How to prevent ground loops with multiple coax connections

[jmw]

I'm doing a project to cobble together a frequency standard and GPS timing module, and I have signals (10 MHz and 1PPS) going between multiple boards and modules, coupled over coax, which is a recipe for ground loops. Some of the modules have isolated DC/DC converters for power, but connecting them with coax between them breaks isolation, and connecting two creates a loop.

Here are some ideas I thought of to get rid of ground loops. Are any of these particularly good or bad?

1. Tie together all the ground/common nodes of the components, then float one end of the coax connection between boards. Leave the shield unconnected to the ground plane at one of the PCBs. This forces the current return though the ground or chassis path. Seems ok with high-impedance terminations like 1PPS but less good with 50 Ω where there might be non-negligible current.

2. Do the above but capacitively couple the shield and PCB ground plane. Maybe 100 kΩ in parallel with 10 nF and this will push DC and mains frequency off the coax while allowing high frequency to return through the coax shield.

3. Transformers at the BNC ports. Seems ok if the signal is narrowband like 10 MHz, just choose a transformer specced to that frequency. I'm not sure if this will work well with a pulse like 1PPS without distortion. The GPS module puts out a crazy long pulse (100 ms) which I'm concerned would saturate a transformer core too. Any part suggestions?

[T3sl4co1l]

1 is simply outright bad: ambient noise adds in series with a given cable, and without a ground at the receiving end to pick up that ground reference, the noise is added directly onto the signal.  As you say, it might be okay with slow signals -- given the additional provision that the signal is well filtered first!  Which is a good principle in general, only use what bandwidth you need to.

2 is the preferred way, and you can improve it further by using a differential receiver so that ground level is not ignored.  The relevant case here being, ambient noise (in this case not so much RF thanks to the AC grounding, but mains and slowly varying offsets ("DC")) skews the baseline, adding phase noise.

Use lots of caps in parallel, to approximate a wide ground connection.  Just one will be adequate for low frequencies (a few MHz), but you will need many in parallel to cover the full range of interference.  Again, this is give or take signal filtering.

This is the preferred solution for dealing with ground loop, in audio applications.  (For analog signals, the differential receiver is simply a diff amp.)  Or, it would be if they knew about it, but no one seems to care.  Go figure?

3 is good of course, when it's applicable.  PPS could be converted to a RZ code first, or modulated in any number of ways (ASK, FSK, etc.).  Note that an RZ code (basically short pulses up or down) is high bandwidth, so is very susceptible to noise, especially impulsive noise.  This can be addressed somewhat with a windowed filter (disable the input until very near the expected arrival time of the next pulse), so there are digital solutions, but ASK and especially FSK offers strong benefits from signal filtering, and you can still do the window filter after detection if you like.

Tim

[Zero999]

As far as the long pulses and transformers are concerned. How about using DC blocking capacitors, so only the rising/falling edges are passed? A Schmitt trigger could be used to restore the long pulses, if necessary. This will only work, if the DC amplitude of the pulses is unimportant and doesn't need to be transmitted.

[jmelson]

Ummm, first, is this ACTUALLY a problem?  I've built all sorts of things that are connected by multiple coaxes, and rarely had an issue with it.  Unless you are using millivolt-level signals, it just doesn't usually cause any issue.

Jon

[jmw]

Ummm, first, is this ACTUALLY a problem?  I've built all sorts of things that are connected by multiple coaxes, and rarely had an issue with it.  Unless you are using millivolt-level signals, it just doesn't usually cause any issue.

But muh phase noise ... I do want to keep jitter as low as possible in the 1PPS signal.

[jmelson]

But muh phase noise ... I do want to keep jitter as low as possible in the 1PPS signal.

Well, you could run the 1 PPS on twisted-pair cable via LVDS transmitters and receivers.
That would get the jitter down pretty far.  It could be less than with coax as the coax adds a lot of capacitance.

Jon