50 Hz family/component

This is as expected from an imperfectly filtered SMPS input. Furthermore, the spectral width of the lines is consistent with jitter/phase noise of a switcher's rate and it appears to have stronger even harmonic components, as is common from full-wave rectification.

But as to the actual source and coupling mechanism - that's harder to say. There are so many uses for SMPS supplies these days, some of them for lighting related power that an after-dark correlation is not unlikely.

To put things in perspective, while your noise floor is no doubt raised considerably by this, you are in pretty good company and a lot kiwiSDRs reveal similar and worse results. If you are using a well matched antenna, your noise is about 30 dB above kTB which isn't so dreadful at that frequency. OTOH, it may be higher than that if your system is not well matched to the radiation resistance. You could be seeing CM currents, current in the earth below or other coupling mechanisms. I suggest not first considering it to be inverse-square radiation. That is actually much rarer, in my experience.

Glenn n6gn

@la6lu that made me smile...
I too had problems with the factory fan and replaced several with Clint's suggestion https://www.alltronics.com//cgi-bin/item/28F082/search/Sunon-5V-Fan-Size--30mm-x-30mm-x-6mm-

I've now replaced four fans with this alternative but have just recently had to replace one of the alternatives when it too started to get noisy at the remote, poorly climate-conditioned site. It gets both hot and cold at that remote site so maybe this is to be expected. If someone finds an even better substitute or can identify a good maintenance plan for these (change oil every 20k km?) I'd be interested. Maybe this is just going to be part of the cost of keeping a Kiwi running continuously...

Actually, I think the power has been on the whole time. The issue seems to be a poor quality router that loses its way and no one having access to help it out. There may be plans for a replacement but there is perhaps a little politics in that one. Perhaps Rob may be able to elaborate.