grounding equipment, whats's the best way to get a low resistance connection?

[cdev]

A couple of years ago I acquired a large collection of old school binding posts and Ive decided I want to use some of them to make a permanent easy to use, low resistance ground connection on at least my computer and SDR receivers cases and then strap them together so that the measurable resistance between the two cases is the absolute minimum possible. Some parts of my computers case are aluminum and other parts are steel. I guess the gist of my question is where is the best place on the back of the computer to ground it, and whats the best way to do it materials wise?

The case on my computer was a fairly cheap one and now I realize its RF properties are quite bad, as even though the case is closed, cables routed near the computer pick up nasty RFI. Which wont go away. After trying a milion things Ive become convinced that the grounding - low residual resistances and crappy connections on the stereo miniplug audio cables are the area where a little bit of work is likely to bring the biggest improvements.

They are just generic old bind posts, made out of brass. But they will be better than what I have now on the computer for grounding which is just a simple lug.

They have a screw on the back that accepts crimp lugs and I have a bunch of crimp lugs and wire that fits them. Is it possible to get a good connection crimping? an alternative I feel more comfortable with is soldering better ground lugs. Using a loop of copper wire and not using a crimp connection, soldering. I have some "phone style" (they are the kind of terminals found inside old phones in the US) - crimpable lugs that will also take solder, they are tinned. They are pretty small but they support soldering. Do you think they would be preferable? Improving the ground may make a significant difference in the noise floor.

What I thought I should do is use maybe 16 gauge bare copper wire inside the case to connect the back of the binding post to multiple points while observing the noise at the end of the audio cables with the SDR removed.

The motherboard has supports which have grounding symbols on them. Logical place to ground it. Also, on the SDR I thought I could replace the positive power connector with a feed through capacitor for the power and strap the negative power to the case and the board inside via a thick 16 gauge wire.  This approach has worked well in a phono cartridge situation for me in the past. Also its what I needed to do to get maximal peformance out of my LNA.

Cleaning the contacts is also important. I have some deoxit. To be honest its been decades since I last used the stuff. (back in the day we used "cramolin" (sp?) which is similar.) These connections will rarely get messed with. Are you supposed to clean connectors and then wipe it off completely or leave it on?

[MisterDiodes]

This might not be the best section since not really metrology related but since nobody else is answering I'll take a stab at it for what we do in the lab (You might ask in the ham radio section and that might get you a better-quality answer from those guys):

1) Your brass posts are probably fine for simple grounds.

2) Your cheap PC case is splattering RF everywhere. Probably should address that if you can.  Good cases don't cost too much that are rated for better EMI shielding. Shield your SDR also.  Also any switching power supplies (most everything) are noise generators, so move those away.  Most SDR's I've seen are not shielded very well themselves, if at all.

Also watch out for fluorescent or LED lights - those can spew RF absolutely everywhere.

3)  You might try a star ground, not a looped-around-the-shack shared ground. i.e. Pick ONE good ground point and run separate ground leads to that single point - best to use a ground rod nearby if possible.  We like braided ground lead, not solid wire, for a little better performance - but use whatever works.  The really good stuff is the copper braided downlead for very broadband lightning protection, if you can get that.  But experiment with whatever you have, and try different setups to see which is best.

4) Deoxit goes on, work in to clean, then wipe off excess. Be practical about it - you don't need a lot.  If you leave a lot of excess on it can get gooey and attracts dust.

5) Use a good -solid and tight- crimp connection, and maybe solder afterwards if it makes you feel better - a good, tight crimp is usually best.  Make sure you have the right crimper for your wire size and connector.  Or use a GOOD screw-clamp connector for your ground leads.  Nice and snug but don't over-do it.

If it helps, there's an article in Circuit Cellar a couple months ago that addresses trying to run an SDR from a PC.  It turns out the better result was running the SDR remotely using a Raspberry-Pi mini PC:  You put the RaspPi and SDR up in the attic (or wherever) as far away from all the RF noise as much as possible, and control the SDR with the RaspPi using WiFi or Ethernet cable from your PC to Rasp-Pi.  The Rasp-Pi is small and can be shielded in a small metal box, and makes a lot less noise than your desktop PC.  In other words get the SDR as far away as possible from that noisy PC.  You'll have to look up the article but its from a month or two ago - fairly recent.

[WastelandTek]

If you want to get serious you end up doing it like this.  Yes it is a pain in the ass when you need to top off the water loop...lol

[cdev]

These are all really good suggestions. The only thing I worry about with braided wire is small pieces of wire being liberated and causing shorts elsewhere. What I'm trying right now is two lengths of around 16 gauge multistrand copper wire in parallel with the wires going in opposite directions.  Crimped end connectors right now. And that did help.

I do need a better case, and need a better ground outdoors.

Ive also thought about putting the SDR on a RPI. The issue as I see it is my lack of a high end USB audio card. The best I can do right now is a little Syba USB audio dongle that supports 96 kHz sampling rate and around 96 db SNR (although it claims 110 db, its nowhere close) I think my PC's built in sound card works better.

There are much better external USB "sound cards" but they cost >$70  I would rather wait and save that money towards my ham rig. The problem is, the sound card's performance is critical with the kind of setup I have and any little glitches really show up in the spurs and the hump in the middle.

That said its very sensitive when everything is running right.

[MisterDiodes]

Wasteland's got it about right for shielding!!  Nice!

For a couple PC's in the lab, at the state surplus outlet we bought some older Dell commercial workstaion boxes for $15 each that have a good amount of steel and lots of ground fingers / shield gaskets - I think they come in around 20 ~25 lbs empty.  The RFI / EMI noise level is over an order of magnitude attenuated over that cheap stamped thin foil crap on the generic cheap cases.

Also - a lot of noise is being transferred from the MoBo to the SDR over the USB cable, if it uses that.  You might try an isolated USB cable box or use an optical to USB interface.  Use a battery to power the SDR and see if that helps reduce some of the noise.

You have to try a mix of isolation and good grounding.  16 Awg might be a little on the thin side for a decent RF ground, but certainly use whatever gets results for you.  We would use a minimum of something like #10 or #8 - usually #6 or larger - and that's why the braided stuff is handy since you get the flexibility but very low resistance and very broadband.  Yes, those braided leads can drop their fur sometimes if flexed a lot but we put a plastic split wire loom conduit (or sleeving, tubing, etc) over the braid to keep it clean-room safe.  Just an idea for the toolbox.

[cdev]

Actually, it could improve quite a bit from what I got with 16AWG. I have some older silver (tinned wire) braid that I could use, its quite thick- and you gave me the idea to maybe let it run within some cable management spiral stuff I have, which could also enclose the audio cable to keep he two right next to one another. The only problem I have is making the connection to it but I just thought, I could gather the braid at the ends together, wrap it tightly with unenameled copper wire, put some flux on it and solder it with a loop to go around the screws at each end. Another alternative, solder it to a tab of flashing copper with a hole in it for a ground screw. Basically a strap of copper and strap=less inductance.

[CopperCone]

soldering your case shut might be less work and cheaper then buying a new one. or using rf tape.