apache-labs.com

I've been having the occasional sequence error (2 or 3 per hour)...usually 264 or 265. I have ignored them since the radio recovers after a couple of seconds, and I wasn't sure when they had started ......been changing hardware connections and also continuously upgrading Thetis for quite awhile.
I decided that perhaps my ethernet cable (direct connection to the computer on a dedicated cable) was a place to start. The cat 6 that I am using is individually screened pairs plus overall screening. It was a 3 metre long cable which was way too long for the setup that I have. I also had more than a few ferrite cores clamped on to it.
I cut the cable down to just over a metre in length, in doing so I realised that all I had for termination was a standard RJ 45 connector so the cable although it would still be screened would not now connect the rig 7000 mkII to my computer via the the ethernet cable screening. I popped a couple of ferrites back on the cable and reconnected the cable.

Since I have done this I have had zero sequence errors.

So what is the best cat cable for connection and should one be using ferrite clamps on the cable ?

The (humble) cat 5 cable has no screening there is no DC connection between rig and computer.

A good cat 6 cable has full screening but connects the rig to the computer via the braid and apparently a couple of the cores.

I do have a good ground connection between rig computer and amp (50mm wide copper foil ~ 8mm^2) so there should not be an issue there.

I wonder if more folks have had issues with similar errors and what they have done to fix the problem. Keeping cable screening but cutting the DC link apparently fixed my problem. I'm interested to know what solutions are out there.

In general, and over typical residential distances, there is little to no difference in your choice of network cabling. This is especially true for very short connections in a shack/on a bench/in a rack/etc. Cat 5 will happily and reliably run at GigE speeds under those circumstances.

It sounds instead like you have a grounding/bonding issue in your station configuration.

Shielded Ethernet cable requires shielded connectors on both the cable and the connected equipment to have any value. I don't believe the G2 uses a shielded Ethernet connector but might be wrong (I don't have a G2). Most PCs do not provide shielded connections, but some do.

Rarely do ferrites on Ethernet cables offer any benefits.

In most cases creating proper shielded connections is the right thing to do. However, if your primary grounding and bonding schemes is not providing equal and effective paths to ground for all equipment then it is easily possible to get unwanted currents on important cables and connections.

It would probably be time well spent investigating what improvements and changes could be made to your station grounding and bonding configuration.

Scott, I agree whole heartedly with you.

I've been in the very high power drives business for over 50 years. Earthing, bonding, grounding have always been of great concern. We've gone from the "audio" days of an earthing tree with everything connected in the correct order. To connecting everything together, ground the screening at both ends, plate all steelwork and bond all panels together. Finally fibre optic connections have proven the best way of doing high speed data connections with the likes of very high power V.S.D. drives producing very high levels of interference.

What I've got / had shouldn't have been a problem. 50mm wide copper foil is about as good as it gets from the point of low inductance grounding/bonding connection. I have never had a problem during transmit with RF feedback or RF getting in where it shouldn't be. This problem occured during recieve with (alledgedly) nothing else going on.

I went from a CAT 5 cable to a full CAT 6 cable based on what I had read on this web site .... and of course the thought that CAT 6 had to better than CAT 5. Similarly the habit of adding a couple of ferrite clamp-on chokes to all inputs/outputs has been a good one to keep RF out of sensitive parts.

However the combination of CAT 6 cabling and ferrite cores seems to have been a problem. (Hence the post) I have seen a couple of posts asking about sequence errors 264/5 and wondered if there was a common issue buried in there somewhere.

Should clamp-on ferrite cores be able to affect ethernet cable operation ? I don't know but I hope that someone in the group would know and be able to supply guidance.

Bob

Quick update. I measured that the screen of CAT 7 cable is connected to the chassis of my 7000 mkII. As was my HP EliteDesk PC I 5.